High-efficiency protein expression mediated by enterovirus 71 internal ribosome entry site

COMMUNICATION TO THE EDITOR High-Efficiency Protein Expression Mediated by Enterovirus 71 Internal Ribosome Entry Site Jin-Ching Lee,1,2 Tzong-Yuan Wu,3 Chien-Fu Huang,1 Feng-Mine Yang,1 Shin-Ru Shih,4 John T.A. Hsu1,5 1 Division of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 128 Yen-Chiu-Yuan Road, Sec. 2, Taipei, Taiwan, Republic of China; telephone: + 886 2 2652 4135; fax: + 886 2 2789 0264; e-mail: tsuanhsu@ nhri.org.tw 2 Faculty of Biotechnology, Kaohsiung Medical University, Kaohsiung, Taiwan, Republic of China 3 Department of Bioscience Technology, Chung Yuan Christian University, Chung Li, Taiwan, Republic of China 4 School of Medical Technology, Chang Gung University, Taoyuan, Taiwan 5 Department of Chemical Engineering, National Tsing Hua Unversity, Hsinchu, Taiwan, Republic of China

Received 25 October 2004; accepted 7 December 2004 Published online 7 April 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.20440

Abstract: An internal ribosome entry site (IRES) has been used to facilitate the expression of more than one protein in a single transcript. In this study, we examined the translational activities of several IRES elements derived from different RNA viruses. The protein expression of encephalomyocarditis virus (EMCV) IRES is similar to that of hepatitis C virus (HCV) IRES in mammalian cells. Notably, the protein expression of enterovirus 71 (EV71) IRES was 23-fold higher than the efficiency of EMCV IRES following normalization of mRNA transcriptional level. Thus, expression of the secreted alkaline phosphatase (SEAP) reporter protein in mammalian cells may be controlled at desirable levels by using appropriate IRES in the expression vector. B 2005 Wiley Periodicals, Inc. Keywords: IRES; polycistronic; protein expression; EV71; EMCV; HCV

INTRODUCTION Expression of many viral and cellular proteins was controlled at the translational level. Initiation of mRNAs translation can be mediated by an Internal Ribosomal Entry Site (IRES) element, which was first identified in picornavirus, in a cap-independent manner (Jackson, 1988; Vagner et al., 2001). The viral genome of picornavirus consists of a postive-strand RNA molecule with a single long open reading frame encoding a polyprotein. During virus infection, the cap-dependent host cell translation is

Correspondence to: John T.A. Hsu

B 2005 Wiley Periodicals, Inc.

drastically inhibited and viral protein expression continues via a cap-independent mechanism that is mediated by IRES; with the aid of transacting factors, such as the translation initiation factors (eIFs; Bedard and Semler, 2004; Pestova et al., 2001). In picornavirus, a region of nucleotides located within the 5V-end untranslated region (5VUTR) of viral RNA promoted internal initiation of protein expression. Some unique nucleotide sequences and stable secondary RNA structure in the IRES elements were suggested to be the major determinants of IRES function (Davies and Kaufman, 1992; Jackson and Kaminski, 1995). In fact, IRES-mediated protein expression is not a unique feature in picornavirus. Many IRES elements have also been identified in the RNAs from other viruses (e.g., hepatitis C virus) and higher organisms such as yeasts, insects, and mammals (Gan and Rhoads, 1996; Iizuka et al., 1994; Macejak and Sarnow, 1991; Oh et al., 1992; TsukiyamaKohara et al., 1992; Vagner et al., 1995). Additionally, IRES has been successfully applied to a variety of biotechnological applications, such as heterologous protein expression, production of transgenic animals, and gene therapy (Borman et al., 1995; Moser et al., 2001; Pao et al., 2003; Peroni et al., 2002; Schlatter and Fussenegger, 2003; Weber and Fussenegger, 2002; Wen et al., 2003). Though different genes can be expressed under separate expression cassettes in one vector, this strategy may encounter problems such as promoter attenuation. Further, the vector size can sometimes be too big to increase handling difficulty during cloning. Alternatively, IRES can be

employed to design bicistronic constructs to express at least two genes in a single transcription cassette under the control of a single upstream promoter (Ciafre et al., 2002; De Felipe and Izquierdo, 2000; Fussenegger et al., 1997; Mielke et al., 2000; Ramesh et al., 1996). The bicistronic or polycistronic approach offers a high degree of flexibility for regulating gene expression. The IRES elements from poliovirus and encephalomyocarditis virus (EMCV) are most commonly used for the construction of bicistronic expression constructs (Martinez-Salas, 1999). However, compared to the cap-dependent translation of the upstream gene, the EMCV IRES-mediated translation is less efficient (Hennecke et al., 2001; Lee et al., 1997; Zhou et al., 1998). The aim of this study was to compare the efficiencies of several IRES elements in mediating the initiation of the internal translation in different cell lines. In this work, we designed a bicistronic reporter vector, comprising the h-galactosidase gene as the upstream cistron and the secreted alkaline phosphatase (SEAP) gene as the downstream cistron, to examine the IRES translational efficiency. The protein expression of three different viral IRESes was evaluated in mammalian cell lines. Our results provide valuable information to facilitate the selection of appropriate IRES elements to achieve desired expression levels of recombinant proteins.

MATERIALS AND METHODS Construction of Bicistronic Vectors All of the bicistronic plasmids used in the present study were based on the parental construct, pTriEx-MTEGFP, derived from pTriEx-4 (Novagen, Madison, WI). pTriEx-MTEGFP contains an EGFP (enhanced green fluorescent protein) gene cassette where EGFP is expressed under the control of the metallothionein promoter (MT) promoter. This MTEGFP expression cassette was blunt-end cloned into the SphI site of pTriEx-4. To make the bicistronic reporter constructs, both PCR-amplified h-galactosidase gene from ph-gal-control (Clontech, Palo Alto, CA) and secreted alkaline phosphatase (SEAP) gene from pSEAP2-control (Clontech) were cloned into the EcoRI and XhoI sites of pTriEx-MTEGFP plasmid, respectively. The resulting plasmid was termed pGS. Subsequently, both pGS and the PCR-amplified IRES elements from encephalomyocarditis virus (EMCV), hepatitis C virus (HCV), and enterovirus 71 (EV71) were digested with NotI, and inserted in between the h-galactosidase and SEAP genes of pGS plasmid to create pGS-EMCV, pGS-HCV and pGS-EV71, respectively. The inserted sequences were corresponding to the complete IRESes of EMCV (nucleotides 1130-1710 of pIRES; Clontech, Palo Alto, CA), hepatitis C virus (nucleotides 58-424; GenBank Accession No. AF054247), and enterovirus 71 (nucleotide 1-750; GenBank Accession No. AF117633). To construct monocistronic plasmid expressing h-galactosidase or SEAP, the reporter gene was

introduced into pTriEx-MTEGFP in the EcoRI site to generate pTri-GAL or pTri-SEAP. In pTri-GAL and pTriSEAP, gene expression is under the control of cytomegalovirus (CMV) immediate early promoter. Clones were confirmed by DNA sequencing and restriction analysis.

Cell Culture Huh-7 (human hepatoma cell line, obtained from Apath, LCC, USA), COS-7 (monkey kidney cell line, ATCC number: CRL-1651), and CHO (Chinese Hamster Ovary, ATCC number: CCL-61) cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) with 10% heat-inactivated fetal bovine serum (FBS, GIBCOy, Lot no. 458112; Invitrogen, La Jolla, CA) and incubated at 37jC in the presence of 5% CO2.

Transfection and Analysis of IRES Activities in Mammalian Cells The procedure for examining IRES activity by transient transfection experiments is basically as described (Lee et al., 2003). Briefly, cells in 24-well plates at 70% confluence were transfected in triplicate with 0.5 Ag DNA using LipofectAMINE 2000TM reagent (Life Technologies, Grand Island, NY) for COS-7 and CHO cells or FuGENEy 6 (Roche, Indianapolis, IN) for Huh-7 cells. Two days post-transfection, cells or supernatant were harvested and analyzed for reporter gene expression. The SEAP activity in culture media and the h-galactosidase activity in cell lysates were measured using Phospha-Lighty and Galacto-Stary assay kits (TROPIX, Foster, CA), respectively. The chemiluminescent intensities reflecting relative SEAP and galactosidase activities were detected with Top Counter Microplate Scintillation and Luminescence Counter (PACKARD, Meriden, CT). The relative translational efficiency of IRES was determined by normalizing the level of downstream reporter gene product (SEAP) against the level of upstream reporter gene product (h-galactosidase).

Analysis of In Vitro Translation Products The in vitro coupled transcription/translation (TNT) reactions were performed with commercially available rabbit reticulocyte lysates (Promega, Madison, WI). Transcription/translation reactions can be readily conducted using the pGS-plasmids since these plasmids contain bacterial phage T7 promoter in front of the cloned genes. The reactions were carried out at 30jC for 3 hours in the presence of [35S]methionine and [35S]cysteine (Institute of Isotopes Co., Ltd, Budapest, Hungary) at 100 ACi/mL. Transcription/translation reactions were terminated by the addition of SDS-PAGE sample buffer (50 mM TrisCl, pH 6.8, 100 mM dithiothreitol, 2% SDS, 0.1% glyc-

LEE ET AL.: PROTEIN EXPRESSION MEDIATED BY ENTEROVIRUS 71

657

erol, 0.2% bromophenol blue) and boiled for 5 min. The 35 S-labeled products were analyzed by 10% SDS-PAGE. Quantification of Transcribed Bicistronic RNA Level Huh-7 cells were seeded at 3  105 cell/well in 6-well cell culture plates in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum (Gemini Bio-Products), 100 units/mL penicillin G, and 100 mg/mL streptomycin sulfate at 37jC in a humidified atmosphere of 5% CO2. After 24 h, transfections were performed with Fugeney 6 (Roche, Indianapolis, IN) according to the instructions of the manufacturer. Briefly, transfection mixtures for each well contained 1.5 Al of Fugene 6, and 0.9 Ag of pGS-EMCV, pGS-HCV, or pGS-EV71 each. Cells were incubated in the transfection mixture for 6 h at 37jC in an atmosphere of 5% CO2. The cells were then incubated for c48 h in fresh Dulbecco’s modified Eagle’s medium. Total RNA

was prepared from cells using Trizol RNA isolation kit (Invitrogen, Carlsbad, CA), and RNA concentration was estimated from absorption at 260 nm. The transcription of bicistronic RNA was quantified and normalized by quantitative PCR (qPCR) method with seap specific primers (Forward: 5V-CCC ACC TTG GCT GTA GTC AT-3V; Reverse: 5V-GGT GAA CCG CAA CTG GTACT-3V) and Actin specific primers (Forward: 5V-CATCCTGTCGGCAATGCCAGG-3V; Reverse: 5V-CTTCCTGGGCATGGAGTCCTG3V) using the Roche LightCycler instrument (Roche Diagnostic, Inc. USA). Briefly, real-time PCR was set up using a 20 AL volume, with a 100 ng sample of DNA added to 2 AL LightCycler FastStart DNA Master SYBR Green I, 0.5 AM primer pair, and 4 mM MgCl2 stock solution. The PCR consisted of denaturation at 95jC for 10 min, and 45 cycles of denaturation at 95jC for 10 s, annealing at 59jC for 5 s, and extension at 72jC for 10 s. Five concentrations (1.5 ng/AL, 150 pg/AL, 15 pg/AL, 1.5 pg/AL,

Figure 1. Schematic diagrams of the mono- and bicistronic reporter vectors; pTri-GAL, pTri-SEAP, pGS-EMCV, pGS-HCV, and pGS-EV71. The plasmid pTriEx4-MTEGFP, in which enhanced green fluorescent protein (EGFP) gene was under the control of metallothionein promoter (MT prom), was used as vector backbone for insertion of various internal ribosomal entry sites (IRESes) and reporters. In these vectors, pCMV represents the human cytomegalovirus (CMV) immediate early promoter, p10 indicates the p10 gene promoter of AcMNPV, and T7 indicates the T7 promoter of bacteria phage. IRESes indicated the IRES element from EMCV (Encephalomyocarditis virus), HCV (Hepatitis C virus), or EV71 (Enterovirus 71), respectively. The genes encoding secreted alkaline phosphatase (SEAP) and h-galactosidase (h-gal) were separated by various IRES elements and were expressed under control of the CMV promoter.

658

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 90, NO. 5, JUNE 5, 2005

0.15 pg/AL per reaction, respectively) of standard plasmid DNA were used for the preparation of a standard curve. All the real-time PCR was carried out in triple-replication by using three reaction wells for each template DNA.

RESULTS AND DISCUSSION There are many advantages to utilize IRES for coexpression of proteins using one single mRNA transcript. The EMCV IRES has been frequently employed for such a purpose but availability of alternative IRES elements for protein expression is limited. Thus, it is advantageous to identify more IRES elements with various translational

efficiencies for tuning protein expression levels. We show here that IRES elements with different translational efficiencies may be utilized to yield different protein expression levels. To compare the protein expression of various IRESes, we developed a bicistronic expression vector in which a single transcript containing the h-galactosidase and SEAP reporter genes is under transcriptional control of a single promoter element, the human cytomegalovirus (CMV) major intermediate early promoter/enhancer sequence. The investigated IRES elements were inserted into the junction of these two reporter genes. The configurations of the bicistronic vectors utilized in this study are shown in Figure 1. The translations of h-galactosidase and SEAP genes were

Figure 2. Translational levels of the reporter genes in vitro. (A) TNT experiments were conducted as described in Materials and Methods and the reaction mixtures were separated on SDS-PAGE. 35S-labeled TNT products were analyzed by autoradiography, and the positions of the h-galactosidase and SEAP proteins are revealed by arrows. (B), (C) The bars indicate the relative h-galactosidase an SEAP expression levels as determined by densitometric scannings.

LEE ET AL.: PROTEIN EXPRESSION MEDIATED BY ENTEROVIRUS 71

659

controlled by cap-dependent and IRES-mediated mechanisms, respectively. These constructs were firstly examined with TNT experiments. The results are shown in Figure 2. Lanes 1 to 5 of Figure 2A indicate that all vectors produced high levels of h-galactosidase proteins (predicted molecular weight c119 kDa). As revealed by densitometric scanning, there was no difference in h-galactosidase levels from TNT experiments using different vectors (Fig. 2B). However, different IRES elements gave rise to different expression levels of SEAP proteins in the TNT experiments (Fig. 2A and 2C). Compared to the cap-dependent translation in a monocistronic construct (pTri-SEAP, lane 1), the expression of SEAP proteins (predicted molecular weight c56 kDa) driven by various viral IRES elements was very low (lanes 3 to 5), and the translated product of HCV IRES was even undetectable (lane 4). In these TNT experiments, protein translation mediated by EMCV IRES was more efficient than the translations mediated by EV71 and HCV IRES elements as revealed by the analysis of densitometric scanning (Fig. 2C). To compare the protein expression of various IRES elements in a cellular environment, transient protein expression experiments were performed in Huh-7, COS-7, and CHO cells. Cells were transfected by pGS-EMCV, pGS-HCV, and pGS-EV71, respectively. Cells and culture supernatants were harvested at 48 hours post-transfection and h-galactosidase and SEAP activities were measured. In all cell lines examined, there was no significant difference in h-galactosidase activities among these bicistronic constructs (Fig. 3A). The h-galactosidase activities from cell lysates of Huh-7 cells transfected with these bicistronic plasmids varied from 400,000 to 600,000 Light Emission Units. Similarly, the h-galactosidase activities varied approximately from 1,000,000 to 1,600,000 in COS-7 cell lysates and from 700,000 to 900,000 Light Emission Units in CHO cell lysates. These results were similar to the gene expression levels as examined by TNT in vitro conditions (Fig. 2). To evaluate the IRES-mediated protein expression, SEAP is used as the reporter gene. As revealed by the measured SEAP activities, EV71 IRES was found to be the most efficient element in driving protein translation. After normalizing SEAP activities with the corresponding h-galactosidase activities in cell lysates, we observed that protein expression of EV71 IRES is approximately 10- to 15-fold higher than that of EMCV IRES in the tested cell lines (Fig. 3B). In this experiment, the efficiency of HCV IRES was comparable to that of EMCV IRES in all these cell lines and showed that EMCV and HCV IRESes exhibited similar protein expression in mammalian cells. Interestingly, the amount of seap mRNA under enterovirus 71 (EV71) IRES control is about 1.8-fold lower than EMCV IRES and 3.2-fold lower than HCV IRES, respectively (Fig. 4A). So the expression activity is powerful enough under enterovirus 71 (EV71) IRES control at about 23-fold than EMCV IRES (Fig. 4B). However, the translated product of EV71 IRES was around threefold lower than that of EMCV IRES (Fig. 2) in the in vitro TNT

660

Figure 3. Comparison of protein expression of different IRESes in mammalian cells. Huh-7, COS-7, and CHO cells were transfected with the indicated bicistronic vectors. Cells and culture media were harvested at 48 h post-transfection and h-galactosidase and SEAP activities were measured. (A) Cap-dependent translation was reflected by the measured hgalactosidase activity. (B) Cap-independent translation of IRES was indicated by the measured SEAP activity. SEAP activities were normalized against the h-galactosidase activities in cell lysates. The error bars represent standard deviations from three independent experiments.

experiments. Such discrepancy might be due to the fact that there were some cellular factors required for enhancing EV71 IRES activity, and these factors might be either absent or present in low concentrations in rabbit reticulocyte lysates. Similar phenomenon has been observed for hepatitis A virus IRES, whose activity was stimulated up to 12-fold when some proteins of liver cells were added to rabbit reticulocyte translation extracts (Glass et al., 1993). Thus, caution should be taken when studying IRES-mediated translation using the in vitro translation system. An internal ribosome entry site has many biotechnological applications. Several studies have developed bicistronic retroviral vectors co-expressing a drug-selectable marker gene (MDR1) with a second gene that has therapeutic efficacy for gene therapy (Lee et al., 1997; Sugimoto et al., 1995; Zhou et al., 1998). In these studies, IRES was utilized to facilitate co-expression of the dominant drugselectable marker with the therapeutic gene in a single bicistronic mRNA. It was shown that both the selectable

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 90, NO. 5, JUNE 5, 2005

tronic vectors created in this study would facilitate the preparation of recombinant protein for a variety of biotechnological applications.

References

Figure 4. Comparison of the amount of bicistronic mRNA linked with EMCV, HCV, and enterovirus 71 (EV71) IRES. Huh-7 cells were transfected with the same amounts of the indicated bicistronic vectors. Cells were harvested 48 h post-transfection and total RNA samples were prepared and used for q-PCR analysis with seap- and actin- specific primers. (A) Relative amount of bicistronic mRNA. The bicistronic RNAs were expressed under control of CMV promoter and normalized with the RNA of actin gene. (B) Cap-independent translation of IRES was indicated by the measured SEAP activity. SEAP activities were normalized against the RNA level ratio from indicated constructs. The error bars represent standard deviations from three independent experiments.

marker and the therapeutic gene could be efficiently expressed in a bicistronic retroviral vector. When the MDR1 gene was placed downstream of EMCV IRES, there was substantial reduction (1 to 1.5 log) in retroviral titers presumably due to the low level of IRES-mediated translation. Thus, it would be helpful to identify alternative IRES elements with higher protein expression. Along the same line, modulating protein expression levels is an important technique in the area of metabolic engineering where successful multiple gene expression in a heterologous host is often a prerequisite for success (Fussenegger and Betenbaugh, 2002). In addition, since several viral IRESes have been proposed as potential antiviral drug targets, our bicistronic construct can also be modified by inclusion of a target IRES for drug screening (Gallego and Varani, 2002). Finally, the expression of recombinant proteins in mammalian cell lines is a fundamental technique in modern biotechnology. The incorporation of IRES elements in expression plasmids provides a powerful tool to co-express multiple proteins within a single transcript. The IRES element of EMCV has been reported as the most efficient IRES elements among all piconavirus (Borman et al., 1997; Ramesh et al., 1996). By altering the spacing of the initiation codons present at the 3V end of the IRES would enhance the translation of the IRES element (Davies and Kaufman, 1992). In this study, we have identified a much more potent IRES derived from EV71, which is also a member of Picornaviridae (Hsiung and Wang, 2000). This finding may offer an alternative method for tuning protein expression levels by selecting appropriate IRES elements in the bi- or poly-cistronic expression systems. The bicis-

Bedard KM, Semler BL. 2004. Regulation of picornavirus gene expression. Microbes Infect 6(7):702–713. Borman AM, Bailly JL, Girard M, Kean KM. 1995. Picornavirus internal ribosome entry segments: Comparison of translation efficiency and the requirements for optimal internal initiation of translation in vitro. Nucleic Acids Res 23(18):3656–3663. Borman AM, Le Mercier P, Girard M, Kean KM. 1997. Comparison of picornaviral IRES-driven internal initiation of translation in cultured cells of different origins. Nucleic Acids Res 25(5):925–932. Ciafre SA, Barillari G, Bongiorno-Borbone L, Wannenes F, Izquierdo M, Farace MG. 2002. A tricistronic retroviral vector expressing natural antiangiogenic factors inhibits angiogenesis in vitro, but is not able to block tumor progression in vivo. Gene Ther 9(4):297–302. Davies MV, Kaufman RJ. 1992. The sequence context of the initiation codon in the encephalomyocarditis virus leader modulates efficiency of internal translation initiation. J Virol 66(4):1924–1932. De Felipe P, Izquierdo M. 2000. Tricistronic and tetracistronic retroviral vectors for gene transfer. Hum Gene Ther 11(13):1921–1931. Fussenegger M, Betenbaugh MJ. 2002. Metabolic engineering II. Eukaryotic systems. Biotechnol Bioeng 79(5):509–531. Fussenegger M, Moser S, Mazur X, Bailey JE. 1997. Autoregulated multicistronic expression vectors provide one-step cloning of regulated product gene expression in mammalian cells. Biotechnol Prog 13(6): 733–740. Gallego J, Varani G. 2002. The hepatitis C virus internal ribosome-entry site: A new target for antiviral research. Biochem Soc Trans 30(2): 140–145. Gan W, Rhoads RE. 1996. Internal initiation of translation directed by the 5V-untranslated region of the mRNA for eIF4G, a factor involved in the picornavirus-induced switch from cap-dependent to internal initiation. J Biol Chem 271(2):623–626. Glass MJ, Jia XY, Summers DF. 1993. Identification of the hepatitis A virus internal ribosome entry site: In vivo and in vitro analysis of bicistronic RNAs containing the HAV 5V noncoding region. Virology 193(2):842–852. Hennecke M, Kwissa M, Metzger K, Oumard A, Kroger A, Schirmbeck R, Reimann J, Hauser H. 2001. Composition and arrangement of genes define the strength of IRES-driven translation in bicistronic mRNAs. Nucleic Acids Res 29(16):3327–3334. Hsiung GD, Wang JR. 2000. Enterovirus infections with special reference to enterovirus 71. J Microbiol Immunol Infect 33(1):1–8. Iizuka N, Najita L, Franzusoff A, Sarnow P. 1994. Cap-dependent and capindependent translation by internal initiation of mRNAs in cell extracts prepared from Saccharomyces cerevisiae. Mol Cell Biol 14(11):7322–7330. Jackson RJ. 1988. RNA translation. Picornaviruses break the rules. Nature 334(6180):292–293. Jackson RJ, Kaminski A. 1995. Internal initiation of translation in eukaryotes: The picornavirus paradigm and beyond. RNA 1(10): 985–1000. Lee CG, Jeang KT, Martin MA, Pastan I, Gottesman MM. 1997. Efficient long-term coexpression of a hammerhead ribozyme targeted to the U5 region of HIV-1 LTR by linkage to the multidrug-resistance gene. Antisense Nucleic Acid Drug Dev 7(5):511–522. Lee JC, Shih YF, Hsu SP, Chang TY, Chen LH, Hsu JT. 2003. Development of a cell-based assay for monitoring specific hepatitis C virus NS3/4A protease activity in mammalian cells. Anal Biochem 316(2):162–170. Macejak DG, Sarnow P. 1991. Internal initiation of translation mediated by the 5V leader of a cellular mRNA. Nature 353(6339):90–94.

LEE ET AL.: PROTEIN EXPRESSION MEDIATED BY ENTEROVIRUS 71

661

Martinez-Salas E. 1999. Internal ribosome entry site biology and its use in expression vectors. Curr Opin Biotechnol 10(5):458–464. Mielke C, Tummler M, Schubeler D, von Hoegen I, Hauser H. 2000. Stabilized, long-term expression of heterodimeric proteins from tricistronic mRNA. Gene 254(1–2):1–8. Moser S, Rimann M, Fux C, Schlatter S, Bailey JE, Fussenegger M. 2001. Dual-regulated expression technology: a new era in the adjustment of heterologous gene expression in mammalian cells. J Gene Med 3(6): 529–549. Oh SK, Scott MP, Sarnow P. 1992. Homeotic gene Antennapedia mRNA contains 5V-noncoding sequences that confer translational initiation by internal ribosome binding. Genes Dev 6(9):1643–1653. Pao W, Klimstra DS, Fisher GH, Varmus HE. 2003. Use of avian retroviral vectors to introduce transcriptional regulators into mammalian cells for analyses of tumor maintenance. Proc Natl Acad Sci USA 100(15): 8764–8769. Peroni CN, Soares CR, Gimbo E, Morganti L, Ribela MT, Bartolini P. 2002. High-level expression of human thyroid-stimulating hormone in Chinese hamster ovary cells by co-transfection of dicistronic expression vectors followed by a dual-marker amplification strategy. Biotechnol Appl Biochem 35(Pt 1):19–26. Pestova TV, Kolupaeva VG, Lomakin IB, Pilipenko EV, Shatsky IN, Agol VI, Hellen CU. 2001. Molecular mechanisms of translation initiation in eukaryotes. Proc Natl Acad Sci USA 98(13):7029–7036. Ramesh N, Kim ST, Wei MQ, Khalighi M, Osborne WR. 1996. High-titer bicistronic retroviral vectors employing foot-and-mouth disease

662

virus internal ribosome entry site. Nucleic Acids Res 24(14): 2697–2700. Schlatter S, Fussenegger M. 2003. Novel CNBP- and La-based translation control systems for mammalian cells. Biotechnol Bioeng 81(1):1–12. Sugimoto Y, Aksentijevich I, Murray GJ, Brady RO, Pastan I, Gottesman MM. 1995. Retroviral coexpression of a multidrug resistance gene (MDR1) and human alpha-galactosidase A for gene therapy of Fabry disease. Hum Gene Ther 6(7):905–915. Tsukiyama-Kohara K, Iizuka N, Kohara M, Nomoto A. 1992. Internal ribosome entry site within hepatitis C virus RNA. J Virol 66(3):1476–1483. Vagner S, Galy B, Pyronnet S. 2001. Irresistible IRES. Attracting the translation machinery to internal ribosome entry sites. EMBO Rep 2(10):893–898. Vagner S, Gensac MC, Maret A, Bayard F, Amalric F, Prats H, Prats AC. 1995. Alternative translation of human fibroblast growth factor 2 mRNA occurs by internal entry of ribosomes. Mol Cell Biol 15(1): 35–44. Weber W, Fussenegger M. 2002. Artificial mammalian gene regulation networks—Novel approaches for gene therapy and bioengineering. J Biotechnol 98(2–3):161–187. Wen F, Cecena G, Munoz-Ritchie V, Fuchs E, Chambon P, Oshima RG. 2003. Expression of conditional cre recombinase in epithelial tissues of transgenic mice. Genesis 35(2):100–106. Zhou Y, Aran J, Gottesman MM, Pastan I. 1998. Co-expression of human adenosine deaminase and multidrug resistance using a bicistronic retroviral vector. Hum Gene Ther 9(3):287–293.

BIOTECHNOLOGY AND BIOENGINEERING, VOL. 90, NO. 5, JUNE 5, 2005