Search tips
Search criteria 


Logo of jvirolPermissionsJournals.ASM.orgJournalJV ArticleJournal InfoAuthorsReviewers
J Virol. 1999 August; 73(8): 7031–7034.
PMCID: PMC112791

E4ORF3 Requirement for Achieving Long-Term Transgene Expression from the Cytomegalovirus Promoter in Adenovirus Vectors


Analysis of transgene expression under the control of the cytomegalovirus (CMV) promoter from adenovirus vectors in which the E4 region was modified indicated that E4ORF3 is required for long-term expression in the murine lung. CMV promoter truncation led to the persistence of expression in the absence of E4, thus eliminating the ORF3 requirement.

The influence of a wild-type E4 region on achieving persistent transgene expression in vivo from either the cytomegalovirus (CMV) or Rous sarcoma virus promoter in the context of recombinant adenovirus vectors has been reported (2, 6, 8). Our results indicated that E4 influences expression from the CMV promoter in trans, implicating the involvement of one or more E4 gene products. In this study, a series of vectors in which the E4 region was modified (Fig. (Fig.1A)1A) was constructed to determine which open reading frame(s) (ORF[s]) is required for achieving long-term expression from the CMV promoter in the mouse lung. Ad2/βgal-4 contains a wild-type E4 region, whereas Ad2/βgal-7 and -8 contain only ORF4 and ORF6 and -6/7, respectively. Ad2/βgal-9, -10, -11, -12, and -13 contain knockout mutations of ORF1, ORF2, ORF3, ORF4, and ORF6/7, respectively. Ad2/βgal-7, -8, -9, -10, -11, -12, and -13 were derived from dl366+ORF4, E4dlORF1-4, in351, in352, E4inORF3, dl358, and dl356, respectively, which were obtained from Tom Shenk and Pat Hearing (10, 11). Viruses were propagated and purified and the titers of the virus were determined, all as previously described (1). BALB/c nude mice, 7 to 16 weeks old (Taconic Farms, Germantown, N.Y.), were given intranasal instillations of 3 × 109 infectious units (i.u.) of recombinant virus in 100 μl of phosphate-buffered saline–3% sucrose. Mice were sacrificed on days 3 and 14, and β-galactosidase activity in the lungs was measured by using a chemiluminescent substrate (Galactolight Plus; Tropix, Bedford, Mass.) and was expressed as relative light units (2).

FIG. 1
Genomic structure of E4-modified adenovirus vectors. Expression cassettes replacing the E1 region are depicted on the left and variations of the E4 region are depicted on the right. (A) Vectors contain a CMV promoter element to drive β-galactosidase ...

In previous studies, β-galactosidase expression levels in the mouse lung on day 14 were higher than the day 3 levels from Ad2/βgal-4 (wild-type E4); in contrast, day 14 levels were lower than day 3 levels from analogous vectors in which E4 was deleted (2). Shown in Fig. Fig.22 are the day 14 β-galactosidase activity levels, represented as percentages of day 3 expression levels for the vector group shown in Fig. Fig.1A.1A. Expression levels of the different vectors were compared by analysis of variance by using the Tukey post hoc test for significance. The increases in expression levels in mice that received Ad2/βgal-9, -10, -12, and -13 were not statistically different from those in mice that received Ad2/βgal-4, indicating that ORF1, ORF2, ORF4, and ORF6 are dispensable for prolonged CMV promoter activity. In contrast, expression levels on day 14 were significantly lower than those on day 3 in animals that received Ad2/βgal-7, -8, or -11 when compared to Ad2/βgal-4 (P < 0.05). These results indicate that neither ORF4 nor the combination of ORF6 and ORF6/7 is sufficient but that ORF3 is required to achieve longevity of expression from the CMV promoter in the murine lung.

FIG. 2
β-Galactosidase expression in the lungs of immunodeficient mice that received vectors in the Ad2/βgal series. BALB/c nude mice were given intranasal instillations of 3 × 109 i.u. of each virus. Mice were sacrificed on days 3 and ...

Long-term expression of the human cystic fibrosis transmembrane conductance regulator (hCFTR) under the control of the CMV promoter has been achieved in the airway epithelia of immunocompetent mice with a first-generation adenovirus vector (wild-type E4) but not with an E4-modified vector (17). On the basis of results with the Ad2/βgal vector series, we hypothesized that an analogous CFTR expression vector retaining ORF3 would yield long-term expression in the mouse lung. Vectors with the following E4 modifications (Fig. (Fig.1B)1B) were tested in parental BALB/c mice: Ad2/CFTR-16 (wild-type E4) (17), Ad2/CFTR-5 (ORF6) (12), and Ad2/CFTR-18 (ORF3 and ORF4). In Ad2/CFTR-18, E4 sequences (positions 32891 to 35470) were deleted and replaced with a fragment (positions 33998 to 34804) containing ORF3 and ORF4 coding sequences and 3′ mRNA splice sites. ORF4 protects transduced cells from lysis by cytotoxic T lymphocytes in vitro (13) and was included for potential beneficial effects in immunocompetent mice. As shown in Fig. Fig.3,3, expression based on reverse transcription-PCR (2) persisted to day 42 in animals that received either Ad2/CFTR-16 or Ad2/CFTR-18 but was transient in animals that received Ad2/CFTR-5. This indicates that the retention of ORF3 in an adenovirus vector allows expression from the CMV promoter to persist. A vector containing only ORF3 has not yet been obtained; however, the data with the Ad2/βgal series (Fig. (Fig.2)2) indicate that ORF4 is neither sufficient nor required. While a contribution from ORF4 cannot completely be ruled out, the data collectively suggest that ORF3 may be sufficient for achieving long-term expression from the CMV promoter.

FIG. 3
Persistence of hCFTR expression in the lungs of immunocompetent mice. BALB/c mice were given intranasal instillations of 3 × 109 i.u. of Ad2/CFTR-5, -16, or -18 and were sacrificed on days 3, 21, and 42 postinstillation. RNA samples from lungs ...

The above-described studies demonstrate that ORF3 activity is required for prolonged expression from the CMV promoter; however, it is not clear if this results in promoter activation or prevention of promoter repression. The CMV promoter has been extensively studied and many transcription factor binding sites have been identified (4, 15). For example, it has been demonstrated that YY1 can mediate repression of expression from a full-length, but not from a truncated, CMV promoter in nonpermissive cells (14). We therefore chose to determine if truncation of the CMV promoter (Fig. (Fig.1C)1C) would prevent the down regulation of expression in the context of an adenovirus vector containing a complete E4 deletion. The expression from ΔCMVβgal-1 was compared to that from Ad2/βgal-4 in the lungs of BALB/c nude mice. In order to determine if the enhancer elements remaining in the ΔCMV promoter were still responsive to E4, Ad2/CFTR-16 that could supply E4 in trans was coinstilled into one group of mice. As shown in Fig. Fig.4,4, the expression remained elevated in mice that received either Ad2/βgal-4, ΔCMVβgal-1, or the ΔCMVβgal-1–Ad2/CFTR-16 combination. This indicates that CMV promoter truncation was sufficient to prevent a decline in the expression which has been observed previously with Ad2/βgal-5 (complete E4 deletion) (2). Moreover, the data suggest that the promoter fragment from positions −295 to −14 does not require E4 gene products for continued expression and that sequences within positions −523 to −296 are involved in down regulation. The results also imply that E4ORF3 may act to relieve CMV promoter repression in the mouse lung. In addition, the day 3 levels of expression did not differ significantly among the groups. This is somewhat surprising since a considerable portion of the enhancer is deleted in the ΔCMV promoter; however, a similar truncation did not appear to affect expression in cultured cells (14).

FIG. 4
β-Galactosidase expression in the lungs of immunodeficient mice. BALB/c nude mice were given intranasal instillations of 3 × 109 i.u. of either Ad2/βgal-4 or ΔCMVβgal-1. One cohort of animals received ΔCMVβgal-1 ...

Although a role for ORF3 in the regulation of expression is suggested by our results, the mechanism remains obscure. Despite knowledge of ORF3 action and function with respect to adenovirus replication (5, 7, 9, 11, 16), it is unclear how they relate to the effect observed in our studies. Nonetheless, the down regulation of the CMV promoter in the context of adenovirus vectors is consistent with the tissue distribution of the CMV-driven LacZ gene (3) or neomycin resistance gene (18) expression observed in transgenic mice. Reporter expression was detected in a variety of tissues but not in the lung or liver, suggesting active CMV promoter silencing in these tissues.

Gene therapy applications focusing on the correction of genetic defects will require prolonged transgene expression to avoid problems associated with vector readministration. Achievement of long-term CFTR expression in the lungs of BALB/c mice with both first- and second-generation adenovirus vectors is demonstrated here and elsewhere (17). Although the results are encouraging, it is likely that immune responses to viral antigens (1921) that eliminate transduced cells will be a limiting factor in more permissive models. Our results suggest that E4ORF3 is required to achieve prolonged expression from the CMV promoter in the murine lung. Because ORF3 influences expression from the CMV promoter, cellular gene expression may similarly be affected. Until more is known about ORF3 function, it is not possible to predict consequences of ORF3 expression in other models. We also demonstrate that expression from a truncated CMV promoter can be maintained in the absence of any E4 gene product, suggesting that the combination of this promoter with a complete E4 deletion might be used as an alternative to vectors with a longer CMV promoter that requires the retention of ORF3. A better understanding of the interplay between ORF3 and the CMV promoter in a variety of tissues may help define strategies for controlling transgene expression. While many gene therapy applications focus on achieving long-term expression, others may require transient expression. Therefore, the retention or deletion of ORF3 in gene delivery vectors might be a useful mechanism for controlling the outcome of transgene expression.


We thank Kathy Hehir and Denise Pratt in Virus Production for preparations of all vectors used in this work. We also thank Margaret Stedman and Malinda Plog for expression analysis in mice and Carol Sacks and Amy Gates for technical assistance with animal studies.


1. Armentano D, Sookdeo C C, Hehir K M, Gregory R J, St. George J A, Prince G A, Wadsworth S C, Smith A E. Characterization of an adenovirus gene transfer vector containing an E4 deletion. Hum Gene Ther. 1995;6:1343–1353. [PubMed]
2. Armentano D, Zabner J, Sacks C, Sookdeo C C, Smith M P, St. George J A, Wadsworth S C, Smith A E, Gregory R J. Effect of the E4 region on the persistence of transgene expression from adenovirus vectors. J Virol. 1997;71:2408–2416. [PMC free article] [PubMed]
3. Baskar J F, Smith P P, Nilaver G, Jupp R A, Hoffmann S, Peffer N J, Tenney D J, Colberg-Poley A M, Ghazal P, Nelson J A. The enhancer domain of the human cytomegalovirus major immediate-early promoter determines cell type-specific expression in transgenic mice. J Virol. 1996;70:3207–3214. [PMC free article] [PubMed]
4. Boshart M, Weber F, Jahn G, Dorsch-Hasler K, Fleckenstein B, Schaffner W. A very strong enhancer is located upstream of an immediate early gene of human cytomegalovirus. Cell. 1985;41:521–530. [PubMed]
5. Bridge E, Ketner G. Redundant control of adenovirus late gene expression by early region 4. J Virol. 1989;63:631–638. [PMC free article] [PubMed]
6. Brough D E, Hsu C, Kulesa V A, Lee G M, Cantolupo L J, Lizonova A, Kovesdi I. Activation of transgene expression by early region 4 is responsible for a high level of persistent transgene expression from adenovirus vectors in vivo. J Virol. 1997;71:9206–9213. [PMC free article] [PubMed]
7. Carvalho T, Seeler J S, Ohman K, Jordan P, Pettersson U, Akusjarvi G, Carmo-Fonseca M, Dejean A. Targeting of adenovirus E1A and E4-ORF3 proteins to nuclear matrix-associated PML bodies. J Cell Biol. 1995;131:45–56. [PMC free article] [PubMed]
8. Dedieu J-F, Vigne E, Torrent C, Jullien C, Mahfouz I, Caillaud J-M, Aubailly N, Orsini C, Guillaume J-M, Opolon P, Delaère P, Perricaudet M, Yeh P. Long-term gene delivery into the livers of immunocompetent mice with E1/E4-defective adenoviruses. J Virol. 1997;71:4626–4637. [PMC free article] [PubMed]
9. Doucas V, Ishov A M, Romo A, Juguilon H, Weitzman M D, Evans R M, Maul G G. Adenovirus replication is coupled with the dynamic properties of the PML nuclear structure. Genes Dev. 1996;10:196–207. [PubMed]
10. Halbert D N, Cutt J R, Shenk T. Adenovirus early region 4 encodes functions required for efficient DNA replication, late gene expression, and host cell shutoff. J Virol. 1985;56:250–257. [PMC free article] [PubMed]
11. Huang M-M, Hearing P. Adenovirus early region 4 encodes two gene products with redundant effects in lytic infection. J Virol. 1989;63:2605–2615. [PMC free article] [PubMed]
12. Jiang C, O’Connor S P, Armentano D, Berthelette P B, Schiavi S C, Jefferson D M, Smith A E, Wadsworth S C, Cheng S H. Ability of adenovirus vectors containing different CFTR transcriptional cassettes to correct ion transport defects in CF cells. Am J Physiol. 1996;271:L527–L537. [PubMed]
13. Kaplan J M, Armentano D, Scaria A, Woodworth L A, Pennington S E, Wadsworth S C, Smith A E, Gregory R J. Novel role for E4 region genes in protection of adenovirus vectors from lysis by cytotoxic T lymphocytes. J Virol. 1999;73:4489–4492. [PMC free article] [PubMed]
14. Liu R, Baillie J, Sissons J G, Sinclair J H. The transcription factor YY1 binds to negative regulatory elements in the human cytomegalovirus major immediate early enhancer/promoter and mediates repression in nonpermissive cells. Nucleic Acids Res. 1994;22:2453–2459. [PMC free article] [PubMed]
15. Niller H H, Hennighausen L. Formation of several specific nucleoprotein complexes on the human cytomegalovirus immediate early enhancer. Nucleic Acids Res. 1991;19:3715–3721. [PMC free article] [PubMed]
16. Nordqvist K, Öhman K, Akusjärvi G. Human adenovirus encodes two proteins which have opposite effects on accumulation of alternatively spliced mRNAs. Mol Cell Biol. 1994;14:437–445. [PMC free article] [PubMed]
17. Scaria A, St. George J A, Jiang C, Kaplan J M, Wadsworth S C, Gregory R J. Adenovirus-mediated persistent cystic fibrosis transmembrane conductance regulator expression in mouse airway epithelium. J Virol. 1998;72:7302–7309. [PMC free article] [PubMed]
18. Schmidt E V, Christoph G, Zeller R, Leder P. The cytomegalovirus enhancer: a pan-active control element in transgenic mice. Mol Cell Biol. 1990;10:4406–4411. [PMC free article] [PubMed]
19. Tripathy S K, Black H B, Goldwasser E, Leiden J M. Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replication-defective adenovirus vectors. Nat Med. 1996;2:545–550. [PubMed]
20. Yang Y, Nunes F A, Berencsi K, Gonczol E, Engelhardt J F, Wilson J M. Inactivation of E2a in recombinant adenoviruses improves the prospect for gene therapy in cystic fibrosis. Nat Genet. 1994;7:362–369. [PubMed]
21. Yang Y, Su Q, Wilson J M. Role of viral antigens in destructive cellular immune responses to adenovirus vector-transduced cells in mouse lungs. J Virol. 1996;70:7209–7212. [PMC free article] [PubMed]

Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)