- Email: [email protected]
1142. Regulation of Transgene Expression in Mouse Liver
GENE REGULATION: PROMOTER AND CONSTRUCT DESIGN agents (5.0-, 3.3- and 6.0-fold, respectively, at 50 mM 5FU and 6.6, 4.8-, and 8.0-fold, respectively, at 5 mM hydroxyurea) but not by alkylating agents (100 mM carboplatinum and 100 mM cisplatinum). Viral uptakes in cells are determined by viral DNA contents using real-time PCR and appear to be upregulated by pre-treatment of cells with 5FU (2.0-, 1.5-, and 2.0-fold, respectively) and hydroxyurea (2.5-, 1.8-, and 2.1-fold, respectively). In addition, viral infections in cells prior to drug treatment result in greater induction for b-gal activities, suggesting that intracellular activation of viral gene is also involved. By using real-time PCR, we found that viral DNA replicates in LoVo and SMMC7721 at 48 h after infection (5.3- and 5.7-fold, respectively) whereas no replication is detected in WB cells or drug-treated cells. In contrast, mRNA levels for b-gal determined by real-time RT-PCR analysis are significantly elevated in LoVo, SMMC7721 and WB cells treated by 5FU (36.0-, 6.0-, and 5.0-fold, respectively) and hydroxyurea (4.0-, 11.0-, and 10-fold, respectively) as compared to non-treated control cells. The level of mRNA for host gene a-actin remains unchanged, indicating these drugs may affect viral mRNA in dependent of cellular transcription. Our findings provide scientific basis for the combination of enzyme/ prodrug adenovirus gene therapy and chemotherapy in improving selectivity and efficiency in cancer treatment.
1140. Characterization of Sub-Nuclear and PostMitotic Plasmid Trafficking in Microinjected Cells Joshua Z. Gasiorowski,1 David A. Dean.1 1 Pulmonary Medicine, Northwestern University, Chicago, IL, United States. One of the least understood aspects of non-viral gene therapy is the role of sub-nuclear plasmid localization. Plasmids successfully delivered across the plasma membrane must still traffic through the cytoplasm and into the dense, highly organized nucleus before any transgene expression can occur. Even if gene delivery is successful, episomal plasmids will cease to express if they fail to come into contact with the appropriate cellular transcription factors or are later excluded from the nucleus. We used nuclear microinjection experiments in conjunction with several different fluorescent labeling techniques and in situ hybridizations to track sub-nuclear plasmid movement. pEGFP-N1 injected directly into the nucleus moves into an organized speckle pattern over time. After injection, most of these plasmid speckles co-localize with transcriptional machinery within 30-240 minutes whereas a plasmid containing no promoter or transgene (pBR322) shows little or no co-localization with the same factors. The timeframe that transcriptional machinery colocalized with pEGFP-N1 also corresponded with the first visible detection of transgene. Therefore, expression level may be linked to co-localization of the plasmid with a host of cellular proteins within discreet sub-nuclear regions. It is currently unclear as to whether the plasmid actually moves into areas that are dense in transcriptional machinery or instead recruits the necessary cellular proteins to it. We also tracked nuclear injected plasmids after a round of cell division using several different fluorescent labeling methods. We show that the post-mitotic location of plasmids differed depending on the type of labeling method used. Unlabeled plasmids detected by in situ hybridization or labeled with Cy3 protein nucleic acid (PNA) were distributed throughout the nuclei of divided cells while plasmids fluorescently labeled with commercial kits were excluded from the nucleus after cell division indicating that different fluorophores and labeling techniques may interfere with certain protein-DNA interactions that maintain the post-mitotic nuclear retention of episomal plasmids.
S440
1141. Insulin-Like Growth Factor-1 Enhances Plasmid Transgene Expression through a Phosphatidylinositol 3-Kinase-Dependent Pathway Edward Chaum,1,2,3 Xiuying Yang,1 Huaitao Yang.1 Ophthalmology; 2Pediatrics; 3Anatomy and Neurobiology, University of Tennessee Health Science Center, Memphis, TN, United States. 1
Introduction. The goal of our work is to modulate the phenotype of the human retinal pigment epithelium (RPE) using transgenic growth factors, to treat degenerative diseases of the retina. We have previously shown that expression of an IGF-1 transgene enhances the proliferative potential of cloned RPE cells in vitro in a dosedependent manner. Recent studies of these clones showed that IGF1-transduction also enhances transient transgene expression via a phosphatidylinositol 3-kinase (PI3-K)-dependent pathway. Methods. Naive human RPE cells and IGF-1-transduced RPE clones expressing increased levels of a biologically active IGF-1 transgene were transfected by lipofection with the pGeneGrip plasmid that encodes a CMV-promoted green fluorescent protein (GFP). The efficiency of plasmid-mediated transgene expression in the RPE was measured by quantifying GFP-fluorescence using flow cytometry 48 hours after transfection. Transfection studies were performed in parallel under low-serum conditions, at confluence and in the presence of the PI3-K inhibitor, LY294002. Results. IGF-1-treated RPE cells and IGF-1-transduced RPE clones c14, c62, and c49 demonstrated an enhanced and dosedependent increase in GFP transgene expression following transfection, compared to naive controls (treated, P < 0.001; c14, P < 0.003, c62, P < 0.001, c49, P < 0.001; paired samples, t-test). The enhanced efficiency of transgene expression was more pronounced in cycling subconfluent cells transfected under low-serum conditions, but was also seen in non-cycling, confluent RPE clones (c14, P < 0.038, c62, P < 0.003, c49, P < 0.007). Culturing the cells in the presence of the PI3-K inhibitor, LY294002 significantly inhibited the increased efficiency of transgene expression in naive cells and IGF-1-transduced RPE clones (naive, P < 0.022, c14, P < 0.006, c62, P < 0.008, c49, P < 0.006). Conclusions. These studies show a direct and dose-dependent correlation between the level of IGF-1 expression and enhanced transgene expression following transient lipofection in RPE cells and IGF-1 transduced RPE clones. Inhibition of the PI3-K pathway, a downstream arm of the IGF-1 signal transduction cascade, blocks this biological effect of IGF-1 gene expression. Enhancement of transgene expression is a previously unrecognized biological effect of IGF-1 signal transduction. Possible molecular mechanisms of action are discussed.
1142. Regulation of Transgene Expression in Mouse Liver Mohammed Al-dosari,1 Guisheng Zhang,1 Joseph Knapp,1 Dexi Liu.1 1 Pharmaceutical Sciences, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, United States. Understanding the mechanisms involved in regulating gene expression in a target specific manner is a critical part of developing practical applications in gene therapy. Transgene expression was studied in the liver of mice transfected with various plasmid constructs by the hydrodynamics-based procedure. Thirteen constructs were examined, including those containing viral promoters/ enhancer sequences (CMV, CMV/EBV, RSV), or those of liver specific regulatory sequences (human alpha 1-antitrypsin promoter, bovine serum albumin promoter, 5’-end flanking sequences of cytochrome P450 genes of 1A2, 2B10, 2C9, 2C18, 2D6, and 3A4). In addition, Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts
Copyright ® The American Society of Gene Therapy
GENE REGULATION: PROMOTER AND CONSTRUCT DESIGN promoters of HS70 and NFkB gene were also tested for their function in regulating the level and persistency of reporter gene expression. Among those constructs examined, the CMV and hAAT promoters exhibited the highest level of reporter gene expression 6 hours post transfection. Although originally derived from the liver, promoters of serum albumin and those of cytochrome P450 genes were 10 to 100 fold less effective. The lowest level of reporter gene expression was seen in construct containing 5’-flanking sequence of cytochrome P450 2C18 gene. Gene product level with time for all of the constructs tested was similar, with a rapid initial decline followed by a slow declining phase at a lower level. Southern blot analysis revealed that plasmids injected stayed as episomal form in all of transfected animals, regardless the type of promoter used. However, Northern blot analysis revealed that the mRNA level of reporter gene fell below the detection limit one day after transfection for all of the plasmid constructs with the exception of plasmid containing EBV sequences, which remained at high level for more than a week. These results suggest that transgene shut-down is a common phenomenon for all types of promoters. EBV sequences in the plasmid construct are able to delay the shut-down process. Regulation of transgene expression is not dependent on whether a tissue specific promoter is utilized. The fact that constructs containing NFkB or HS70 gene promoter exhibited the most transient gene expression indicates that transgene expression is regulated by components involving in stress-related pathways.
1143. Biological Efficacy and Toxicity of Naked Plasmid DNA and Cationic Liposome-DNA Complexes in Ovine Lung Louise Renwick,1 Steve Tate,1 Hazel Painter,2 Deborah Gill,2 Uta Griesenbach,3 Chris Boyd,1 Micheal Emerson,1 Seng Cheng,5 David Collie.4 1 Department of Medical Sciences, The University of Edinburgh, Edinburgh, United Kingdom; 2Gene Medicine Group, Nuffield Department of Clinical Laboratory Sciences, University of Oxford, Oxford, United Kingdom; 3Gene Therapy, Imperial College London, London, United Kingdom; 4Veterinary Clinical Studies, University of Edinburgh, Edinburgh, United Kingdom; 5 Genzyme Corporation, Frammingham. The overall efficiency of liposome-mediated gene transfer to the respiratory epithelium is sub optimal with respect to both the extent and duration of transgene expression. To identify the factors likely to contribute to this inefficiency, we defined, using a novel large animal model system, the acute pathological response to local lung administration of plasmid DNA expressing chloramphenicol acetyl transferase (CAT) either as naked plasmid DNA (pDNA) or cationic liposome pDNA complexes (pDNA: GL67) and related such responses to concomitant indicators of biological efficacy, namely levels of CAT protein and mRNA in specific lung tissue compartments. We instilled doses of 0.2, 1, 5 and 25 mg pDNA to spatially distinct lung segments in six anaesthetised sheep and doses of 0.2, 1 and 5 mg pDNA: GL67 to a further six sheep. Twenty four hours after gene delivery the sheep were euthanased and necropsy examination with sampling of relevant tissues carried out. Levels of plasmid derived CAT-specific mRNA and CAT protein in samples derived from segments treated with either pDNA or pDNA:GL67, increased in relation to the administered dose. Levels of mRNA and protein expression were greater for pDNA:GL67 than for pDNA alone. A significant correlation was observed between mRNA and protein expression in samples derived from airways treated with pDNA and pDNA:GL67. Histopathological changes following the administration of both pDNA and pDNA:GL67 were characterised by a neutrophilic inflammation Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts Copyright ® The American Society of Gene Therapy
predominantly orientated on airways. The severity of the inflammatory response appeared to correlate with the administered dose of DNA and was generally more severe for pDNA:GL67. This study demonstrates a within-organ dose response at the level of mRNA and protein expression following local lung segmental instillation of either pDNA or DNA complexed with cationic liposomes. Furthermore, such transfection, in particularly that mediated through cationic liposomes, was associated with significant toxicity. To our knowledge this is the first reported instance of a correlation existing between transgene-specific mRNA and protein expression following gene transfer in a large animal model. Thus, our findings support the practical relevance of studies involving local delivery to the ovine lung in evaluating different transfer strategies.
1144. Evaluation of Viral and Mammalian Promoters for Use in Gene Delivery to Salivary Glands Changyu Zheng,1 Bruce J. Baum.1 Gene Therapy and Therapeutics Branch, NIDCR, NIH, Bethesda, MD, United States. 1
Many studies on the development of gene delivery vectors have demonstrated the importance of promoter selection for optimizing therapeutic outcomes. For example, promoter choice can lead to enhanced as well as persistent transgene expression (e.g., Van Linthout et al, Hum. Gene Ther., 2002). Additionally, use of a cell or tissue-specific promoter can provide a significant level of patient safety in the event of unwanted vector dissemination beyond the targeted tissue. Over the last decade, studies from our laboratory have established several potential clinical applications of gene transfer to salivary glands (see Baum et al, Int. Rev Cytol., 2002). As a key step in optimizing vector delivery to salivary glands, we have begun to examine the effect of promoter selection on transgene expression. Initially, we screened the efficacy of several viral and mammalian cellular promoters in driving luciferase activity in an established, well-characterized rat submandibular gland cell line, A5. Four viral promoters were tested: cytomegalovirus (CMV, 0.59 kb), Rous Sarcoma Virus (RSV, 0.41 kb), SV40 (0.31 kb) and the long terminal repeat from Moloney murine leukemia virus (LTR, 0.61 kb). Six mammalian cellular promoters were tested including the promoters for human elongation factor 1a (EF1a, 1.2 kb), human cytokeratin 18 (K18, 2.5 kb), human cytokeratin 19 (K19, 2.95 kb), rat aquaporin-5 (rAQP5, 4.5 kb), human amylase (AMY, 1 kb) and human kallikein (KALL, 0.35 kb). The AMY promoter, although weak, is relatively salivary gland specific, while the kallikrein promoter is stronger, but active in several other epithelial tissues (Zheng et al., Hum. Gene Ther., 2001). All promoters were placed upstream of a luciferase cassette in pAC. This cassette included approximately 1.8 kb luciferase cDNA (obtained from pGL2-Basic, Promega) as a reporter gene, along with the SV40 polyadenylation signal. Luciferase activity was measured in a chemiluminescent assay. Results (n=3 experiments, each in triplicate) showed that the order of promoter activity in A5 cells was, from highest to lowest, CMV>EF1a>SV40=RSV>K18>K19>KALL>LTR=AMY. Future experiments will include examination of promoter activity in rat submandibular glands in vivo using transgene transfection methods (O’Connell et al., Amer. J. Physiol., 1995). We anticipate that these and related studies will permit development of optimized expression cassettes for vector-mediated gene transfer to salivary glands.
S441
1140. Characterization of Sub-Nuclear and PostMitotic Plasmid Trafficking in Microinjected Cells Joshua Z. Gasiorowski,1 David A. Dean.1 1 Pulmonary Medicine, Northwestern University, Chicago, IL, United States. One of the least understood aspects of non-viral gene therapy is the role of sub-nuclear plasmid localization. Plasmids successfully delivered across the plasma membrane must still traffic through the cytoplasm and into the dense, highly organized nucleus before any transgene expression can occur. Even if gene delivery is successful, episomal plasmids will cease to express if they fail to come into contact with the appropriate cellular transcription factors or are later excluded from the nucleus. We used nuclear microinjection experiments in conjunction with several different fluorescent labeling techniques and in situ hybridizations to track sub-nuclear plasmid movement. pEGFP-N1 injected directly into the nucleus moves into an organized speckle pattern over time. After injection, most of these plasmid speckles co-localize with transcriptional machinery within 30-240 minutes whereas a plasmid containing no promoter or transgene (pBR322) shows little or no co-localization with the same factors. The timeframe that transcriptional machinery colocalized with pEGFP-N1 also corresponded with the first visible detection of transgene. Therefore, expression level may be linked to co-localization of the plasmid with a host of cellular proteins within discreet sub-nuclear regions. It is currently unclear as to whether the plasmid actually moves into areas that are dense in transcriptional machinery or instead recruits the necessary cellular proteins to it. We also tracked nuclear injected plasmids after a round of cell division using several different fluorescent labeling methods. We show that the post-mitotic location of plasmids differed depending on the type of labeling method used. Unlabeled plasmids detected by in situ hybridization or labeled with Cy3 protein nucleic acid (PNA) were distributed throughout the nuclei of divided cells while plasmids fluorescently labeled with commercial kits were excluded from the nucleus after cell division indicating that different fluorophores and labeling techniques may interfere with certain protein-DNA interactions that maintain the post-mitotic nuclear retention of episomal plasmids.
S440
1141. Insulin-Like Growth Factor-1 Enhances Plasmid Transgene Expression through a Phosphatidylinositol 3-Kinase-Dependent Pathway Edward Chaum,1,2,3 Xiuying Yang,1 Huaitao Yang.1 Ophthalmology; 2Pediatrics; 3Anatomy and Neurobiology, University of Tennessee Health Science Center, Memphis, TN, United States. 1
Introduction. The goal of our work is to modulate the phenotype of the human retinal pigment epithelium (RPE) using transgenic growth factors, to treat degenerative diseases of the retina. We have previously shown that expression of an IGF-1 transgene enhances the proliferative potential of cloned RPE cells in vitro in a dosedependent manner. Recent studies of these clones showed that IGF1-transduction also enhances transient transgene expression via a phosphatidylinositol 3-kinase (PI3-K)-dependent pathway. Methods. Naive human RPE cells and IGF-1-transduced RPE clones expressing increased levels of a biologically active IGF-1 transgene were transfected by lipofection with the pGeneGrip plasmid that encodes a CMV-promoted green fluorescent protein (GFP). The efficiency of plasmid-mediated transgene expression in the RPE was measured by quantifying GFP-fluorescence using flow cytometry 48 hours after transfection. Transfection studies were performed in parallel under low-serum conditions, at confluence and in the presence of the PI3-K inhibitor, LY294002. Results. IGF-1-treated RPE cells and IGF-1-transduced RPE clones c14, c62, and c49 demonstrated an enhanced and dosedependent increase in GFP transgene expression following transfection, compared to naive controls (treated, P < 0.001; c14, P < 0.003, c62, P < 0.001, c49, P < 0.001; paired samples, t-test). The enhanced efficiency of transgene expression was more pronounced in cycling subconfluent cells transfected under low-serum conditions, but was also seen in non-cycling, confluent RPE clones (c14, P < 0.038, c62, P < 0.003, c49, P < 0.007). Culturing the cells in the presence of the PI3-K inhibitor, LY294002 significantly inhibited the increased efficiency of transgene expression in naive cells and IGF-1-transduced RPE clones (naive, P < 0.022, c14, P < 0.006, c62, P < 0.008, c49, P < 0.006). Conclusions. These studies show a direct and dose-dependent correlation between the level of IGF-1 expression and enhanced transgene expression following transient lipofection in RPE cells and IGF-1 transduced RPE clones. Inhibition of the PI3-K pathway, a downstream arm of the IGF-1 signal transduction cascade, blocks this biological effect of IGF-1 gene expression. Enhancement of transgene expression is a previously unrecognized biological effect of IGF-1 signal transduction. Possible molecular mechanisms of action are discussed.
1142. Regulation of Transgene Expression in Mouse Liver Mohammed Al-dosari,1 Guisheng Zhang,1 Joseph Knapp,1 Dexi Liu.1 1 Pharmaceutical Sciences, University of Pittsburgh School of Pharmacy, Pittsburgh, PA, United States. Understanding the mechanisms involved in regulating gene expression in a target specific manner is a critical part of developing practical applications in gene therapy. Transgene expression was studied in the liver of mice transfected with various plasmid constructs by the hydrodynamics-based procedure. Thirteen constructs were examined, including those containing viral promoters/ enhancer sequences (CMV, CMV/EBV, RSV), or those of liver specific regulatory sequences (human alpha 1-antitrypsin promoter, bovine serum albumin promoter, 5’-end flanking sequences of cytochrome P450 genes of 1A2, 2B10, 2C9, 2C18, 2D6, and 3A4). In addition, Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts
Copyright ® The American Society of Gene Therapy
GENE REGULATION: PROMOTER AND CONSTRUCT DESIGN promoters of HS70 and NFkB gene were also tested for their function in regulating the level and persistency of reporter gene expression. Among those constructs examined, the CMV and hAAT promoters exhibited the highest level of reporter gene expression 6 hours post transfection. Although originally derived from the liver, promoters of serum albumin and those of cytochrome P450 genes were 10 to 100 fold less effective. The lowest level of reporter gene expression was seen in construct containing 5’-flanking sequence of cytochrome P450 2C18 gene. Gene product level with time for all of the constructs tested was similar, with a rapid initial decline followed by a slow declining phase at a lower level. Southern blot analysis revealed that plasmids injected stayed as episomal form in all of transfected animals, regardless the type of promoter used. However, Northern blot analysis revealed that the mRNA level of reporter gene fell below the detection limit one day after transfection for all of the plasmid constructs with the exception of plasmid containing EBV sequences, which remained at high level for more than a week. These results suggest that transgene shut-down is a common phenomenon for all types of promoters. EBV sequences in the plasmid construct are able to delay the shut-down process. Regulation of transgene expression is not dependent on whether a tissue specific promoter is utilized. The fact that constructs containing NFkB or HS70 gene promoter exhibited the most transient gene expression indicates that transgene expression is regulated by components involving in stress-related pathways.
1143. Biological Efficacy and Toxicity of Naked Plasmid DNA and Cationic Liposome-DNA Complexes in Ovine Lung Louise Renwick,1 Steve Tate,1 Hazel Painter,2 Deborah Gill,2 Uta Griesenbach,3 Chris Boyd,1 Micheal Emerson,1 Seng Cheng,5 David Collie.4 1 Department of Medical Sciences, The University of Edinburgh, Edinburgh, United Kingdom; 2Gene Medicine Group, Nuffield Department of Clinical Laboratory Sciences, University of Oxford, Oxford, United Kingdom; 3Gene Therapy, Imperial College London, London, United Kingdom; 4Veterinary Clinical Studies, University of Edinburgh, Edinburgh, United Kingdom; 5 Genzyme Corporation, Frammingham. The overall efficiency of liposome-mediated gene transfer to the respiratory epithelium is sub optimal with respect to both the extent and duration of transgene expression. To identify the factors likely to contribute to this inefficiency, we defined, using a novel large animal model system, the acute pathological response to local lung administration of plasmid DNA expressing chloramphenicol acetyl transferase (CAT) either as naked plasmid DNA (pDNA) or cationic liposome pDNA complexes (pDNA: GL67) and related such responses to concomitant indicators of biological efficacy, namely levels of CAT protein and mRNA in specific lung tissue compartments. We instilled doses of 0.2, 1, 5 and 25 mg pDNA to spatially distinct lung segments in six anaesthetised sheep and doses of 0.2, 1 and 5 mg pDNA: GL67 to a further six sheep. Twenty four hours after gene delivery the sheep were euthanased and necropsy examination with sampling of relevant tissues carried out. Levels of plasmid derived CAT-specific mRNA and CAT protein in samples derived from segments treated with either pDNA or pDNA:GL67, increased in relation to the administered dose. Levels of mRNA and protein expression were greater for pDNA:GL67 than for pDNA alone. A significant correlation was observed between mRNA and protein expression in samples derived from airways treated with pDNA and pDNA:GL67. Histopathological changes following the administration of both pDNA and pDNA:GL67 were characterised by a neutrophilic inflammation Molecular Therapy Vol. 7, No. 5, May 2003, Part 2 of 2 Parts Copyright ® The American Society of Gene Therapy
predominantly orientated on airways. The severity of the inflammatory response appeared to correlate with the administered dose of DNA and was generally more severe for pDNA:GL67. This study demonstrates a within-organ dose response at the level of mRNA and protein expression following local lung segmental instillation of either pDNA or DNA complexed with cationic liposomes. Furthermore, such transfection, in particularly that mediated through cationic liposomes, was associated with significant toxicity. To our knowledge this is the first reported instance of a correlation existing between transgene-specific mRNA and protein expression following gene transfer in a large animal model. Thus, our findings support the practical relevance of studies involving local delivery to the ovine lung in evaluating different transfer strategies.
1144. Evaluation of Viral and Mammalian Promoters for Use in Gene Delivery to Salivary Glands Changyu Zheng,1 Bruce J. Baum.1 Gene Therapy and Therapeutics Branch, NIDCR, NIH, Bethesda, MD, United States. 1
Many studies on the development of gene delivery vectors have demonstrated the importance of promoter selection for optimizing therapeutic outcomes. For example, promoter choice can lead to enhanced as well as persistent transgene expression (e.g., Van Linthout et al, Hum. Gene Ther., 2002). Additionally, use of a cell or tissue-specific promoter can provide a significant level of patient safety in the event of unwanted vector dissemination beyond the targeted tissue. Over the last decade, studies from our laboratory have established several potential clinical applications of gene transfer to salivary glands (see Baum et al, Int. Rev Cytol., 2002). As a key step in optimizing vector delivery to salivary glands, we have begun to examine the effect of promoter selection on transgene expression. Initially, we screened the efficacy of several viral and mammalian cellular promoters in driving luciferase activity in an established, well-characterized rat submandibular gland cell line, A5. Four viral promoters were tested: cytomegalovirus (CMV, 0.59 kb), Rous Sarcoma Virus (RSV, 0.41 kb), SV40 (0.31 kb) and the long terminal repeat from Moloney murine leukemia virus (LTR, 0.61 kb). Six mammalian cellular promoters were tested including the promoters for human elongation factor 1a (EF1a, 1.2 kb), human cytokeratin 18 (K18, 2.5 kb), human cytokeratin 19 (K19, 2.95 kb), rat aquaporin-5 (rAQP5, 4.5 kb), human amylase (AMY, 1 kb) and human kallikein (KALL, 0.35 kb). The AMY promoter, although weak, is relatively salivary gland specific, while the kallikrein promoter is stronger, but active in several other epithelial tissues (Zheng et al., Hum. Gene Ther., 2001). All promoters were placed upstream of a luciferase cassette in pAC. This cassette included approximately 1.8 kb luciferase cDNA (obtained from pGL2-Basic, Promega) as a reporter gene, along with the SV40 polyadenylation signal. Luciferase activity was measured in a chemiluminescent assay. Results (n=3 experiments, each in triplicate) showed that the order of promoter activity in A5 cells was, from highest to lowest, CMV>EF1a>SV40=RSV>K18>K19>KALL>LTR=AMY. Future experiments will include examination of promoter activity in rat submandibular glands in vivo using transgene transfection methods (O’Connell et al., Amer. J. Physiol., 1995). We anticipate that these and related studies will permit development of optimized expression cassettes for vector-mediated gene transfer to salivary glands.
S441