Pharmacological expression in rat hepatocytes of a gene transferred by an adenovirus vector enabled by a chimeric promoter containing multiple cyclic adenosine monophosphate response elements

Gene transfer to the pleural mesothelium as a strategy to deliver proteins to the lung parenchyma

The pleura covers the lung parenchyma, chest wall, and diaphragm with a single layer of flat cells that are easy to genetically modify with adenovirus (Ad) vectors. Although intrapleural gene therapy has been used to treat intrapleural disorders, we hypothesized that it may also be used to deliver extracellular gene products to the lung parenchyma. In this context, this study is based on the concept that administration of adenovirus gene transfer vectors into the pleural cavity will mediate expression of gene products in mesothelial cells, and that the extracellular products produced by these genetically modified cells will reach the lung parenchyma. To assess this concept, Ad(beta)gal (expressing beta-galactosidase [beta-Gal]) or AdLuc (expressing luciferase) was administered into the right pleural cavity of BALB/c mice, as compared with intravenous injection via the jugular vein or the intratracheal route. Histologic assessment of lungs and pleural surface after intrapleural administration of Ad(beta)gal demonstrated beta-Gal expression limited to the pleural mesothelium and cells adjacent to the pleural surface. Right intrapleural administration of AdLuc showed higher level of luciferase in both the right and left lung (right vs. left, p > 0.8), compared with the intratracheal (p < 0.05) or intravenous routes (p < 0.02), that is, unilateral intrapleural administration is sufficient to transfer genes bilaterally to the pleura. To assess the ability of intrapleural gene transfer to modify lung parenchymal processes, CT26.CL25 tumor cells (3 x 10(5)) were injected via the jugular vein to generate tumor metastases in the lung parenchyma followed 24 hr later by administration to the right pleura of 5 x 10(8) PFU of Adsflt (an Ad "antiangiogenesis" vector expressing a soluble, secreted, extracellular portion of the Flt-1 receptor for vascular endothelial growth factor). Compared with phosphate-buffered saline, or the control vector AdNull (no transgene), mice receiving Adsflt by the intrapleural route had a marked suppression of tumor growth in the parenchyma of both lungs as assessed 12 days after tumor administration (p < 0.005). Treatment of preestablished lung metastases with Adsflt administered by the intrapleural route significantly improved long-term survival as compared with control animals (p < 0.0001). Thus, although intrapleural administration of an Ad vector encoding an intracellular protein mediates gene expression only in mesothelial cells and the local tissues, intrapleural delivery of a vector expressing a secreted protein can be used to modify processes throughout the lung parenchyma. In the context that intravascular gene transfer is an ineffective strategy to deliver gene products to the lung parenchyma, and that intratracheal administration is associated with alveolar inflammation secondary to host defenses against Ad vectors, these findings demonstrate that intrapleural administration represents a strategy that can be used to effectively deliver extracellular gene products to the lung parenchyma via a site that is readily accessible, and where inflammation against the vector will not have significant pathophysiologic consequences.

Adenovirus vector-mediated transfer of the vascular endothelial growth factor cDNA to healing abdominal fascia enhances vascularity and bursting strength in mice with normal and impaired wound healing

BACKGROUND

We hypothesized that adenovirus-mediated transfer of the vascular endothelial growth factor (VEGF121) complementary DNA (cDNA) to murine laparotomy fascial wounds would enhance vascularity and bursting strength.

METHODS

Microfibrillar collagen sponges saturated with adenovirus (Ad) vectors encoding for the human VEGF121 cDNA (Ad(CU)VEGF121.1), a control marker gene (Ad beta gal, AdLuc) or no transgene (AdNull) were sutured to fascial edges during laparotomy closure in normal mice and mice treated with dexamethasone. Endpoints addressed included transgene expression in the fascia and surrounding tissue, the number of blood vessels in the healing wound determined using immunostaining, and wound bursting strength using a calibrated tensinometer.

RESULTS

Transgene expression was detected readily in the fascial edges, but only marginally detectable in neighboring tissues. In normal mice and mice treated with dexamethasone, no differences were observed at 7 days. Strikingly, however, 21 days after wound closure/therapy, significantly more blood vessels were present in the wounds that received the VEGF121 vector compared with controls (normal: AdNull: 4.2 +/- 1.8; Ad(CU)VEGF121.1: 11.2 +/- 1.2; P <.05; dexamethasone: AdNull: 1.4 +/- 0.8; Ad(CU)VEGF121.1: 5.4 +/- 1.2; P <.05), and bursting strength was significantly higher in VEGF121-treated wounds (normal: AdNull: 665 +/- 68 mN/mm; Ad(CU)VEGF121.1: 956 +/- 82 mN/mm; P <.0005; dexamethasone: AdNull: 234 +/- 111 mN/mm; Ad(CU)VEGF121.1: 592 +/- 121 mN/mm; P <.03).

CONCLUSIONS

Adenovirus-mediated gene transfer to healing fascial wounds is achieved readily using a microfibrillar collagen sponge, with transfer of the human VEGF121 cDNA significantly enhancing wound vascularity and bursting strength in normal mice, as well as in mice treated with dexamethasone.

Adenovirus gene transfer vectors inhibit growth of lymphatic tumor metastases independent of a therapeutic transgene

Adenovirus (Ad) gene transfer vectors traffic to regional lymph nodes (RLNs) after footpad injections in mice, resulting in localized production of interferon gamma (IFN-gamma). With this background, we evaluated the hypothesis that Ad vector administration may inhibit RLN tumor metastasis independent of the transgene in the expression cassette. Tumors of MM48, a cell line with a propensity toward lymphogenous metastasis, were established in the footpads of syngeneic C3H mice, and E1(-)E3(-) Ad vectors encoding no transgene (AdNull) or encoding an irrelevant transgene (AdCD; Escherichia coli cytosine deaminase with no 5-fluorocytosine administration) were administered (10(10) particles) in a peritumoral location. Both vectors suppressed the growth of tumor in the regional (popliteal) lymph node. This effect was localized to the regional, but not distant, lymph nodes (p < 0.05). Heat inactivation of the vector or decreasing the dose of the vector to 10(9) particles did not suppress RLN growth of the tumor when compared with 10(10) particles of active AdNull (p < 0.05 and p < 0.01, respectively). The ability of an E1(-)E4(-) vector expressing beta-galactosidase (AdRSVbetagal.11) to suppress RLN tumor growth showed that the E4 region of the Ad vector was not responsible for the effect. Blocking either IFN-gamma or natural killer (NK) cells with systemic antibody treatment in immunocompetent mice allowed rapid growth of RLN metastases despite Ad vector administration, and Ad vector injection into the footpads of tumor-free mice induced the accumulation of NK cells in the RLN. These data demonstrate that, in a metastatic murine tumor model, a low dose (10(10) particles) of replication-deficient Ad vectors inhibits RLN metastases independent of a therapeutic transgene, an effect that is mediated, at least in part, by IFN-gamma and NK cells.

Adenovirus vector-mediated gene transfer to regional lymph nodes

Regional lymph nodes (RLNs) possess important immune functions and represent a major pathway of metastasis for solid tumors. Given these facts, the ability to transfer exogenous genes to the RLNs with the goal of manipulating the local immunological milieu would be desirable. On the basis of the hypothesis that a significant proportion of adenovirus (Ad) gene transfer vectors traffic through the lymphatics, E1-E3- Ad vectors were injected into the hind footpad of C3H/He mice and the RLNs assessed for vector trafficking and transgene expression. A low dose (10(9) particles) of an Ad vector encoding the firefly luciferase gene (Ad-CMV.Luc) resulted in luciferase expression only in the injection site and RLNs, with no detectable systemic (liver, spleen, lung) expression. At a higher dose (10(11) particles), some expression could be detected systemically in addition to the RLNs, but at levels in liver 14-fold less than in the RLNs. Transgene expression in the RLNs was transient, peaking at 1 day, decreasing markedly by 7 days. At high doses (10(11) particles), interruption of draining lymphatics decreased the amount of systemic dissemination 22-fold, suggesting that a large proportion of the vector trafficks through the lymphatics before reaching the systemic circulation. Administration of a vector encoding the jellyfish green fluorescent protein gene (AdCMV.GFP, 10(11) particles) showed that transgene expression in the RLNs was primarily in the cortical area. After footpad injection of a fluorescent-labeled Ad vector (Cy3-AdCMV.Null), fluorescent virions were visualized in the draining lymph. Regional lymph collected from animals injected in the footpad with AdCMV.Luc (10(11) particles) contained functional vector. Augmentation of local immune function in the RLNs was achieved by footpad administration of an Ad vector encoding murine IL-12, resulting in high mIL-12 and IFN-gamma levels in the regional, but not distant, nodes. These data demonstrate that expression of exogenous genes in RLNs is easily accomplished with Ad vectors, Ad vector dissemination occurs primarily via the lymphatics after footpad administration in mice, and basic immune functions in the RLNs can be manipulated by Ad-mediated gene transfer in vivo.

Adenovirus vector-mediated perforin expression driven by a glucocorticoid-inducible promoter inhibits tumor growth in vivo

To evaluate the concept that in vivo transfer of perforin complementary DNA (cDNA) will suppress tumor growth, we constructed an adenovirus vector (AdGRE.PFP) carrying perforin cDNA driven by the glucocorticoid response element (GRE) promoter. We infected A549 lung carcinoma cells with this vector in vitro and in vivo, and evaluated cell growth over time. In the presence of dexamethasone, in vitro infection of A549 cells with the AdGRE.PFP vector yielded perforin messenger RNA (mRNA) transcripts and effectively suppressed A549 cell growth. In accord with these in vitro observations, administration of dexamethasone following direct injection of AdGRE. PFP into established subcutaneous A549 tumors in nude mice resulted in a marked reduction in tumor growth as compared with AdGRE.PFP infection without dexamethasone or with dexamethasone alone. These observations suggest that regulable, adenovirus-mediated gene expression of perforin cDNA may have potential as a strategy for local control of tumor cell growth.

Intermittent, Repetitive Corticosteroid-Induced Upregulation of Platelet Levels After Adenovirus-Mediated Transfer to the Liver of a Chimeric Glucocorticoid-Responsive Promoter Controlling the Thrombopoietin cDNA

For many in vivo gene therapy clinical applications, it is desirable to control the expression of the transferred transgene using pharmacologic agents. To evaluate the feasibility of accomplishing this using corticosteroids, pharmacologic agents widely used in clinical medicine, we constructed replication deficient adenoviral (Ad) vectors containing an expression cassette with a chimeric promoter comprised of five glucocorticoid response elements (GRE) and the chloramphenicol acetyltransferase reporter gene (AdGRE.CAT) or the murine thrombopoietin cDNA (AdGRE.mTPO). In vitro studies showed the vectors functioned as expected, with marked glucocorticoid-induced upregulation of the CAT or mTPO transgenes. To evaluate the inducibility of the GRE promoter in vivo, the AdGRE.CAT vector was administered intravenously to C57B1/6 mice, and CAT activity was quantified in liver before and after intraperitoneal administration of dexamethasone. The GRE promoter activity was dependent on the dexamethasone dose, with a 100-fold increase in CAT expression with 50 μg dexamethasone, similar to the levels observed in vivo with the Rous sarcoma virus long terminal repeat constitutive promoter. After dexamethasone administration, maximum CAT activity was observed at day 2, with a slow decline to baseline levels by 2 weeks. Based on these observations, we hypothesized that a single administration of an Ad vector-mediated transfer of the chimeric GRE inducible promoter driving the mTPO cDNA would enable repetitive administration of corticosteroids to repetitively upregulate platelet levels for 1 to 2 weeks. The data show that this occurs, with dexamethasone administration every 3 weeks associated with 1-week elevations (at each 3-week interval) of serum mTPO levels, megakaryocyte numbers in bone marrow, and platelet levels fourfold to sixfold over baseline. Thus, with the appropriate promoter, it is possible to use a commonly used pharmacologic agent to upregulate the expression of a newly transferred gene on demand.

© 1998 by The American Society of Hematology.

Intermittent, Repetitive Corticosteroid-Induced Upregulation of Platelet Levels After Adenovirus-Mediated Transfer to the Liver of a Chimeric Glucocorticoid-Responsive Promoter Controlling the Thrombopoietin cDNA

For many in vivo gene therapy clinical applications, it is desirable to control the expression of the transferred transgene using pharmacologic agents. To evaluate the feasibility of accomplishing this using corticosteroids, pharmacologic agents widely used in clinical medicine, we constructed replication deficient adenoviral (Ad) vectors containing an expression cassette with a chimeric promoter comprised of five glucocorticoid response elements (GRE) and the chloramphenicol acetyltransferase reporter gene (AdGRE.CAT) or the murine thrombopoietin cDNA (AdGRE.mTPO). In vitro studies showed the vectors functioned as expected, with marked glucocorticoid-induced upregulation of the CAT or mTPO transgenes. To evaluate the inducibility of the GRE promoter in vivo, the AdGRE.CAT vector was administered intravenously to C57B1/6 mice, and CAT activity was quantified in liver before and after intraperitoneal administration of dexamethasone. The GRE promoter activity was dependent on the dexamethasone dose, with a 100-fold increase in CAT expression with 50 μg dexamethasone, similar to the levels observed in vivo with the Rous sarcoma virus long terminal repeat constitutive promoter. After dexamethasone administration, maximum CAT activity was observed at day 2, with a slow decline to baseline levels by 2 weeks. Based on these observations, we hypothesized that a single administration of an Ad vector-mediated transfer of the chimeric GRE inducible promoter driving the mTPO cDNA would enable repetitive administration of corticosteroids to repetitively upregulate platelet levels for 1 to 2 weeks. The data show that this occurs, with dexamethasone administration every 3 weeks associated with 1-week elevations (at each 3-week interval) of serum mTPO levels, megakaryocyte numbers in bone marrow, and platelet levels fourfold to sixfold over baseline. Thus, with the appropriate promoter, it is possible to use a commonly used pharmacologic agent to upregulate the expression of a newly transferred gene on demand.

© 1998 by The American Society of Hematology.