As a step toward preventing and curing Staphylococcus aureus mastitis, an adenoviral-mediated gene transfer technique was used to enable mammary cells to synthesize and secrete lysostaphin, an anti-staphylococcal protein. A lysostaphin gene, modified for eukaryotic expression of the bioactive variant, Gln125,232-lysostaphin, was inserted into a replication deficient adenovirus by homologous recombination in 293 cells. The resulting adenoviral vector containing the modified lysostaphin gene (Ad-lys) was used to infect bovine mammary epithelial cells in vitro and caprine mammary cells in vivo. A similar adenoviral vector containing the Escherichia coli gene encoding beta-galactosidase (Ad-lacZ) was also evaluated. Transduction of cultured bovine cells by Ad-lacZ was confirmed by the presence of beta-galactosidase in fixed cells 48 h postinfection. Bovine cells transduced by Ad-lys secreted immunoreactive Gln125,232-lysostaphin (0.8 microg/ml) into media that had approximately 20% bioactivity compared with native lysostaphin. To evaluate transduction in vivo, udder halves of four nonlactating goats were exposed to 10(10) plaque-forming units (pfu) ofAd-lacZ by two intramammary infusions given 48 h apart. The animals were euthanized 24 h later, and extensive expression of beta-galactosidase was detected in cells lining the teat canals, with more moderate expression observed in adjoining mammary parenchyma. Udder halves of two other nonlactating goats were infused with 10(10) pfu of Ad-lys while contralateral udder halves received Ad-lacZ. The animals were euthanized 48 h postinfusion. In both animals, extensive expression of beta-galactosidase was detected in Ad-lacZ exposed teats. Immunoreative Gln125,232-lysostaphin was detectable in secretions from Ad-lys exposed glands 24 h postinfusion, increasing to approximately 1 microg/ml at 48 h postinfusion. As with cultured bovine mammary epithelial cells, the bioactivity of goat-derived Gln125,232-lysostaphin was approximately 20% of native lysostaphin. These results demonstrate that an adenoviral vector can be used to introduce a gene into the ruminant mammary gland, enabling the secretion of a bioactive form of lysostaphin.

Gene therapy provides a promising approach to curing diabetes. However, an effective route for islet-specific targeting has yet to be established. Toward this end, the pancreatic blood circulation system in Balb/c mice was determined by the injection of rhodamine-containing beads. The efficiency of islet targeting was then measured by the injection of adenoviral vectors carrying a green fluorescence gene via the celiac trunk (C.T.). The results showed that >95% of islets and about 60% of β cells within the pancreatic body and tail could be labeled 3 days after surgery. α-Cell labeling was not as efficient, whereas labeling of nonendocrine tissues was barely detectable. For proof of principle, adenoviral vectors carrying a Sirtuin transgene were injected similarly to test the islet protection effect in the streptozotocin (STZ)-induced type 1 diabetic model. The results demonstrated that overexpression of Sirtuin in STZ-treated mice reduced the level of β-cell death and extent of glucose intolerance. This study reports on efficient islet-specific targeting by using adenoviral injection. This procedure could be invaluable to the treatment of diabetes and the study of islet biology.

Gene therapy has great potential to enable synthesis of protein molecules in targeted cells of an animal. One application may be the production of antibacterial enzymes by the mammary gland as a means of preventing or treating mastitis. We have previously demonstrated that goat mammary cells are capable of producing lysostaphin, an antistaphylococcal enzyme, after being transduced in vivo with a recombinant adenoviral vector containing a modified lysostaphin gene (Ad-lys). The current study examined duration of expression, and antibody response to lysostaphin and the adenoviral vector. Following intramammary infusion into nonlactating goats (n = 4), recovery of transducible adenoviral vector in mammary secretions persisted for 11 d. Transducible vector was not detected in serum, saliva, urine, or feces. Peak lysostaphin concentrations (< 20 microg/mL) in mammary secretions of infused udders were detected approximately 1 wk postinfusion, and generally returned to undetectable levels after an additional 1 to 2 wk. The poor persistency of expression was likely due to the very potent immune response to both the adenovirus and the expressed lysostaphin. Serum IgG antibodies recognizing the adenoviral vector developed within 7 d of the infusion, and titers rose dramatically to greater than 1:1 x 10(5). Similar titers of serum IgG antibodies to lysostaphin developed in 3 goats, with more moderate titers in the fourth goat. The antibody response to lysostaphin was delayed by approximately 4 d in comparison to the response to the adenovirus. Serum IgG antibody profiles were reflected in mammary secretions. No IgA antibodies to adenovirus or lysostaphin were detected in sera or mammary secretion. We demonstrate that while the lysostaphin gene can be introduced to the mammary gland using an adenoviral-mediated gene transfer technique, the strong immune response that it provokes makes the approach unsuitable for combating mastitis.

The present study has established an approach for efficient pancreatic capillary targeting by using microsphere-adenoviral complexes. This procedure could be invaluable for the treatment of capillary-related diseases.

Ectopically expressed Cre recombinase in extrapancreatic tissues in RIP-Cre mice has been well documented. The objective of this study was to find a simple solution that allows for improved beta-cell specific targeting. To this end, the RIP-Cre and reporter CMV-loxP-DsRed-loxP-EGFP expression cassettes were configurated into a one-plasmid and two-plasmid systems, which labeled approximately 80% insulin-positive INS-1 cells after 48 h transfection. However, off-target labeling was robustly found in more than 15% insulin-negative Ad293 cells. When an IRES element was inserted in front of Cre to reduce the translation efficiency, the ratio of recombination between INS-1 and Ad293 cells increased 3-4-fold. Further, a series of Cre mutants were generated by site-directed mutagenesis. When one of the mutants, Cre(H289P) in both configurations, was used in the experiment, the percentage of recombination dropped to background levels in a number of insulin-negative cell lines, but decreased only slightly in INS-1 cells. Consistently, DNA substrate digestion assay showed that the enzymatic activity of Cre(H289P) was reduced by 30-fold as compared to that of wild-type. In this study, we reported the generation of constructs containing RIP and Cre mutants, which enabled enhanced beta-cell specific labeling in vitro. These tools could be invaluable for beta-cell targeting and to the study of islet development.