A relatively high percentage of epiphytic bacteria on pear leaf and fruit surfaces had the ability to produce indole-3-acetic acid (IAA) in culture media supplemented with tryptophan. While over 50% of the strains produced at least small amounts of IAA in culture, about 25% of the strains exhibited high IAA production as evidenced by both colorimetric and high-performance liquid chromatography analysis of culture supernatants. A majority of the strains that produced high amounts of IAA were identified as Erwinia herbicola (Pantoea agglomerans), while some strains of Pseudomonas syringae, Pseudomonas viridiflava, Pseudomonas fluorescens, Pseudomonas putida, and Rahnella aquaticus that produced high amounts of IAA also were found on pear. Fruit russeting was significantly increased in 39 out of 46 trials over an 8-year period in which IAA-producing bacteria were applied to trees compared with control trees. A linear relationship was observed between fruit russet severity and the logarithm of the population size of different IAA-producing bacteria on trees in the 30 days after inoculation, when normalized for the amount of IAA produced by each strain in culture. On average, the severity of fruit russet was only about 77% that on control trees when trees were treated at the time of bloom with Pseudomonas fluorescens strain A506, which does not produce IAA. Both total bacterial populations on pear in the 30-day period following full bloom and fruit russet severity varied greatly from year to year and in different commercial orchards over a 10-year period. There was a strong linear correlation between the logarithm of total bacterial population sizes and fruit russet severity.

Abstract-The apolipoprotein E (apoE)-deficient mouse is a relevant animal model of human atherosclerosis. Although the prevention of atherosclerosis development has been documented after somatic gene transfer into animal models, regression of lesions remains to be demonstrated. Thus, we used this genetically defined mouse model nn the nude background to show atherosclerosis regression. ApoE-deficient nude mice were infected with 5ϫ10 8 or 10 9 plaque-forming units of a first-generation adenovirus encoding human apoE cDNA. The secretion of human apoE resulted in a rapid decrease of total cholesterol, which normalized the hypercholesterolemic phenotype within 14 days (from 600Ϯ100 to Ͻ100 g/mL). Transgene expression was observed during a period of Ͼ4 months, with a normalization of cholesterol and triglyceride levels during 5 months. At that time, we successfully reinjected the recombinant adenovirus and observed the appearance of the human protein as well as the correction of lipoprotein phenotype. In mice killed 6 months-after the first infection, we observed a dose-dependent regression of fatty streak lesions in the aorta. We showed sustained expression of a transgene with a first-generation adenoviral vector and a correction of dyslipoproteinemia phenotype leading to lesion regression. These data demonstrate that somatic gene transfer can induce plaque regression.