Investigation of a novel dendritic derivative of 5-aminolaevulinic acid for photodynamic therapy

Photodynamic therapy is a treatment for malignant and certain non-malignant lesions that involves administration of a photosensitising drug. The use of 5-aminolaevulinic acid-induced porphyrins has become one of the most active fields of photodynamic therapy research. Since the efficacy of the treatment is somewhat limited by the hydrophilic nature of 5-aminolaevulinic acid, chemical modifications such as esterification with aliphatic alcohols have been made to induce higher porphyrin production. In an attempt to improve delivery of 5-aminolaevulinic acid to tissue, we have investigated the use of dendritic derivatives capable of bearing several drug molecules. The aim of this work was to evaluate in vivo and in vitro the efficacy of the first generation dendron, aminomethane tris-methyl 5-aminolaevulinic acid (containing three 5-aminolaevulinic acid residues) in terms of porphyrin synthesis. In LM3 cells, the dendron induced similar porphyrin levels compared to equimolar concentrations of 5-aminolaevulinic acid. Although the dendron is taken up with comparable efficiency to 5-aminolaevulinic acid, we found that there is only partial intracellular liberation of 5-aminolaevulinic acid residues. Both systemic and topical administration of the dendron to tumour-bearing mice induced higher porphyrin levels than the widely investigated hexyl ester derivative in most tissues studied, although it was not possible to surpass the levels induced by 5-aminolaevulinic acid. In conclusion, aminomethane tris-methyl 5-aminolaevulinic acid is capable of being taken up by cells efficiently, and liberating the active residues, although in vivo it was not possible to improve upon the efficacy of 5-aminolevulinic acid. Studies of accessibility and regulation of the esterases are needed to improve the design of these dendritic derivatives.     

Genetic Structure of Avian Influenza Viruses from Ducks of the Atlantic Flyway of North America

Wild birds, including waterfowl such as ducks, are reservoir hosts of influenza A viruses. Despite the increased number of avian influenza virus (AIV) genome sequences available, our understanding of AIV genetic structure and transmission through space and time in waterfowl in North America is still limited. In particular, AIVs in ducks of the Atlantic flyway of North America have not been thoroughly investigated. To begin to address this gap, we analyzed 109 AIV genome sequences from ducks in the Atlantic flyway to determine their genetic structure and to document the extent of gene flow in the context of sequences from other locations and other avian and mammalian host groups. The analyses included 25 AIVs from ducks from Newfoundland, Canada, from 2008–2011 and 84 available reference duck AIVs from the Atlantic flyway from 2006–2011. A vast diversity of viral genes and genomes was identified in the 109 viruses. The genetic structure differed amongst the 8 viral segments with predominant single lineages found for the PB2, PB1 and M segments, increased diversity found for the PA, NP and NS segments (2, 3 and 3 lineages, respectively), and the highest diversity found for the HA and NA segments (12 and 9 lineages, respectively). Identification of inter-hemispheric transmissions was rare with only 2% of the genes of Eurasian origin. Virus transmission between ducks and other bird groups was investigated, with 57.3% of the genes having highly similar (≥99% nucleotide identity) genes detected in birds other than ducks. Transmission between North American flyways has been frequent and 75.8% of the genes were highly similar to genes found in other North American flyways. However, the duck AIV genes did display spatial distribution bias, which was demonstrated by the different population sizes of specific viral genes in one or two neighbouring flyways compared to more distant flyways.     

Sample size planning for developing classifiers using high-dimensional DNA microarray data

Many gene expression studies attempt to develop a predictor of pre-defined diagnostic or prognostic classes. If the classes are similar biologically, then the number of genes that are differentially expressed between the classes is likely to be small compared to the total number of genes measured. This motivates a two-step process for predictor development, a subset of differentially expressed genes is selected for use in the predictor and then the predictor constructed from these. Both these steps will introduce variability into the resulting classifier, so both must be incorporated in sample size estimation. We introduce a methodology for sample size determination for prediction in the context of high-dimensional data that captures variability in both steps of predictor development. The methodology is based on a parametric probability model, but permits sample size computations to be carried out in a practical manner without extensive requirements for preliminary data. We find that many prediction problems do not require a large training set of arrays for classifier development.