Fructan biosynthesis in transgenic plants
Andrew J Cairns
DOI: https://doi.org/10.1093/jxb/erg056
Abstract:Data from plants transformed to accumulate fructan are assessed in the context of natural concentrations of reserve carbohydrates and natural fluxes of carbon in primary metabolism: Transgenic fructan accumulation is universally reported as an instantaneous endpoint concentration. In exceptional cases, concentrations of 60-160 mg g(-1) fresh mass were reported and compare favourably with naturally occurring maximal starch and fructan content in leaves and storage organs. Generally, values were less than 20 mg g(-1) for plants transformed with bacterial genes and <9 mg g(-1) for plant-plant transformants. Superficially, the results indicate a marked modification of carbon partitioning. However, transgenic fructan accumulation was generally constitutive and involved accumulation over time-scales of weeks or months. When calculated as a function of accumulation period, fluxes into the transgenic product were low, in the range 0.00002-0.03 nkat g(-1). By comparison with an estimated minimum daily carbohydrate flux in leaves for a natural fructan-accumulating plant in field conditions (37 nkat g(-1)), transgenic fructan accumulation was only 0.00005-0.08% of primary carbohydrate flux and does not indicate radical modification of carbon partitioning, but rather, a quantitatively minor leakage into transgenic fructan. Possible mechanisms for this low fructan accumulation in the transformants are considered and include: (i) rare codon usage in bacterial genes compared with eukaryotes, (ii) low transgene mRNA concentrations caused by low expression and/or high turnover, (iii) resultant low expression of enzyme protein, (iv) resultant low total enzyme activity, (v) inappropriate kinetic properties of the gene products with respect to substrate concentrations in the host, (vi) in situ product hydrolysis, and (vii) levan toxicity. Transformants expressing bacterial fructan synthesis exhibited a number of aberrant phenotypes such as stunting, leaf bleaching, necrosis, reduced tuber number and mass, tuber cortex discoloration, reduction in starch accumulation, and chloroplast agglutination. In severe cases of developmental aberration, potato tubers were replaced by florets. Possible mechanisms to explain these aberrations are discussed. In most instances, the attempted subcellular targeting of the transgene product was not demonstrated. Where localization was attempted, the transgene product generally mis-localized, for example, to the cell perimeter or to the endomembrane system, instead of the intended target, the vacuole. Fructosyltransferases exhibited different product specificities in planta than in vitro, expression in planta generally favouring the formation of larger fructan oligomers and polymers. This implies a direct influence of the intracellular environment on the capacity for polymerization of fructosyltransferases and may have implications for the mechanism of natural fructan polymerization in vivo.