Over-expression of the apple <Emphasis Type="Italic">spermidine synthase</Emphasis> gene in pear confers multiple abiotic stress tolerance by altering polyamine titers

An apple spermidine synthase (SPDS) gene (MdSPDS1) was verified to encode a functional protein by the complementation of the spe3 yeast mutant, which lacks the SPDS gene. To justify our hypothesis that apple SPDS is involved in abiotic stress responses and to obtain transgenic fruit trees tolerant to abiotic stresses as well, MdSPDS1-over-expressing transgenic European pear (Pyrus communis L. ‘Ballad’) plants were created by Agrobacterium-mediated transformation. A total of 21 transgenic lines showing various spermidine (Spd) titers and MdSPDS1 expression levels were obtained. Selected lines were exposed to salt (150 mM NaCl), osmosis (300 mM mannitol), and heavy metal (500 μM CuSO₄) stresses for evaluating their stress tolerances. Transgenic line no. 32, which was revealed to have the highest Spd accumulation and expression level of MdSPDS1, showed the strongest tolerance to these stresses. When growth increments, electrolyte leakage (EL), and values of thiobarbituric acid reactive substances (TBARS) were monitored, line no. 32 showed the lowest growth inhibition and the least increase in EL or TBARS under stress conditions. Spd titers in wild-type and transgenic lines showed diverse changes upon stresses, and these changes were not consistent with the changes in MdSPDS1 expressions. Moreover, there were no differences in the sodium concentration in the shoots between the wild type and line no. 32, whereas the copper concentration was higher in the wild type than in line no. 32. Although the mechanism(s) underlying the involvement of polyamines in stress responses is not known, these results suggest that the over-expression of the SPDS gene substantially increased the tolerance to multiple stresses by altering the polyamine titers in pear. Thus, MdSPDS1-over-expressing transgenic pear plants could be used to improve desert land and/or to repair polluted environments. Xiao-Peng Wen and Xiao-Ming Pang contributed equally to this work.