To determine whether overexpression of Fe-superoxide (SOD) dismutase would increase superoxide-scavenging capacity and thereby improve the winter survival of transgenic alfalfa (Medicago sativa L.) plants, two genotypes were transformed with the vector pEXSOD10, which contains a cDNA for Arabidopsis Fe-SOD with a chloroplast transit peptide and cauliflower mosaic virus 35S promoter. A novel Fe-SOD was detected by native PAGE in both greenhouse-and field-grown transgenic plants, but activity varied among independent transgenic plants. The increased Fe-SOD activity was associated with increased winter survival over 2 years in field trials, but not with oxidative stress tolerance as measured by resistance of leaves to methyl viologen, a superoxide generator. Total shoot dry matter production over 2 harvest years was not associated with Fe-SOD activity. There was no detectable difference in the pattern of primary freezing injury, as shown by vital staining, nor was there additional accumulation of carbohydrates in field-acclimated roots of the transgenic alfalfa plants. We did not detect any difference in growth of one transgenic plant with high Fe-SOD activity compared with a non-transgenic control. Therefore, the improvement in winter survival did not appear to be a consequence of improved oxidative stress tolerance associated with photosynthesis, nor was it a consequence of a change in primary freezing injury. We suggest that Fe-SOD overexpression reduced secondary injury symptoms and thereby enhanced recovery from stresses experienced during winter.

Winterhardiness is a composite of tolerances to freezing, desiccation, ice-encasement, flooding and diseases. From one point of view, winterhardiness may not be easily manipulated by genetic engineering technology because many different genes are involved in the tolerance of these diverse stresses. However, these various stresses have similarities. They promote formation of activated forms of oxygen, promote membrane lipid and protein degradation, cause similar biophysical changes in membrane structure, and culminate with increased leakage of cytoplasmic solutes and loss of cellular membrane functions. These similarities led to the hypothesis that winter injury might be reduced in crop plants if their tolerance of oxidative stress was increased.Towards that objective we created transgenic alfalfa (Medicago sativa L.) plants that overexpress either Mn-SOD or Fe-SOD cDNA (provided by Dirk Inzt, Universiteit Gent). Petiole explants were transformed using Agrobacterium tumefaciens and plants were regenerated by somatic embryogenesis. The primary transgenic plants were screened using PCR (polymerase chain reaction), Southern hybridization and native PAGE for SOD activity. Greenhouse and laboratory studies showed a minimal difference in stress tolerance between the primary transgenic and non-transgenic plants. In the first field trial, four primary transgenic plants expressing two forms of the Mn-SOD cDNA had greater survival after two winters than the non-transgenic RA3. Similar results were obtained in a second field trial, comparing 18 independent transformants with Mn-SOD targeted to the mitochondria, 11 independent transformants with Mn-SOD targeted to the chloroplast and 39 independent transformants with Fe-SOD targeted to the chloroplast, expressed in three different non-transgenic plants. The transgenic plants averaged over 25% higher survival than the non-transgenic controls after one winter. There was no effect of subcellular targeting or SOD type on field survival, but there was variation among independent transformants containing the same SOD construct. Activated oxygen therefore appears to be one of the possible causes of winter injury, and it should be possible to reduce winter injury in transgenic plants by constitutive overexpression of SOD.