Search

We tested the efficacy of an attenua- tion system developed to preclude the deleteri- ous effects of floral promoter::cytotoxin genes on vegetative growth of transgenic sterile plants. We tested the promoter (2.6 kb 5¢ region) of the poplar LEAFY gene PTLF driving barstar, combined on the same T-DNA with barstar dri- ven by either the CaMV 35S basal promoter +5 to -72 fragment (35SBP), 35SBP fused to the TMV omega element (35SBP omega), or the NOS promoter. The unattenuated pPTLF::barnase construct failed to give rise to any transgenic events, suggesting substantial non-reproductive expression from this promoter. The barstar- attenuated constructs enabled transformation, but the rate was reduced by nearly one-third. Four events (7% of attenuated events) had highly abnormal morphology, and were identified during the early phases of propagation; these events had significantly higher barnase:barstar expression ra- tios based on quantitative RT-PCR. A greenhouse study showed that phenotypically normal attenu- ated plants grew at the same rate as wild-type and barnase-lacking transgenic plants. A statistically significant positive linear association was found between relative growth rate (RGR) and bar- star:barnase ratio in the attenuated events, and graphical analysis suggested a threshold for barstar attenuation of barnase, above which additional levels of barstar did not provide further attenua- tion. Surprisingly, the appearance and growth rate of the nearly all of the attenuated events were substantially reduced after one or two growing seasons in the field, and the extent of growth reduction was associated with barstar:barnase expression ratio. These results demonstrate the importance of field testing during early phases of research to identify pleiotropic effects of transgenic sterility genes in trees.