5.3.8
PATHOGEN-DERIVED TRANSGENES CONFER VIRUS RESISTANCE IN SUGARCANE

PA JOYCE1, RB McQUALTER1, RM HARDING2, JA HANDLEY2, JL DALE2 and GR SMITH1

1David North Plant Research Centre, Bureau of Sugar Experiment Stations, Indooroopilly, Brisbane, Q4068, Australia; 2Centre for Molecular Biotechnology, Queensland University of Technology, George St, Brisbane Q4001, Australia

Background and objectives
Sugarcane is a genetically complex, aneuploid polyploid Saccharum interspecific hybrid within the Gramineae family. Contemporary commercial cultivars are believed to have chromosome numbers varying between 100 and 130. This genetic complexity makes breeding for specific pest and disease resistances, whilst retaining elite agronomic characters, very difficult. Contributing to this difficulty is that selected crosses can generate progeny whose phenotype for a particular character is outside the phenotype of the parents, making predictions of the performance of individual seedlings from a cross difficult, and back-crossing practically impossible. Genetic engineering offers a practical solution to the problem of introducing resistance genes into existing agronomically elite varieties, without the genetic reassortment that results from crossing. In the 1980s an epidemic of sugarcane mosaic, caused by sugarcane mosaic potyvirus, resulted in substantial yield losses of up to 40% in the Isis region of southern Queensland. The lack of any agronomically suitable, SCMV-resistant plants in the breeding/selection programmes and the demonstration that pathogen-derived genes could be used to confer virus resistance provided a requirement and a strategy to develop transgenic SCMV-resistant sugarcane.

Materials and methods
Embryogenic callus of cultivars including Q155 was produced on media supplemented with 2,4-D. Plasmid constructs were based on either the 'Emu' or maize ubiquitin promoters, and Nos termination sequence. The CP gene was translatable [1], while translatable and untranslatable replicase constructs used sequence derived from analysis of the NIb region of Australian isolates of SCMV [2]. Transformed callus was selected on a medium supplemented with geneticin, and then placed in the light to initiate shoot regeneration. When the shoots were approximately 10 cm tall and had a well established root system they were transferred to pots in a containment glasshouse. When the plants were 6-7 weeks old they were mechanically challenged with SCMV. Standards of non-transgenic Q155 were also inoculated. Samples were removed for analysis of viral titre by ELISA. 2-week-old maize (cv. Iochief) plants were back-inoculated with extracts from symptomatic and asymptomatic transgenic plants. Development of symptoms on the maize plants was noted, and the viral titre analysed by ELISA.

Results and conclusions
Transgenic sugarcane plants transformed with the coat protein of sugarcane mosaic potyvirus are resistant to subsequent challenge inoculation. Lines carrying replicase constructs are yet to be evaluated. Three resistance phenotypes in response to mechanical infection have been identified: immunity, where the lines show no symptoms of SCMV infection, contain no virus (ELISA data), and inoculum prepared from them is not infectious to the maize indicator plants; recovery, typified by symptoms on the leaves that had emerged at the time of inoculation and that decrease in intensity until the newly emerging leaves have no symptoms; and atypical, characterized by small flecks on the leaf blade, at the leaf margin or on the midrib, or small patchy chlorotic regions across the leaf. ELISA and back-inoculation experiments indicated that no virus was present in these plants. A number of the transgenic lines remained susceptible to SCMV, with high viral titres and extracts infectious to maize in the back-inoculation experiments. Many of the resistant lines have a high transgene copy number, no measurable expression of coat protein from the translatable constructs, and transgene-specific transcript levels are very low. Transgenic resistance in this genetically complex monocotyledon appears to be RNA-mediated, and research to confirm this hypothesis continues.

References
1. Smith GR, Ford R, Frenkel MJ et al., 1992. Archives of Virology 125, 15-23.
2. Handley JA, Smith GR, Dale JL, Harding RM, 1996. Archives of Virology 141, 2289-2300.