Abstract
The aim of the project was to develop a strategy towards anthracnose resistance in lupin using molecular techniques. Colletotrichum species are considered to be major plant pathogens of cereals and legumes around the world, causing significant crop losses. Colletotrichum acutatum causes anthracnose disease on lupin.
Sweet white lupin (Lupinus albus) is a high protein grain crop that could alleviate protein shortage in South Africa, since it has the highest protein levels (34-45%) compared to Lupinus angustifolius. In an effort to combat the lupin anthracnose threat to the South African lupin industry, which has an annual turnover of approximately 60 million rands, a project was embarked upon to introduce defense genes into a white lupin and a narrow leaf lupin cultivar.
Bean polygalacturonase inhibiting protein (PvPGIP), either extracted from bean or from transgenic tomato expressing the bean pgip1 gene (Pvpgip1), inhibited the C. acutatum polygalacturonase (PG) activity (isolate SHK 788) only by 18-25%, compared to apple PGIP (MdPGIP) that inhibited the C. acutatum PG activity by 70%. These results led to the Mdpgip1 gene, rather than the Pvpgip1 gene, being chosen for genetic engineering of lupin towards anthracnose resistance. However, since plants express more than one PGIP, the protein in the extract prepared from the fruit of apple cv. Granny Smith, could be encoded by any one of at least two closely related copies of pgip genes found in apple. Screening of eight putative first generation Mdpgip1 transformed tobacco plants using PCR, showed that all eight plants contained the Mdpgip1 gene. Inhibition studies, using the C. acutatum PGs, were performed which identified Mdpgip1 transgenic tobacco plant #8 as being the highest expresser of the MdPGIP1, since the MdPGIP1 extract from this plant exhibited the highest level of C. acutatum PG inhibition. The PGIP extract from the non-transgenic tobacco plant, as well as heat denatured MdPGIP1 extracts from the Mdpgip1 transgenic tobacco plants, resulted in no inhibition of C. acutatum PG activity. Mdpgip1 transgenic tobacco plant #8 was chosen for the purification of MdPGIP1. The protein was purified to apparent homogeneity using anion and cation exchange chromatography. N-terminal sequencing deduced the first 15 amino acids, which aligned 100% to the sequence of a pgip gene (called Mdpgip) from Golden Delicious apples (Genbank: accession no. MDU 77041), confirming isolation of MdPGIP1. The protein had a molecular mass of approximately 46kDa as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and an isoelectric point of 8.0. Purified MdPGIP1 inhibited the PGs produced by C. acutatum and the PGs produced by two apple pathogens, B. obtusa and D. ambigua. Results indicated that much less MdPGIP1 is required for effective inhibition of the B. obtusa and D. ambigua PGs, compared to the C. acutatum PGs. However, at higher MdPGIP1 concentrations all three fungal PGs were inhibited equally well. A purified endo-PG from Aspergillus niger was not inhibited by MdPGIP1. This constitutes the first report on the inhibitory activity of MdPGIP1 towards the PGs from C. acutatum, and the two apple pathogens B. obtusa and D. ambigua.
As part of a multigene approach to the production of anthracnose resistant lupin, the use of a yeast exo--1,3-glucanase (EXG1) as an antifungal agent towards C. acutatum was investigated. The exo--1,3-glucanase (exg1) gene had been isolated from Saccharomyces cerevisiae. Yeast cultures transformed with the exg1 gene, as well as untransformed yeast cultures, were obtained from the Institute for Wine Biotechnology, South Africa. Fungal spore suspensions, from isolate SHK 788, were prepared and used in inhibition studies with spore concentrations ranging from 2.5.103 spores to 80.103 spores per flask. Inhibition of C. acutatum mycelial growth ranged from 41%, at a fungal spore concentration of 2.5.103 spores, to 20%, at a fungal spore concentration of 80.103 spores. Ammonium sulphate concentrated yeast extracts containing the glucanase enzyme did not result in increased inhibition of C. acutatum mycelial growth. As an added control, an inhibition study using Botrytis cinerea spores yielded similar results to those obtained for the C. acutatum inhibition studies. An inhibition of at least 50% for all spore concentrations was set as the criterium to decide that the exg1 gene is potent enough for genetic engineering of disease resistance. This extent of inhibition was not obtained and the use of the exg1 gene for protection of lupin against C. acutatum was therefore not considered a worthwhile commercial option.
The defense gene plant transformation vectors prepared for lupin transformation, pCAMBIA 3300-virG, pCAMBIA 3301-virG, pCAMBIA 3300-virG-applePGIP and pCAMBIA 0390:applePGIP were successfully transformed into the A. tumefaciens strains LBA 4404 and AGL1. Lupin transformation was performed by the transformation group at CSIR Bio/Chemtek using A. tumefaciens-mediated transformation of shoot apical meristems. This group showed that the inclusion of the supervirulence virG gene enhanced the levels of transient GUS expression in L. angustifolius by more than two fold. However, transformation efficiency was low, and regeneration of the lupin plant proved to be even more difficult. To overcome the difficulties with plant tissue culture-based transformation systems, an A. tumefaciens seed vacuum infiltration transformation method was utilised. Extracts obtained from Mdpgip1 transgenic tobacco plants produced at CSIR Bio/Chemtek (pCAMBIA 3300-virG-applePGIP as well as pCAMBIA 3300-virG/pCAMBIA 0390:applePGIP transformants) inhibited the C. acutatum PGs. The Mdpgip1 gene thus codes for an active protein in the transgenic tobacco plants, and the defense gene constructs prepared for lupin transformation are functional in planta.
The shpx6a peroxidase gene was isolated from Stylosanthes humulis, as the second defense gene to be used in the strategy towards anthracnose resistance in lupin, and substitute for the yeast exg1 gene. Sequencing data confirmed the successful isolation of the shpx6a peroxidase gene, which was subsequently cloned into pCAMBIA 0390:applePGIP upstream from the NOS terminator to produce pCAMBIA 0390:applePGIP:peroxidase.
Seeing that the constitutive CaMV 35S promoter was going to be used upstream from the selection gene (bar), the Mdpgip1 gene and the additional shpx6a peroxidase gene, there was a concern that one type of gene silencing could occur. Use of one promoter can block expression of another gene being expressed from the same promoter on account of methylation of the promoter DNA. A 4.2kb fragment containing the inducible class-III chitinase (if3) promoter was isolated from L. albus, using the GenomeWalkerTM kit, for use in the pCAMBIA 0390:applePGIP:peroxidase defense gene construct, i.e. upstream from the shpx6a peroxidase gene. The 4.2kb fragment was successfully cloned into the pGEM-T Easy vector and sequenced. The sequence was compared to known sequences in the Genbank database but exhibited no significant homology. Using bioinformatic tools, five possible eukaryotic promoter-containing sites, including the TATA boxes, were identified within the isolated 4.2kb fragment. Deletion studies were performed in order to test for the minimal sequence needed for retaining of promoter activity. The 1.818kb, 1.512kb and 1.138kb if3 promoter-containing fragments were each cloned separately into the pDM327 vector upstream from the bar-gus fusion gene to produce pDM327:Prom1.8, pDM327:Prom1.5 and pDM327:Prom1.1 and used in the BiolisticTM transformation of plant tissue. BiolisticTM transformation of Ornithogalum and bean callus tissue, as well as maize and lupin immature embryos all demonstrated that the if3 DNA fragment isolated from L. albus contains promoter activity, indicated by the efficient stimulation of the expression of the gus reporter gene. Based on these results a provisional patent was filed [Application number: 2003/2405, and entitled “Plant Promoter”].
Bioinformatic analysis indicated the presence of various putative cis-acting regulatory elements, that could be important in controlling the expression of the 1.8kb if3 promoter-containing fragment. A single putative MBS regulatory cis-acting element was present in the 1.13kb promoter-containing fragment. It acts as a Myb transcription factor binding site that regulates transcription of several plant genes in response to various environmental factors, including elicitors and wounding. Several CAAT boxes were also identified within the 1.81kb promoter-containing fragment which play an important role in the determination of promoter efficiency. Most of the putative fungal elicitor activated (Box-W1 and ELI-box3) and wound-inducible [WUN-motif and ERE (ethylene responsive element)] cis-acting elements were present in the 1.13kb promoter-containing fragment. This supports the hypothesis that all regulatory elements needed for the activation of the if3 gene promoter are located within the first 1.13kb fragment upstream from the initiation codon of the if3 gene.
The final evaluation of the main hypothesis that the combinatorial approach, by using two defense genes, will be much more effective than one gene or natural resistance in the suppression of anthracnose in lupin will need to be evaluated once successful transformation and regeneration of lupin has been obtained.
Prof. Ian Dubery