1 CSIRO Plant Industry, Queensland Bioscience Precinct, 306 Carmody Road, St. Lucia, Brisbane, Australia 4067 Email email@example.com
2 Cooperative Research Centre for Sugar Industry Innovation through Biotechnology, http://www.crcsugar.com/
Sugarcane is an important crop in tropical and sub-tropical regions. A major objective of sugarcane improvement programs worldwide is to increase the stem sucrose content. However, unlike yield, the sugar content of commercial cultivars has not been increased in the last 40 years, at least in Australia. This paper describes recent attempts and current strategies to identify genes associated with sucrose accumulation. Differential gene expression studies using microarrays to compare immature internodes with internodes rapidly accumulating sucrose have identified candidate genes associated with sucrose accumulation. Of the genes encoding proteins involved in sugar metabolism and storage a group of sugar transporters have been shown to be the most highly up regulated in the maturing stem. These genes and their gene products have been further studied by northern analysis, in situ hybridisation and immunolocalisation allowing hypotheses to be generated about their potential roles in the sucrose accumulation process. Other differential gene expression studies using microarrays to compare internodes of high and low sucrose content genotypes from a segregating population have also been undertaken and identified candidate genes. The role of the developing area of metabolomics in candidate gene identification by discriminating metabolites and their relative abundance is also discussed.
Increasing sugar content of sugarcane would greatly increase the profitability of growing sugarcane. We are employing genomic and metabolomic techniques to identify the genes that are important in that process.
Saccharum, sugarcane, sugar content, sucrose, differential screening, microarrays, expressed sequence tag, metabolomics
Sugarcane is one of the most important field crops grown in the tropics and sub-tropics. In sugarcane improvement programs, most effort has been spent on increasing yield and sucrose content. Whilst improvements in yield have continued to accrue, in Australia there has been no increase in sugar content over the last 40 years (Jackson 2005). Improvement in sucrose content is desirable because there are greater benefits from producing the same amount of sucrose in less biomass, a result of the reduced costs of harvesting, transport and processing whilst production costs remain the same. Therefore, more strategic approaches have been initiated to increase stem sucrose content by identifying the genetic factors controlling sucrose content. One way in which genetic regions associated with sucrose content have been identified is through quantitative trait locus (QTL) analysis of DNA markers scored in the progeny of segregating populations (Hoarau et al., 2002; Ming et al., 2002). The identification of QTLs does not rely on any understanding of how a region of a particular linkage group contributes to the trait or which particular DNA sequence is involved. Therefore, as a complementary approach, we are using a variety of means to identify particular genes that are associated with the high sucrose phenotype. These candidate genes will be tested for their role in sucrose accumulation through functional assays or by using the DNA sequences as molecular markers. These markers can then be used to explore new QTLs or to test for co-location with previously found QTLs. Possible routes to exploit such information are marker-assisted selection if genetic variation exists, or altered gene expression using a transgenic approach. Genes identified via these approaches are also good candidates for further analysis of genetic variation through SNP analysis as described by Manners et al. (2004). This paper outlines the strategies being used to find genes associated with higher sucrose as the sugarcane stem develops and with genotypes that accumulate high levels of sucrose.
Our initial approach to assess the transcripts associated with sucrose accumulation was to sequence randomly chosen clones from cDNA libraries representing genes expressed in either immature stem internodes or internodes that are rapidly accumulating sucrose. These cDNA sequences are termed expressed sequence tags (ESTs). Genes associated with higher sucrose (accumulating internodes) were determined by (i) the relative abundance of particular ESTs in the two libraries and (ii) the differential expression of particular genes on microarrays when comparing RNA extracted from immature and maturing stem (Casu et al., 2003, 2004). The most up-regulated genes associated with carbohydrate metabolism were sugar transporters, the most abundant being a putative hexose transporter (Casu et al., 2003). The most abundant genes of all were apparently unrelated to sucrose accumulation, rather taking part in the secondary thickening and lignification of internodes as they mature (Casu et al., 2004). There was also a group of up regulated genes that were putatively involved in stress tolerance (Casu et al., 2004). Genes with a role in stress tolerance are being investigated further to help understand the processes at a tissue level that are necessary for the accumulation of high sucrose in the stem storage cells. Similar comparisons can be made between genotypes that accumulate different amounts of sucrose. Such analyses are potentially more informative and more powerful when performed using segregating populations as variation in gene expression in non-relevant traits is minimised. We have recently described this “genetical genomics” approach and believe it warrants further exploration (Casu et al., 2005).
The ability to conduct microarray experiments with greater coverage of the genome or even more targeted micro- or macro-array experiments (Watt et al., 2005) is now feasible due to the large collection of ESTs from sugarcane recently lodged in public databases. It is now possible to create a list of genes for representation on an array, search the sugarcane EST collection for homologous DNA sequences, devise a DNA amplification strategy and generate a matching insert containing the desired DNA sequence for representation on arrays. This approach is currently being used to supplement our own EST collection in the design of our next microarray.
Examining the proteome of the sugarcane stem is another approach that we have not yet fully explored. The development of the field of metabolomics, particularly in the last 5 years, has opened up the possibility of looking at the end products of both gene expression and protein catalysis. Two ways of using data from metabolic studies in sugarcane have immediate appeal in the pursuit of improving sugar content. The first would be analogous to the anonymous DNA marker and QTL approach. Profiles of metabolites would be generated for different groups of genotypes that differ in their level of sucrose accumulation. The data can be used diagnostically without peak identification to determine patterns of metabolites that discriminate between genotypes (Fiehn et al., 2000). As illustrated in Figure 1 for one sugarcane cultivar, the differences between tissues are stark and the techniques have been shown to be sufficiently sensitive to discriminate tissues from different genotypes in other species (Fiehn et al., 2000). Like the DNA-based analysis, individual metabolite peaks responsible for the difference can be identified and the genes coding the enzymes acting on the metabolite, or its precursors, be studied further. This is the second more targeted way of utilising metabolomics data.
Metabolomics is a non-biased approach trying to identify all of the metabolites in a sample. However, its attempt to be all encompassing does mean that there will be some compounds missed as protocols are designed to identify a good cross section of metabolite types. A more targeted approach is the identification of metabolites more closely related to carbohydrate metabolism, which will rely on an adaptation of standard pre-fractionation and treatment of samples. However, as our microarray expression studies have shown, not all of the changes may be directly related to carbohydrate metabolism, so an initial broad sweep may be fruitful.
Several candidate genes identified in the microarray expression studies discussed above have been studied further. The simplest studies involve isolation of RNA from several tissues to analyse expression along the length of the stem and in other non-storage tissues. This method was used to demonstrate the stem specificity of the putative hexose transporter (Casu et al., 2003). We have subsequently been able to fractionate the stem into storage parenchyma and vascular tissue and used RNA from these tissues to further dissect expression patterns (Rae et al., 2005). A similar approach is being used to compare expression levels in different genotypes either from segregating populations or from a diverse set of sugarcane germplasm with different sucrose concentration. These latter experiments are designed to further investigate the association between sucrose content and expression levels of the particular transcript whilst also providing evidence of genetic variation.
Figure 1. GC chromatogram of compounds eluting from polar extracts of mature and immature leaf of sugarcane Q117. The x axis is time in minutes. The extracts were derivatised with methoxlyamine/pyridine and MSTFA (N-Methyl-N-(trimethylsilyl) trifluoracetamide) and run on a Rtx-5Sil MS 30 m x 0.25 μm ID with 10 m Intergra guard (Restek, USA) column with an initial temperature of 80 oC holding for 2 min then increasing by 5 oC a min to 320 oC, which was maintained for 6 min. Note the different profiles for mature leaf (top) and immature leaf (bottom). Example peak identifications, peak at 26.1 min (fructose), 26.6 min (glucose) and 30.4 min (sucrose).
An example of this approach is shown in Figure 2. RNA was extracted from the stems of a variety of sugarcane genotypes including wild progenitor species, such as Erianthus and Saccharum spontaneum, and commercial hybrid cultivars. When the expression of the putative hexose transporter, PST2 (Casu et al., 2003) was examined in these RNA samples, striking differences in the level of expression were found. By correlating the relative expression level with sucrose concentrations, it may be possible to identify genes associated with high sucrose accumulation.
Figure 2. Abundance of transcripts of PST2 in RNA extracted from stems of 14 sugarcane genotypes. The lower panel shows the same membrane probed for ribosomal RNA to demonstrate RNA loading. The varieties used were: 1, Erianthus; 2, SES106 (Saccharum spontaneum); 3, Mandalay (S. spontaneum) ; 4, NG57-54 (S. robustum); 5, NG57-56 (S. robustum); 6, IJ76-237 (S. officianarum); 7, IJ76-567 (S. officianarum); 8, NG77-98 (S. officianarum); 9 Badilla (S. officianarum); 10, Q28; 11, Q117; 12, Q124; 13, Q165; 14, Q200.
The spatial distribution of particular transcripts and their gene products are being explored using in situ hybridisation of RNA and immunolocalisation of proteins. These methods can be used to identify the cells in which particular transcripts are expressed. Using in situ hybridisation, Casu et al. (2003) showed that the PST2 was highly expressed in sugarcane phloem companion cells, while Rae et al. (2004), using immunolocalisation, showed that a sucrose transporter (ShSUT1) could be localised to the layer of cells surrounding the vascular bundle. Through experiments of this type it will be possible to localise many of the putative sugar transporters and other enzymes of sugar metabolism found in our EST collection. The results of these experiments will give us a greater understanding of the spatial arrangement of the sugar transport system and how this changes developmentally down the stem. The results to date are already providing evidence for a highly regulated movement of different sugars through different cells to the site of storage in the stem parenchyma.
The ultimate test of a gene’s function is its expression in a heterologous system. We have expressed the sugarcane sucrose transporter ShSUT1in a yeast mutant that does not take up sucrose. The complemented yeast was shown to take up sucrose at a faster rate than the untransformed yeast (Rae et al., 2004), confirming its function as a transporter of sucrose. Demonstration of function in this way is time consuming and requires individual experimental design for each transcript being tested. Therefore, it is essential that other methods be used first to demonstrate a robust correlation with sucrose accumulation prior to embarking on functional assays.
Genomic expression and metabolic studies are expected to identify many candidate genes involved in sucrose accumulation. These candidate genes can be further tested for their association with sucrose accumulation by comparing expression in organs, tissues and even cell types. Performing these validations in genotypes differing in levels of sucrose accumulation further strengthens the correlation. Functional assays of these genes are more difficult, however, an understanding of the function and the role of a particular gene may allow the development of new strategies for crop improvement that are not possible by anonymous DNA marker-based approaches alone.
Casu RE, Grof CPL, Rae AL, McIntyre CL, Dimmock CM, Manners JM. (2003). Identification of a novel sugar transporter homologue strongly expressed in maturing stem vascular tissues of sugarcane by expressed sequence tag and microarray analysis. Plant Molecular Biology, 52 371-386.
Casu RE, Dimmock, CM, Chapman SC, Grof, CPL, McIntyre, CL, Bonnett, GD and Manners JM (2004) Identification of differentially expressed transcripts from maturing stem of sugarcane by in silico analysis of stem expressed sequence tags and gene expression profiling. Plant Molecular Biology 54 503-517.
Casu RE, Manners JM, Bonnett GD, Jackson PA, McIntyre CL, Dunne R, Chapman SC, Rae AL and Grof CPL (2005) Genomics approaches for the identification of genes determining important traits in sugarcane. Field Crops Research in press.
Fiehn O, Kopka J, Dormann P, Altmann T, Tretheway RN, Willmitzer L. (2000). Metabolite profiling for plant functional genomics. Nature Biotechnology, 18 1157-1161.
Hoarau J-Y, Grivet L, Offmann B, Raboin L-M, Diorflar J-P, Payet J, Hellmann M, D’Hont A, Glaszmann J-C, (2002). Genetic dissection of a modern cultivar (Saccharum spp.). II. Detection of QTLs for yield components. Theor Appl Genet 105: 1027-1037.
Jackson PJ (2005). Breeding for improved sugar content in sugarcane. Field Crops Research in press.
Manners J, McIntyre L, Casu R, Cordeiro G, Jackson M, Aitken K, Jackson P, Bonnett G, Lee S, and Henry R. (2004). Can genomics revolutionise genetics and breeding in sugarcane? Proceedings of the 4th International Crop Science Congress.
Ming R, Wang Y-W, Draye X, Moore PH, Irvine JE, and Paterson AH. (2002). Molecular dissection of complex traits in autopolyploids: mapping QTLs affecting sugar yield and related traits in sugarcane. Theor Appl Genet 105: 332-345
Rae AL, Perroux JM, and Grof, CPL (2004). Sucrose partitioning between cvascular bundles and storage parenchyma in the sugarcane stem: a potential role for the shSUT1 sucrose transporter. Planta 220 in press.
Rae AL, Grof, CPL, Casu RE, and Bonnett GD. (2005). Sucrose accumulation in the sugarcane stem; pathways and control points for transport and compartmentation. Field Crops Research in press.
Watt D, McCormick A, Govender C, Carson D, Cramer M, Huckett B, and Botha F. (2004). Increasing the utility of genomics in unravelling sucrose accumulation. Field Crops Research in press.