Developing a Functional Genomics Approach to Berry Ripening and Defense

Our work with fruit ripening and defense related genes is now complete with regard to collecting EST sequences. With the ones we have already collected and the many thousands that are now planned in efforts on this campus and internationally we feel that it will be possible to clone nearly any grape gene we are interested in by application of sequence information and PCR methods. This means that studies where information about gene expression in grape is needed can be undertaken with confidence that most of the genes of interest can be easily cloned. Nevertheless, the picture that has emerged from our work with the veraison stage library is very informative and is a major research accomplishment and result.

We divided the EST sequences we obtained into three classes. Class one are genes for which the role in fruit development or ripening is already known in other systems (e.g. tomato). An example of class-one genes would be polygalacturonase. The second class includes those genes whose function can be known with a high degree of certainty but where its role in fruit development or ripening is unknown. An example of class-two genes would be a tonoplast intrinsic protein. The third class contains those genes that are unknown (a match was found in the data base but no function has been assigned to the protein) hypothetical proteins (e.g. an open reading frame from Arabidopsis) or no match (i.e. no match at all was found in the data base). In our total EST collection 34%of the genes were assigned to class 1; 44%to class 2; and 22%to class 3.

In order to provide a functional classification for the genes in class 1 and class 2 above, we assigned each gene to a functional category. The major group in the functional category contained genes related to stress responses; either oxidative, osmotic or water stress. We found that 22%of the genes were stress related and 20%were related to protein synthesis, processing and degradation. We found that 18%of the genes to which we could assign a function are related to disease resistance, and 8%involved with signal transduction. There were 9%related to RNA processing or were known transcription factors. Other groups with about 5%each were cell wall chemistry, secondary metabolism and photosynthesis.

The view of grape berry ripening that has emerged from our work is much different than we expected when we began to sequence ESTs from veraison berries. The large number of stress induced genes and disease related genes we found was surprising and unexpected. Nevertheless, this result has drawn our attention to the function of stress responses and plant defense gene expression in fruit ripening and berry composition.

PDF: Developing a Functional Genomics Approach to Berry Ripening and Defense

Enhancement of Stress Tolerance in Vitis vinifera

Abiotic stresses affect important aroma, flavor and color components by altering metabolite composition, improving wine quality and human health benefits. Regulated deficit irrigation has been used successfully to grow grapes with less water, an important feature in arid regions such as Nevada. As a first step toward understanding how growth is affected and wine quality improvements might arise following abiotic stress exposure, we have initiated an expressed sequence tag (EST)-based gene discovery program focused solely on stressed plants. We constructed cDNA libraries from mRNA isolated from leaf and berry tissues of Vitis vinifera cv. Chardonnay, exposed to various abiotic stress conditions. To date, we have sequenced over 3000 leaf ESTs and anticipate completing another 5000 berry sequences over the next few months before funds run out this year. Raw sequence data were processed through an automated EST analysis pipeline (ESTAP) developed at the Virginia Bioinformatics Institute (VBI; Blacksburg, VA) in collaboration with UNR and S.R. Noble Foundation (Ardmore, OK). Initial sequence analysis reveals 36%novel genes and a low redundancy of transcripts. All 1878 unique EST data generated to date have been deposited in GenBank and is freely available to the public.

In the context of genetic engineering more cold tolerant grapes, we have successfully transformed and regenerated V. vinifera in our laboratory. Embryo culture was successfully initiated using 0.5 to 1 mm immature anthers that were excised from flowers of Chardonnay plants. The excised anthers were placed on one of the following callus-initiation mediums: NB, PT or PIV medium. Tissues were subcultured to fresh plates every 6-8 weeks. Embryogenic calli suitable for transformation formed on some cultures. Single cell somatic embryos were transformed with CBF1 and CBF3 constructs including the CAMV35S promoter. The CBF/DRE transcriptional activators, CBF1 (DREB1B), CBF2 (DREB1C) and CBF3 (DREB1A) are some of the master switches for drought, salinity and cold tolerance. Eleven transgenic plants have been regenerated after selection on Kanamycin media. Three plants have been positively identified by PCR for transformation with CBF3. Confirmation of transformation by PCR and Southern analysis for the rest of the plants is underway. A second batch of transformed somatic embryos is currently going through the regeneration process. Degenerate primers designed to motifs of the CBF gene family have produced 8 distinct PCR products. These products represent putative grape CBF orthologs. Additional CBF orthologs are expected to be obtained from our on going EST sequencing program.
Must samples were obtained from well-watered and drought-stressed Chardonnay grapes. Standards were developed for gas chromatography/mass spectroscopy in preparation for must analysis.

PDF: Enhancement of Stress Tolerance in Vitis vinifera

A Genetic Map of Vitis Vinifera: A Foundation for Improving the Management of

We have continued to steadily add information to a genetic linkage map of Vitis vinifera. We have now mapped 145 microsatellite DNA markers to 18 linkage groups. New DNA markers have been developed within the Vitis Microsatellite Consortium, 20 cooperating research groups in 12 countries. Our map is now the most advanced genetic map of Vitis vinifera in the world. It is a resource that can be used by many viticulture researchers to facilitate the identification and isolation of individual grape genes so that we can learn how these genes work and how their functions are influenced by external factors, such as vineyard cultural practices. Our map is now being used by research groups in France, Germany, Italy, Spain and Australia to connect genetic maps being developed in their programs and to further gene identification efforts. These groups are also sharing their own results with us. We have also analyzed quantitative fruit and cluster characteristics that contribute to berry size and cluster tightness. Most of these characteristics map to only 3 regions of the grape genome, suggesting that key genes controlling berry size and cluster structure are located there.

Developing a Functional Genomics Approach to Berry Ripening and Defense

For the past year we have focused our work on gene discovery in grape. We are using a gene library from veraison stage fruit as the source of the new genes. We screened more than 700 plasmids to identify candidates for DNA sequencing, and identified 312 that appeared to have inserts large enough to identify the gene once its sequence was obtained. Preparation for sequencing required growing selected E. coli colonies overnight and then extracting plasmid DNA. One round of sequence was obtained using T3/T7 sites on the pBluescript as sequencing primers. After the sequences were acquired they were examined for open reading frames (ORF) and when an ORF was found it was submitted for a BLAST search to determine if there was homology with known genes. We have obtained many important and interesting genes that are expressed during veraison. The results obtained thus far constitute a list of genes that can be organized into related groups. The groups are based on their physiological and biochemical functions. The groups are: l)Protein Synthesis, Processing and Turnover; 2 Abscisic Acid and Water Stress; 3 Oxidative Stress and Redox; 4 Cell Wall and Cell Wall Management; 5 Plant Hormone and Signal Transduction; 6 Transcription Factors; 7 Enzymes of Primary and Secondary Metabolism/Structural Proteins. Several of the genes have very interesting roles and have been shown to be valuable in other systems. For example we found a pectate lyase that has been shown to cause preactivation of defense genes in transgenic potato, providing resistance to the pathogen Erwinia carotovora. However, the role of the pectate lyase in grape berries is unknown. Some of the proteins expressed at veraison seem to be involved with water stress or related to abscisic acid (ABA), the plant hormone commonly associated with responses to water stress. Finding this category was somewhat surprising but is clearly an important area for further study. It has been recognized for some time that ABA has a role in grape ripening, but it may be that its role is associated with water stress experienced by the berry at veraison. We now have several genes related to ABA and water stress with which to address the role of ABA in ripening, and its possible association with water stress. Our results in the past year have pointed out several important features of berry physiology that were unexpected. We have found several genes related to plant hormones such as auxin, ABA and ethylene. It is well known that hormones are important in berry ripening, and we now have clues as to which of these might be important, and tools to study their effect on expression of specific genes. Taken together, the new view of berry ripening that is emerging from our results may perhaps be the most important accomplishment of the 1999 season.

Genetic Transformation: A Means to Add Disease Resistance to Existing Grape

Methods for initiating embryogenic cultures, the starting material for genetic transformation, have been improved and large numbers of embryogenic cultures suitable for transformation have been produced from Chardonnay and Thompson Seedless. A better strain of Agrobacterium (the gene delivery organism) was obtained. Methods for selecting and isolating putatively transformed somatic embryos have been improved. Transformation experiments with liquid cultures have not been successful. Several experiments with solid cultures of Chardonnay, Thompson Seedless and St. George, however, have yielded cultures that appear to be uniformly transgenic in that they express the blue color of the introduced GUS marker gene throughout the tissue. Some of these putatively transgenic embryos are being induced to germinate into plants so that they can be further tested for the presence and expression of the introduced gene.

Identification and Characterization of Genes that Control the Phenolic

We isolated a cDNA clone for grape 4-coumaryl-CoA ligase (4CL) which is nearly full length. This is an enzyme that is positioned at an important branch-point in phenolic metabolism. We inserted the partial cDNA clone into an expression system, and initial experiments indicate that we can obtain sufficient quantities of purified grape 4CL for antibody production. We also inserted a 4CL clones from poplar into an E.coli expression system and demonstrated that permeabilized cells containing poplar 4CL are able to produce caffeoyl-CoA. Although the yield of caffeoyl-CoA is lower than we had hoped, it will be more than adequate for use in the assay of tartrate O-hydroxycinnamoyltransferase from grape. We isolated cDNA clones for grape dihydroflavonol reductase and a glucosyl transferase. The glucosyl transferase may be involved in glycosylation of phenolic compounds in grape, such as glycosylation of anthocyanidins to give the anthocyanins found in the hypodermal cells of the berry skin. The substrate specificity of the glucosyl transferase has not been determined, thus its exact role in phenolic biosynthesis has not been confirmed. The dihydroflavonol reductase is important because the product of the reaction (flavan-3,4-diols )can be directed into several different classes of phenolics in grape berries such as catechin, epicatechin, tannins and anthocyanins. Like the 4CL enzyme described above, the dihydroflavonon reductase occurs at an important branch point in phenolic metabolism in grape.

A Genetic Map of Vitis vinifera: a Foundation for Improving the Management of

We continued to add markers to a genetic linkage map of Vitis vinifera. During the past year we added primarily microsatellite (SSR) markers. We have now mapped 216 markers (94 AFLP and 122 SSR) and an additional 91 markers have been analyzed but the linkage analysis is not yet complete. The mapped markers form 19 linkage groups, the number expected in grape. Several fruit and cluster characteristics have also been analyzed and they map to three of the 19 linkage groups. New SSR markers are being developed within the Vitis Microsatellite Consortium, 20 cooperating research groups in 11 countries. The genetic map being created will be a resource that can be used by many viticulture researchers. It will facilitate the isolation of individual grape genes so that we can learn how the genes work and how their functions are influenced by external factors, such as vineyard cultural practices.

Identification and Characterization of Genes that Control the Phenolic

Many of the crucial branch points in phenolic biosynthesis occur at the level of the coenzyme A (CoA) esters of the hydroxycinnamic acids, and formation of the CoA esters is a critical step in the biosynthesis of virtually all classes of phenolics in grape. Research on hydroxycinnamate CoA ligases in grape has been very limited. Purification of the enzyme has not been accomplished, and research has been hampered because cDNA probes are not available to study expression of the genes. In the first year of this project we were able to extract and stabilize cinnamoyl CoA ligase (4CL) from grape tissues. We adapted a procedure for enzyme extraction and assay that was described for 4CL from aspen (Populus tremuloidies) so that it can be used with grape tissues and we studied the substrate specificity of the enzyme from leaf and green shoot tissue. In leaves we found that the enzyme showed much more activity with caffeate than with 4-coumaric acid. Activity was also observed with ferulate and sinapate in these extracts. This result is notable because the enzyme from other plants typically shows more activity with 4-coumaric acid than with caffeate. The total enzyme activity extracted from developing shoots was much lower than from leaves and showed substrate specificity more typical of enzymes from other sources, preferring 4-coumaric acid over caffeic acid. Also, with shoot extracts no activity was observed when sinapic acid was used as substrate. These results are significant because they might suggest that there are different isoenzymes present with different substrate specificities in different tissues. Thus, extraction and assay of the enzyme has been successful and we have found that best source of the enzyme from grapevine appears to be developing leaf tissue. We obtained two cDNA clones of 4CL from poplar from a laboratory in British Colombia. These cDNAs were labeled with 32P and used to screen a grape cDNA phage library prepared from mRNA obtained from grape berries at the beginning of ripening. We isolated six strongly positive plaques. The DNA from the positive phage were amplified and have been sent for DNA sequencing. We will know very soon whether or not we have obtained clones of 4CL from grape berries. This would be important because we could then use the grape 4CL clones to study expression of the respective genes in grapevine.

Identifying Varieties and Clones by DNA Typing

We continued to investigate possible ways to distinguish clones within important winegrape varieties. We tested a method called ISSR and were able to detect some differences in Chardonnay and Pinot noir clones but they were not sufficiently reproducible to be useful. We also tried a selection of new SSR markers developed within the Vitis Microsatellite Consortium and found that 8 of 12 markers we tried could detect some differences between clones of Chardonnay or Pinot noir or both. We obtained and analyzed 150 accessions of Plavac Mali from Croatia for comparison with Zinfandel and found that Plavac Mali is not Zinfandel, contrary to some opinions. Surprisingly, we also found that Plavac Mali is not a polyclonal variety as we had presumed and that the accessions were almost completely uniform in their DNA profile.

A Genetic Map of Vitis vinifera: A Foundation for Improving the Management of

The goal of this project is to develop a basic genome map for grape (Vitis vinifera) that will allow us to begin to locate the genes that control important viticultural and enological characteristics, such as disease resistance and fruit composition. This will not only allow us to ultimately move these genes from one variety to another, whether by traditional or biotechnological means, but it will also facilitate the study of how these genes work and how they are affected by environmental and cultural conditions. The development of a genome map requires a population of progeny individuals derived from a cross between two disparate parents. We have used Cabernet Sauvignon and Riesling, two quite different wine grape cultivars, and have a population of 116 seedling vines derived from this cross that is now 4 years old. We have now obtained 178 DNA “markers” that we are placing in positions along the different grape chromosomes. The DNA markers are like signposts along a road. We have located some of these markers on specific chromosomes but others are not yet assigned to a chromosome. Cluster structure (e.g., compactness) is among the many characteristics that are under genetic control. Tight clusters are prone to rot. Loose clusters tend to have smaller berries, which are often preferable for winemaking. It is likely that many genes are involved in determining cluster structure. Some may determine the number of flowers that form on a cluster; others may determine the maximum berry size; and others may determine the length of the pedicel or the branching pattern of the rachis. We are trying to sort out these various components of cluster structure, to determine how many genes control them and to find the genome location for these genes. We are collecting data on 9 berry and cluster characteristics from all of our seedling vines in our mapping population but, because the vines are still quite young, we have only 1 year of data and will need several more before we can begin to interpret this data. Most of the DNA markers that we have been using for our genome map are of the type called AFLP markers. It is relatively easy to generate large numbers of these markers, but they have some limitations and information gained with these markers cannot always be shared with other researchers who are working with different mapping populations. Microsatellite markers, on the other hand, are more powerful and can be used on any mapping population. Unfortunately, these markers are much harder to come by and their discovery and development is very laborious. In order to obtain a large number of microsatellite markers, we have formed an international Vitis Microsatellite Consortium in which researchers in several countries will share in the effort to develop new grape microsatellite markers and will then share in the benefits. After about 10 months of correspondence, organization and the negotiation of a written agreement, the consortium is now underway and up to 20 grape research groups in 10 countries are expected to ultimately join in the effort.