DIRECT ORGANOGENESIS AND AGROBACTERIUM-MEDIATED TRANSFORMATION OF EXACUM STYER GROUP by FARIDE UNDA BSc. (Agriculture Engineering) IASA-ESPE, Ecuador, 1998 A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Plant Science) UNIVERSITY OF BRITISH COLUMBIA March, 2006 ©Faride Unda, 2006 ABSTRACT 11 Interspecific hybrids of Exacum, formally known as Exacum Styer Group, have significant potential as a new ornamental species. However, these hybrids are very susceptible to fungal root pathogens including Fusarium species. Therefore, the development of fungal resistant germplasm is highly desirable. In an attempt to address this goal, a genetic engineering protocol based on Agrobaclerium-medlated transformation using a gene reported to confer fungal resistance was developed. A method for the regeneration of plantlets through direct organogenesis from leaf explants of Exacum, without an intervening callus phase, was developed. Four genotypes were evaluated on M S medium supplemented with T D Z (0.01, 0.05, 0.1 mg 1"') or combinations of B A (0, 0.1, 0.5, 1.0 or 2.0 mg 1"') and N A A (0, 0.01, 0.1 or 0.5 mg 1"'). Leaves of Exacum were very responsive to the plant growth regulators; different combinations induced different organogenesis (i.e. roots, shoots or callus). Significant genotype, media, and genotype x medium interactions were present for several variables. However, genotypes 01-09-01 and 01-37-61 had the highest number of shoots per explant across media (10.2 and 6.6, respectively), while 1.0 mg l " 1 B A plus 0.1 mg 1"' N A A induced the greatest number of shoots among the genotypes evaluated. After establishing organogenesis protocols, I focused on the development of an Agrobacterium-mediated transformation protocol. Using modified fhaumatin-like protein (TLP) constructs, nine transformed lines were produced using the pSM-3 T L P plasmid with the Agrobacterium strain C-58 and three transformed lines were produced using the pCambia-bar-ubi T L P plasmid with Agrobacterium strain LBA4404 . Four tobacco I l l transgenic lines also were produced and used as positive controls. The transformation percentages for Exacum were low (<1%) compared to other plants that have been transformed with similar T L P constructs and bacterial strains. Results from R T - P C R demonstrated that the expression levels of the T L P gene varied significantly among transgenic lines of both Exacum and tobacco. Gene expression patterns showed higher transcript levels in plants transformed with the double 35S promoter/AMV leader sequence (pSM-3 plasmid) compared to plants transformed with the ubiquitin promoter (pCambia-bar-ubi). S D S - P A G E identified a small protein (-14 kDa) present in Exacum transgenic lines but absent from the non-transformed controls. However, it was not confirmed that this protein was T L P due to its smaller than expected size. Immunoblots displayed no signal from either Exacum or tobacco transgenic lines when probed with a rice polyclonal antibody. The lack of binding could be explained by the loss of activity of the antibody or by a post-translational modification of the protein antigens. Future studies should evaluate transgenic plants for enhanced resistance. IV TABLE OF CONTENTS Abstract i i Table of Contents iv List of Tables v i i List of Figures v i i i List of Acronyms x Acknowledgement x ' ' Dedication x i i i Chapter 1 General Introduction and Background 1 1.1 Introduction 1 1.2 Background 2 1.2.1 Exacum Styer Group 2 1.2.2 Pathology 3 1.2.3 Genetic Engineering to Enhance Disease Resistance in Plants 7 1.2.4 PR-Proteins 9 1.3 Literature Cited 13 Chapter II Organogenesis and Plant Regeneration from Leaf Explants of Exacum Styer Group 19 2.1 Introduction 19 2.2 Materials and Methods 21 2.2.1 Experiment #1 21 2.2.1.1 Plant Material 21 2.2.1.2 Treatments 21 2.2.1.3 Statistical Analyses 22 2.2.2 Experiment #2 22 2.2.2.1 Plant Material 22 2.2.2.2 Treatments 23 V 2.2.2.3 Statistical Analyses 23 2.3 Results 23 2.3.1 Experiment #1 ••• 23 2.3.2 Experiment #2 26 2.4 Discussion 38 2.5 Literature Cited 40 Chapter III Agrobacterium-mediated transformation of Exacum Styer Group using a TLP construct 42 3.1 Introduction 42 3.2 Materials and Methods 47 3.2.1 Plant Material 47 3.2.2 Transformation Vectors and Bacterial Strains 48 3.2.3 Hygromycin Sensitivity 51 3.2.4 Agrobacterium-med'iated Transformation 52 3.2.5 Molecular Analysis 54 3.2.5.1 Genomic DNA Extraction 54 3.2.5.2 PCR Analysis of Putative Transfonnants 55 3.2.5.3 RNA Isolation 56 3.2.5.4 Real Time RT-PCR Analysis 57 3.2.5.5 Total Protein Extraction 58 3.2.5.6 Immunoblot Assays 59 3.3 Results '. 60 3.3.1 Transformation Vectors and Bacterial Strains 60 3.3.2 Hygromycin Sensitivity 63 3.3.3 Generation of Transgenic Exacum 65 3.3.4 Generation of Transgenic Tobacco 72 3.3.5 Molecular Analysis of Gene Expression in Transgenic Lines 74 3.3.6 TLP Protein Expression 79 3.4 Discussion 83 3.5 Literature Cited 8 9 vi Chapter IV 95 General conclusions 4.1 Concluding Remarks and Future Research 95 Appendix 1 100 Evaluation of Zinc and charcoal supplements A I . l Introduction 100 A1.2 Materials and Methods 100 A 1.2.1 Plant Material 100 A 1.2.2 Treatments 101 AI.3 Results and Discussion 101 Appendix II 103 Agrobacterium-mediated transformation (previous experiments) A l l . 1 Floral dip 1 0 3 A l l . 1.1 Introduction 103 A I L 1.2 Materials and Methods 104 A l l . 1.2.1 Plant Material 104 A11.1.2.2 Treatments 105 A l l . 1.3 Results and Discussion 105 AII.2 Co-cultivation 106 AII.2.1 Introduction 106 AII.2.2 Materials and Methods 107 All.2.2.1 Plant Material 107 A11.2.2.2 Treatments 108 AII.2.3 Results and Discussion 108 Literature Cited !09 LIST OF TABLES Table 1.1 Distinguishing morphological characters of E. Styer Group Table 1.2 Summary of pathogen-related protein families Table 2.1 Effect of N A A and B A concentrations on shoot production from leaf explants of four Exacum genotypes after 6 weeks of culture Table 2.2 Regression coefficients for B A and N A A concentration and their interaction on shoot regeneration of Exacum genotype 01-09-01 Table 2.3 Regression coefficients for the effects of B A , N A A and their interaction on shoot regeneration of Exacum Table 3.1 Comparison of transformation efficiency among Agrobaclerium strains and Exacum explant sources Table 3.2 Transformation efficiency of Exacum transformed with the Agrobaclerium strain LBA-4404 carrying the pCambia-bar-ubi T L P plasmid Table 3.3 Transformation efficiency of tobacco transformed with the Agrobaclerium strain C-58 carrying the pSM-3 T L P plasmid Table A . 1 Percentages of germination of Exacum genotypes on kanamycin medium 3 12 28 31 68 71 106 LIST OF FIGURES viii Figure 1.1 Symptoms on Exacum Styer Group caused by Fusarium 5 solani Figure 1.2 Macroscopic and microscopic characteristics of Fusarium 6 solani Figure 2.1 Effects of P G R supplements on shoot regeneration from 25 leaf explants of Exacum Figure 2.2 Direct shoot organogenesis from leaf explants of Exacum 25 genotype 01-09-01 Figure 2.3 Overall mean comparisons on shoot regeneration of four 27 genotypes of Exacum. Figure 2.4 Graphical representation of the effects of B A and N A A 34 across genotypes on the expected values of shoot regeneration from leaf explants of Exacum. Figure 2.5 Three dimensional representation of the overall effects of 35 B A . N A A and their interaction on shoot regeneration of Exacum. Figure 2.6 Overall effects of different B A + N A A combinations on root 36 organogenesis of Exacum across genotypes Figure 2.7 Effects of different B A + N A A combinations on Exacum 37 organogenesis Figure 3.1 Gel electrophoresis of the C-58 Agrobaclerium strain 62 transformed with the pSM-3 T L P transformation vector Figure 3.2 Schematic representation of transformation vector pSM-3 62 T L P Figure 3.3 Hygromycin sensitivity results at 6, 15 and 26 days on two 64 Exacum genotypes Figure 3.4 Shoot organogenesis from Agrobaclerium-'mfecied tissue 67 on hygromycin medium IX Figure 3.5 Transgenic Exacum shoot on charcoal elongation medium 67 with antibiotics for selection Figure 3.6 P C R products from transgenic Exacum lines transformed 69 with pSM-3 T L P construct Figure 3.7 P C R products from transgenic Exacum lines transformed 71 with the pCambia-ubi T L P construct Figure 3.8 Putative transgenic tobacco plants on hygromycin medium 73 Figure 3.9 Graphical outputs of gene expression of transgenic Exacum 76 and tobacco lines obtained by real time R T - P C R Figure 3.10 T L P expression levels of transgenic Exacum lines 77 transformed with the Agrobacterium strain C-58 carrying t h e p S M - 3 T L P Figure 3.11 T L P expression levels of transgenic tobacco lines 77 transformed with the Agrobacterium strain C-58 carrying the pSM-3 T L P Figure 3.12 T L P expression levels of transgenic tobacco lines 78 transformed with the Agrobacterium strain LBA4404 carrying the pCambia-bar-ubi T L P Figure 3.13 Protein separation by SDS P A G E of Exacum transgenic 80 lines transformed with the Agrobacterium strain C-58 carrying pSM-3 T L P Figure 3.14 Protein separation by SDS P A G E of Exacum and tobacco 80 transgenic lines transformed with the C-58 and LBA-4404 Agrobacterium strains Figure 3.15 Protein separation by SDS P A G E of Exacum and tobacco 81 transgenic lines transformed with the C-58 Agrobacterium strain carrying the p S M - 3 - T L P plasmid Figure A . l Evaluation of M S medium on Exacum genotype 01-37-61 102 LIST OF ACRONYMS 2,4-D 2,4-dichlorophenoxyacetic acid 2iP 6-(y,y-dimethylallyamino) purine Aa amino acids AMV alfalfa mosaic virus ANOVA analysis of variance Avr avirulance genes BA benzyladenine Bar gene encoding resistance to the herbicide bialaphos BCIP 5-bromo-4 chloro-3-indolylphosphate, toluidine salt Bp base pair CaMV 35S Cauliflower Mosaic Virus 35S promoter cDNA complementary D N A Ct cycle threshold value CTAB hexadecyltrimethylammonium bromide DEPC diethylpyrocarbonate DNA deoxyribonucleic acid DTT dithiothreitol EDTA ethylenediamine tetra acetic acid GE genetic engineering GUS beta-glucuronidase HPT hygromycin phosphotransferase HR hypersensitive response Kb kilobase kDa kilodaltons LB Luria broth medium mRNA messenger R N A MS Murashige and Skoog NAA a-naphthaleneacetic acid NBT nitroblue tetrazolium chloride NCBI National Center for Biotechnology Information OD optical density (absorbance) PAGE polyacrylamide gel electrophoresis PCR polymerase chain reaction Pfu polymerase from Pyrococcus furiosus PGR plant growth regulator PPT phosphinothricin PR pathogenesis related PSMF phenylmethylsulfonylfluoride PVP polyvinylpyrrolidone QPCR quantitative P C R R resistance genes RNA ribonucleic acid Rpm revolutions per minute RT-PCR reverse transcriptase P C R SD standard deviation SDS sodium dodecyl sulfate SE standard error Taq polymerase from Thermits aquaticus TBS Tris-buffered saline T-DNA transfer D N A TDZ thidiazuron (N-phenyl-N' 1,2,3-thidiazol-5ylurea) Ti plasmid tumor inducing plasmid from Agrobacterium tumefaciens TLP thaumatin-like protein TMV Tobacco Mosaic Virus tRNA transfer R N A Ubi ubiquitin promoter X l l ACKNOWLEDGEMENTS I gratefully acknowledge the helpful advice of my supervisor Dr. Andrew Riseman and members of my advisory committee, Dr. Shawn Mansfield, Dr. Zamir Punja, and Dr. Brian Ell is for their invaluable suggestions and expertise. Special thanks to Dr. Mansfield and all members of the Mansfield Lab, particularly to Ji Young Park and Thomas Canan for their unending help, suggestions and protocols. Thanks also to all my colleagues in the Centre for Plant Research, primarily Peter Kalynyak and Diane Edwards for their incessant support and encouragement and to Dr. Kathy Baylis for her statistical guidance. To my parents, words cannot express how grateful I am for the years of support, love and phone calls. Y o u have instilled values in me that wi l l get me through life's hardest challenges. Finally, to my husband and my daughter, your patience, love and understanding have gotten me through many tough times. Y o u are proof that fate has definitely dealt me the winning hand. X Chapter I 1 General Introduction and Background 1.1 Introduction Interspecific hybrids of Exacum, formally known as Exacum Styer Group (Riseman el al, 2005), are derived from several taxa native to Sri Lanka and have significant potential as a new ornamental crop. This germplasm possesses large flowers (>2.5 cm) colored from sky blue to deep violet, and attractive dark green foliage. However, these hybrids are very susceptible to fungal root pathogens, including Fusarium spp. It is likely that once they are released to commercial producers, this susceptibility wi l l inhibit commercial success. Therefore, the development of fungal resistant germplasm is highly desirable. Several methods commonly used to create fungal resistant germplasm include classical intra-specific breeding, interspecific hybridization, exploitation of somaclonal variation, and genetic engineering (GE). In an attempt to introduce fungal resistance into Exacum, I chose to establish suitable genetic engineering protocols, based on Agrobacterium-medlated transformation, using a "fungal resistance' gene that could lead to the production of transgenic plants with an enhanced resistance to fungal pathogens. Use of genetic engineering for Exacum may provide a better alternative to classical breeding when developing germplasm with enhanced disease resistance. Not only could G E provide resistance with a defined mechanism or allow for the creation of novel forms of resistance, it may also allow development of this germplasm quicker than with sexual reproduction and classical breeding. However, it is important to note that G L 2 techniques rely heavily on suitable tissue culture and organogenesis protocols. Therefore, to successfully develop Agrobacterium-mediated transformation protocols for Exacum, optimal organogenesis protocols first need to be established and tested. The following objectives were set for this research: 1) to optimize micropropagation and regeneration protocols for Exacum; 2) to develop a novel thaumatin-like protein (TLP) construct with a hygromycin resistance sequence and a double C a M V 35S promoter; 3) to develop an Agrobacterium-mediated transformation system for Exacum using the novel T L P construct; and 4) to characterize expression patterns of the T L P transgene in transgenic plants. This thesis is divided into four chapters, which detail the steps followed to achieve each of the objectives. Chapter 1 presents a general background on topics directly related to the current research. Chapter 2 describes the development and establishment of direct shoot organogenesis protocols. Chapter 3 details the methodology used to produce a T L P construct and the production of transgenic Exacum plants. The final chapter presents the concluding remarks with suggestions for future research. 1.2 Background 1.2.1 Exacum Styer Group The genus Exacum L . (Gentianaceae Juss.) includes approximately 65 species of annual and perennial herbs (Klackenberg, 1985). Among these species, several taxa native to Sri Lanka were identified as possessing desirable horticultural traits and could serve as parental germplasm for the development of a new commercial crop (Riseman and Craig, 1995). In this context, a classical breeding program was established to 3 combine the most desirable characteristics from among five Sri Lankan Exacum taxa. The collected and intercrossed taxa included: E. pedunculatum L . , E. macranthum Arn . , E. pallidum Trimen., E. trinervium (L.) Druce, and E. trinervium spp. ritigalensis (Willis) Cramer. Through 12 sexual generations, plants were selected for fertility, greenhouse production, and other desirable horticultural traits. Specific combinations of traits are now present in these hybrids, but absent from the parental taxa, thus forming and defining E. Styer Group (Table 1.1) (Riseman et al, 2005). • Table 1.1 Distinguishing morphological characters of E. Styer Group (adapted from Riseman et al., 2005). Stem shape (height at anthesis) Leaf shape (length X width) Flower color RHS code (petal number) Calyx lobe length Petal apex arrangement (length X width) Anther length Cylindrical to quadrangular, with/without wings or lines (12-36 cm) Lanceolate to ovate or narrowly elliptic (5.5-11 x 1.8-3.7 cm) Pale blue to dark violet R H S 89B-C through 93A-C (5-merous) 3-24 mm Rhomboidal to broadly obovate, overlapping or not 20-28 x 16-21 mm 8-18 mm 1.2.2 Pathology Exacum Styer Group is known to be susceptible to fungal attack leading to basal stem and root rots. Typical symptoms initially appear as red to reddish-brown lesions on stems and primary roots that develop into larger necrotic regions. Infection is thought to start in the roots and gradually extend up to the soil surface. Foliar symptoms start with wilting and progress until the foliage turns yellow-brown eventually becoming necrotic (Figure 1.1); plant death is imminent by this stage. The causal agent, based on macroscopic and microscopic observations, has been identified as a member of the 4 Fusarium solani (Mart.) Sacc. complex. (Agriculture and Agri-Food Canada Eastern Cereal and Oilseed Research Center - National Fungal Identification Service, Ottawa). In addition, following Koch's postulates, F. solani was confirmed as the causal agent of this disease (Carrasco, 2002). The genus Fusarium is abundant in soil and is commonly associated with roots and other lower plant parts. Some members of this complex can be found as saprophytes growing on decaying plant parts, while other members are specialized pathotypes that cause vascular wilts, stem rots, fruit rots, grain diseases, and 'damping o f f , a disease of young seedlings (Booth, 1971). Infection of Exacum by F. solani is generally rapid. The pathogen typically penetrates the root hairs or epidermal cells just behind the root tip or within the zone of elongation. It then grows intercellularly and intracellularly through the root cortical tissue, where it enters the protostele and invades the differentiating tracheal vessels (Carrasco, 2002). More recently, additional observations (i.e. macroscopic and microscopic characters, disease progression and symptom development) were made on new isolates obtained from recently infected plants inside the U B C Horticulture greenhouse (Figure 1.2). I found characteristics identical to those described for Fusarium and confirmed the findings of Carrasco (2002). The observations made on our F. solani isolate are also consistent with reports from another gentian, lisianthus (Eustoma grandiflorum (Raf.) Shinners) (Wolcan and Lor i , 1996). In that report, infected lisianthus plants exhibited wilting with root lesions and necrosis of the lower stem; symptoms very similar to the ones produced by this fungus on Exacum Styer Group. 5 Fig. 1.1 Symptoms on Exacum Styer Group caused by Fusarium solani. a & b) chlorosis and wilt on leaves; c & d) stem infection at the plant's base leading to necrotic lesion formation; e) necrotic lesion expanding upward on the stem and f) necrotic roots. 6 Fig 1.2 Macroscopic and microscopic characteristics of Fusarium solani, a) in vitro colony growth on PDA medium; b) macroconidia (banana shape), c) 1. macroconidia and 2. microconidia; and d) chlamydospores (WF 40x). 7 1.2.3 Genetic engineering to enhance disease resistance in plants One of the major problems facing agricultural production is the difficulty in controlling plant diseases to maintain high quality and yield, while also meeting both producers' and consumers' expectations. While chemical treatments to control fungal diseases have proved sufficient in maintaining yield and quality, these treatments are also problematic for producers and consumers. These chemicals tend to be expensive to purchase and time-consuming to apply, and are also implicated in degradation of both human and environmental health. Therefore, 'non-chemical' alternatives are needed to replace these current disease control methods. Classical plant breeding has made significant advancements in developing fungal resistant genotypes by the introduction of 'resistance' genes into susceptible genotypes, thereby lowering the dependance on external chemical treatments. This process typically follows the cycle of germplasm collection and characterization, selection, and hybridization. Overall, this traditional method has been successful but has inherently long time horizons. In addition, classical breeding involves transfer of many other genes, either unrelated to fungal resistance or undesirable for other reasons, and these often take several years to purge. However, new biotechnological methods, in the form of genetic engineering (GE), have permitted the development of resistance genotypes faster with more precise gene combinations. Genetic modification via plant transformation (i.e. recombinant D N A technology) is a powerful tool for the improvement of crop plants. B y using recombinant D N A technology, genes can be cloned and introduced into the genome of a given plant. These genes can come from either distantly related plant species that cannot be hybridized with 8 the crop species, or from any other living organism. With respect to plant disease, transformation technology significantly broadens the number and diversity of genes that can be used for enhancing disease resistance in crop plants (Muehlbauer, 2003). There are thousands of different viruses, bacteria and fungi that cause disease in plants. Among these microbes, fungal pathogens are the most devastating and are responsible for the greatest crop loss (Bent, 2003). With the availability of recombinant D N A techniques, a number of new approaches for the control of plant diseases appear very promising. In a recent review by Punja (2004), various genes used in recombinant D N A research and their downstream effects are summarized. These include: 1) genes encoding products that are directly toxic for the pathogen, such as pathogenesis-related (PR) proteins, antimicrobial peptides, ribosome-inactivation proteins or phytoalexins; 2) gene products that neutralize a component of the pathogen, that is, inhibition of polygalacturonase, oxalic acid, or lipase; 3) gene products that enhance structural defenses in the plant, i.e. elevated levels of peroxidase and/or lignin; 4) products that release signals for defense mechanisms, such as specific elicitors, hydrogen peroxide, salicylic acid or ethylene; and 5) expression of resistance gene (R) products involved in R /Avr interactions and the hypersensitive response (HR). With the development of an efficient transgenic technology for plants, the search for genes that can increase plant resistance to pathogens can become more dynamic. Surveying the genes that have been tested for this purpose, I chose to focus my research on a specific family of proteins (PR-5). A more in-depth description of this family of proteins is provided below. 9 1.2.4 PR- Proteins Pathogenesis-related proteins (PR) were first detected in the early 1970s in tobacco leaves that exhibited a hypersensitive response to inoculation by tobacco mosaic virus ( T M V ) . The PR nomenclature was first proposed by Antoniw et al. (1980) and continues to be the term of choice. The expression of this class of proteins is often induced systemically in many plants as a response to stress, pathogenic attack or wounding (Stintzi el al., 1993). However, PR proteins are also expressed in some fruits in absence of these factors, and they may be involved in fruit development (Clendennen, 1997). PR-proteins have been classified into 14 families (Table 1.2) (Van Loon & Van Strien, 1999). This classification was based on serological relationships, similar molecular weights, and/or shared highly conserved amino acid sequences (Linthorst, 1991; Cutt & Klessig, 1992). In addition to classical PR proteins, new proteins possessing anti-pathogenic action continue to be described and characterized at a rapid pace, thereby increasing the range of resistance mechanisms available for use. Among these PR-proteins, the most widely used for enhancing fungal resistance in plants have been chitinases (PR-3, PR-4, PR-8, and PR-11) and glucanases (PR-2). Another family that exhibits antifungal activity is the PR-5 family of proteins, known as thaumatin-like proteins (TLP). They are so named because of their sequence similarity to thaumatin, a sweet-tasting protein from Thaumatococcus daniellii Benth. (Corneliesen et al, 1986). T L P s have been isolated and characterized from several plant species including corn, (Zea mays L.) (Malehorn et al, 1994), grape (Vitis vimfera L.) (Monteiro el al, 10 2003), and rice (Oryza saliva L.) . From rice, two groups of T L P s have been isolated, and consist of proteins with higher molecular mass (i.e. 22-26 kDa) (Velazhahan et al., 1998), and proteins with lower molecular mass (i.e. -17 kDa) (Reimmann and Dudler, 1993). Functionally, higher concentrations of TLPs actively lyse fungal membranes, while at lower concentrations, they increase membrane permeability, leading to cell leakage and increased uptake of other antifungal compounds (Jayaraj et al., 2004). Although TLPs do display broad antifungal activity (Punja, 2004), their efficacy against particular pathogens varies with the genus of the fungal pathogen itself (Vigers el al, 1991; Abad et al., 1996). One of the rice high M W TLPs (TLP-D34) has been used to enhance fungal resistance of plants via recombinant D N A technology. In transformed carrot (Daucus carota L . ) , expression of the T L P construct reduced both the rate and final disease incidence caused by Bolrytis cinerea Pers. and Sclerotinia sclerotiorum (Lib.) de Bary (Chen and Punja, 2002). In addition, its over-expression in rice reduced lesion development due to Rhizoctonia solani Kuhn (Datta et al, 1999) while in wheat, T L P delayed the development of Fusarium graminearitm Schw. (Chen et al, 1999) and in bentgrass (Agroslis palustris Huds.) it improved resistance to Sclerotinia homoeocarpa F.T. Bennett (Fu et al, 2005). Although the antifungal mechanisms of rice T L P - D 3 4 remain unclear at the molecular level, this gene's product presumably functions in synergy with either other plant antimicrobial product(s) or by interacting with pathogen factors, thereby preventing fungal growth and disease progression (Fu el al, 2005). Sequence analysis of TLP-D34 predicts a protein of 232 amino acids (aa) with a N-terminal signal peptide of 24 aa. In theory, the mature TLP-D34 represents a 21.7-k D A protein with a calculated isoelectric point of 7.5 (Fu el al, 2005). Most PR-5 11 proteins analyzed to date consist of three domains (I, II and III). The common structural feature that is associated with the antifungal isoforms of PR-5 proteins is an acidic cleft, formed by domains I and II, that may participate in the recognition of target molecule(s) (Koiwa et al, 1999; Batalia et al, 1996). Batalia'ef al., (1996) proposed that the PR-5 proteins may interact with other components that facilitate their uptake, or they may modulate osmoregulators, causing a rapid influx of water and leading to hyphal rupture. Based on the available literature and resources, the TLP-D34 gene from rice was chosen for insertion into a construct suitable for use in plant transformation experiments. i 12 Table 1.2 Summary of pathogen-related protein families (taken from Van Loon, L.C. and Van Strien, E.A. 1999). Family Type member Properties Reference PR-1 Tobacco PR-1 a Unknown Antoniw et al., 1980 PR-2 Tobacco PR-2 b-1,3-glucanase Antoniw et al., 1980 PR-3 Tobacco P, Q chitinase type 1,11, Van Loon, 1982 IV,V,VI,VII PR-4 Tobacco 'R' chitinase type 1,11 Van Loon, 1982 PR-5 Tobacco S thaumatin-like Van Loon, 1982 PR-6 Tomato Inhibitor I proteinase-inhibitor Green and Ryan, 1972 PR-7 Tomato P 69 endoproteinase Vera and Conejero, 1988 PR-8 Cucumber chitinase chitinase type III Metraux et al., 1988 PR-9 Tobacco 'lignin- peroxidase Lagrimini et al., 1987 forming peroxidase' PR-10 Parsley 'PR 1' 'ribonuclease-like' Somssich et al., 1986 PR-11 Tobacco 'class V chitinase, type 1 Melchers et al., 1994 chitinase PR-12 Radish Rs-AFP3 defensin Terras et al., 1992 PR-13 Arabidopsis THI2.1 thionin Epplee/tf/., 1995 PR-14 Barley LTP4 lipid-transfer protein Garcia-Olmedo et al., 1995 1.3 Literature Cited •1. Abad, L . , D 'Urzo , M . , L i u , D . , Narasimha, M . , Reuveni, M . , Zhu, J., N i u , X . , Singh, N . , Hasegawa, P., and Bressan, R. 1996. Antifungal activity of tobacco osmotin has specificity and involves plasma membrane permeabilization. Plant Science 118: 11-23. 2. Antoniw, J., Ritter, C , Pierpoint, W. , and Van Loon, L . 1980. Comparison of the pathogenesis-related proteins from plants of two cultivars of tobacco infected with T M V . J. General Virology 47: 79-87. 3. Batalia, M . , Monzingo, A . , Ernst, S., Roberts, W. , Robertus, J. 1996. The crystal structure of the antifungal protein Zeamatin, a member of the thaumatin-like, PR-5 protein family. Nat. Struct. B i o l . 3:19-23. 4. Bent, A . 2003. Crop diseases and strategies for their control. In Chrispeels, M . and Sadava, D . (Ed). Plant, Genes and Crop Biotechnology. Sudbury: American Society of Plant Biologist pp: 390-391. 5. Booth, C . 1971. The genus Fusarium. Commonwealth Mycological Institute, Kew, Surrey, England pp: 237. 6. Carrasco, R. 2002. Development of an in vitro screen for the root pathogen Fusarium spp. and Interspecific Exacum hybrid. Independent Studies Project, U B C pp. 1-17. 7. Clendennen, S. and May, G . 1997. Differential gene expression in ripening banana fruit. Plant Physiol. 115:463-469. 14 8. Chen, W. and Punja, Z . 2002. Transgenic herbicide-and disease tolerant carrot (Daucus carola L.) plants obtained through Agrobacterium-mediated transformation. Plant Cell Reports 20: 929-935 9. Chen, W., Chen, P., L i u , D. , Kynast, R., Friebe, B . , Velazhahan, R., Muthukrishnan, S., and G i l l , B . 1999. Development of wheat scab symptoms is delayed in transgenic wheat plants that constitutively express a rice thaumatin-like protein gene. Theor. Apply Genetics 99: 755-760. 10. Corneliesen, B . , Hooft van Huijduijnen, R., and Bo l , J. 1986. A tobacco mosaic virus-induced tobacco. protein is homologous to the sweet-tasting protein thaumatin. Nature 321: 531-532. 11. Cramer, L . 1981. Gentianaceae. In. Dassanayake, M . (Ed.). A revised handbook to the flora of Ceylon. V o l . 3. Amerind Publ. Co. New Delhi pp: 55-66 12. Cutt, J., and Klessig, D . 1992 Pathogenesis-related proteins. In Meins, F. and Boiler, T. (Ed.). Plant Gene Research: Genes involved in plant defense. New York: Springer-Verlag, pp.. 209-243 13. Datta, K . , Velazhahan, R., Oliva, N . , Ona, I., Mew, T., Khush, G . , Muthukrishnan, S., and Datta, S. 1999. Overexpression of cloned rice thaumatin-like protein (PR-5) in transgenic rice plants enhances environmental-friendly resistance to Rhizoctonia solani causing sheath blight disease. Theor. Apply Genetics 98: 1138-1145. 14. Epple, P., Apel , K . and Bohlmann, H . 1995. A n Arabidopsis thaliana thionin gene is inducible via a signal transduction pathway different from that for' pathogenesis-related proteins. Plant Physiol. 109: 813-820. 15. Fu, D . , Tisserat, N . , Xiao , Y . , Settle, D . , Muthukrishnan, S., Liang, G . 2005. Overexpression of rice TLPD34 enhances dollar-spot resistance in transgenic bentgrass. Plant Science 168: 671-680. 16. Garcia-Olmedo, F., Molina, A . , Segura, A . and Moreno, M . 1995. The defensive role of non-specific lipid-transfer proteins in plants. Trends Microbiol . 3: 72-74. 17. Green, T. and Ryan, C. 1972. Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175: 776-777. 18. Jayaraj, J., Anand, A . , and Muthukrishnan, S. 2004. Pathogenesis-related Proteins and their roles in Resistance to Fungal Pathogens. In Punja, Z . (Ed). Fungal Disease Resistance in Plants. Food Product Press, pp: 139-177. 19. Klackenberg, J. 1985. The genus Exacum (Gentianaceae). Opera botanica. A i O print Ltd. Copenhagen pp. 1-144 20. Koiwa , H . , Kato, H . , Nakatsu, T., Oda, J., Yamada, Y . , Sato, F. 1999. Crystal structure of tobacco PR-5d protein at 1.8 A resolution reveals a conserved acidic cleft structure in antifungal thaumatin-like proteins. J. M o l . B io l . 286: 1137-1145 21. Lagrimini, L . , Burkhart, W., Moyer, M . , and Rothstein, S. 1987. Molecular cloning of complementary D N A encoding the lignin-forming peroxidase from tobacco: molecular analysis and tissue-specific expression. Proc. Natl . Acad. Sci. U S A 84: 7542-7546. 22. Melchers, L . , Apotheker-de Groot, M . , Van der Knaap, J., Ponstein, A . , Sela-Buurlage, M . , B o l , J., Cornelissen, B . , Van den Elzen, P. and Linthorst, H . 1994. A new class of tobacco chitinases homologous to bacterial exo-chitinases displays antifungal activity. Plant J. 5: 469-480. 16 23. Linthorst, H . 1991. Pathogenesis-related proteins of plants. Crit. Rev. Plant Sci. 10:123-150. 24. Malehorn, D . , Borgmeyer, J., Smith, C , and Shah, D. 1994. Characterization and expression of antifungal zeamatin-like protein (Zip) gene from Zea mays. Plant Physiol. 106: 1471-1481. 25. Metraux, J., Streit, L . and Staub, T. 1988. A pathogenesis-related protein in cucumber is a chitinase. Physiol. M o l . Plant Pathol. 33: 1-9. 26. Muehlbauer, G . , Bushnell, W. 2003. Transgenic approaches to Fusarium Head Blight resistance. In Leonard, K . and Bushnell, W . (Ed). Fusarium Head Blight of Wheat and Barley. St. Paul: The American Phytopathological Society, pp: 318-362. 27. Monteiro, S., Barakat, M . , Picarra-Pereira, M . , Teixeira, A . , and Ferreira, R. 2003. Osmotin and thaumatin from grape: A putative general defense mechanism against pathogenic fungi. Phytopathology 93: 1505-1512. 28. Punja, Z . 2004. Genetic Engineering of Plants to Enhance Resistance to Fungal Pathogens. In Punja, Z . K . (Ed). Fungal Disease Resistance in Plants. New York, Food Product Press pp: 207-258. 29. Reimmann, C. and Dudler, R. 1993. c D N A cloning and sequence analysis of a pathogen-induced thaumatin-1 ike protein from rice (Oryza saliva). Plant Physiol. 101:1113-1114. 30. Riseman, A . and R. Craig. 1995. Interspecific Exacum Hybrids- Novel germplasm for the production of a new floricultural crop. XVIII th E U C A R P I A symposium Section Ornamentals, Tel A v i v , Israel. 5-9 Mar. 1995. Acta Hort. 420:132-134. 17 31. Riseman, A , Sumanasinghe, V . , Justice, D . and Craig, R. 2005. New name 'Styer Group' proposed for interspecific hybrids of Exacum species native to Sri Lanka. Hortscience 40: 1580-1583. 32. Somssich, I., Schmelzer, E. , Bollmann, J., and Hahlbrock, K . 1986. Rapid activation by fungal elicitor of genes encoding "pathogenesis-related" proteins in cultured parsley cells. Proc. Natl. Acad. Sci. U S A 83: 2427-2430. 33. Stintzi, A . , Heitz, T., Prasad, V . , Wiedemann-Merdinoglu, S., Kauffmann, S., Geoffroy, P., Legrand, M . , and Fritig, B . 1993. Plant "pathogenesis-related" proteins and their role in defense against pathogens. Biochem. 75:687-706. 34. Terras, F., Schoofs, H . , De Bolle, M . , Van Leuven, F., Rees, S., Vanderleyden, J., Cammue, B . , and Broekaert, W.F . 1992. Analysis of two novel classes of plant antifungal proteins from radish (Raphanus sativus L.) seeds. J. B io l . Chem. 267: 15301-15309. 35. Van Loon, L . 1982. Regulation of changes in proteins and enzymes associated with active defense against virus infection. In Wood R. (Ed.). Active Defense Mechanisms in Plants. Plenum Press, New York, U S A pp. 247-273. 36. Van Loon, L . and Van Strien, E . 1999. The families of pathogenesis-related proteins, their activities and comparative analysis of PR-1 type proteins. Physiol. M o l . Plant Pathol. 55: 85-97 37. Velazhahan, R., Cole, K . , Anuratha, C , and Muthukrishnan, S. 1998. Induction of thaumatin-like proteins (TLPs) in Rhizoclonia so/aw'-infected rice and characterization of two new c D N A clones. Physiologia Plantarum 102: 21-29 18 38. Vera, P. and Conejero, V . 1988. Pathogenesis-related proteins of tomato. P-69 is an alkaline endoproteinase. Plant Physiol. 87: 58-63 39. Vigers, A . , Roberts, W. , and Selitrennikoff, C. 1991. A new family of plant antifungal proteins. M o l . Plant-Microbe Interaction 4: 315-323. 40. Wolcan, S., Lor i , G . 1996. Basal rot of Eustoma grandiflorum in Argentina. Etiology. Invest. Agr. Prot. Veg. 11: 465. Chapter II 19 Organogenesis and Plant Regeneration from Leaf Explants of Exacum Styer Group 2.1 Introduction Exacum Styer Group (Gentianaceae Juss.) (Riseman et al. 2005), derived from directed hybridizations among five taxa native to Sri Lanka, has great potential as a potted, bedding or cut flower crop. Its large blue flowers contrast attractively with its' bright yellow anthers to give this gentian important horticultural value. However, difficulties in traditional vegetative propagation prompted the evaluation of alternative approaches to generating the required propagules for in vitro studies, genetic transformation, or commercial production. The development and introduction of a new horticultural crop is often facilitated by the development of various in vitro methods that include micropropagation, culture-virus indexing, induction of somaclonal variation, and/or genetic transformation. Efficient micropropagation techniques are generally required for these activities and are typically used to produce a large number of true-to-type plants in a relatively short period of time. However, a micropropagation system that includes a callus phase is undesirable due to the possible introduction of genetic variability in the resulting propagules. Therefore, to facilitate Exacum Styer Group's utility, research is required to identify factors that affect the growth and development of this germplasm in culture and that allows for the faithful production of specific genotypes. 20 The development of protocols for direct shoot organogenesis not only has application in micropropagation but also in the introduction of novel genes via Agrobacterium-mediated transformation. In this application, direct shoot organogenesis from individual transformed cells is an indispensable step in the recovery of transgenic plantlets. To date, there are no reports of direct organogenesis for Exacum. However, protocols for the establishment and multiplication of Exacum Styer Group genotypes have been reported (Riseman, 2004), and were used as the starting point for this study. The methodology followed in this research consisted of two experiments. The first experiment was designed to broadly evaluate different plant growth regulator (PGR) combinations for their ability to induce direct shoot organogenesis. Based on these preliminary results, a second follow up experiment was initiated to evaluate the most effective P G R combinations across a wider range of concentrations. Treatment selection for Experiment #1 was based on previous work that evaluated base medium compositions (i.e. M S and B5) supplemented with either 2ip (6-(y.y,dimethylallyamino purine), 2,4-D (2,4-dichlorophenoxyacetic acid), or B A (benzyladenine) plus N A A (naphthaleneacetic acid). The results showed that M S medium supplemented with B A (2 mg l"1) and N A A (0.2 mg l"1) induced greater quantities and quality of organogenetic shoots compared to the other treatments (unpublished data). In addition to evaluating an expanded concentration range of these PGRs , Experiment #1 also evaluated the efficacy of T D Z (thidiazuron). T D Z was included based on its strong ability to induce shoot organogenesis from leaf explants on various ornamental plants (Turk et al., 1994; Dubois and de Vries, 1995; L i n el al., 1997). The objective of this work was to develop efficient protocols for direct shoot organogenesis for Exacum Styer Group leaf explants for use in either micropropagation or genetic transformation research. 2.2 Materials and Methods 2.2.1 Experiment #1: 2.2.1.1 Plant material Four Exacum Styer Group genotypes were evaluated: 01-09-01, 01-37-61, 01-47-49, and 01-50-46. These genotypes were 10 t h-generation interspecific hybrids derived from directed hybridizations among the five taxa native to Sri Lanka (E. trinervium (L.) Druce ssp. macranthum (Arn.) Cramer, E. trinervium ssp. pallidum (Trimen) Cramer, E: trinervium ssp. ritigalensis (Will is) Cramer, E. trinervium ssp. trinervium, and E. pedunculatum L . (Cramer, 1981)). Leaf explants were aseptically excised from in vitro-grown plants. The stock plants were cultured on M S medium (Murashige and Skoog, 1962) supplemented with 0.01 mg l " 1 N A A and 2 mg l " 1 2iP, 30 g 1"' sucrose and 3 g 1"' agar and 1.1 g 1"' Phytogel. The pH was adjusted to 5.8 before autoclaving. 2.2.1.2 Treatments The base medium was full-strength M S salts supplemented with 30 g 1"' sucrose and 3 g f1 agar and 1.1 g 1"' of Phytogel (Sigma Chemical Co. , St. Louis, M O ) . The pH was adjusted to 5.8. P G R treatments included: 1) 0.01 mg f1 T D Z , 2) 0.05 mg f1 T D Z , 3) 0.1 mg l " 1 T D Z , and 4) 2.0 mg 1"' B A + 0.2 mg 1"' N A A . 22 Four to six young leaves per stock culture were harvested, cut into 1 c m 2 explants, and placed abaxial side down in petri plates containing 20 ml of medium. Four to eight explants were cultured per petri plate. The petri plates were sealed with parafilm, and incubated at 23-25° C with a 16 h photoperiod at 350 |JW cm" 2 supplied by cool white fluorescent lamps. Variables observed weekly included number of shoots per explant, percent explants forming callus, percent necrotic explants, and percent contamination. 2.2.1.3 Statistical analyses A randomized complete block design was employed with four replications. Due to contamination, some replications were incomplete. Analysis of variance and descriptive statistics was performed using S T A T A 6.0 (StataCorp L P . Austin, T X ) . 2.2.2 Experiment #2: 2.2.2.1 Plant material Four Exacum genotypes (01-09-01, 01-37-37, 01-37-61, and 01-42-3) were evaluated in this experiment. Leaf explants were excised from in vitro grown plantlets that were in cultured for six weeks on optimized elongation medium (Appendix I) supplemented with 4 g 1"' charcoal. A l l the other materials and protocols were as previously described in Experiment #1, except that leaf explants were cut using a cork borer #4 (7 mm diameter) instead of scissors. 23 2.2.2.2 Treatments A l l P G R supplements were evaluated on full-strength M S medium supplemented with 30 g l " 1 sucrose, 3 g l " 1 agar and 1.1 g l " 1 of Phytogel. The pH was again adjusted to 5.8. P G R combinations included B A (0, 0.1, 0.5, 1 or 2 mg l"1) and N A A (0, 0.01, 0.1 or 0.5 mg f ' ) . The same procedures and culture conditions were applied as in Experiment #1. Variables observed included number of shoots per explant, percent of responsive explants, and production of roots (scale 1 -5). 2.2.2.3 Statistical analyses For maximal efficiency, a factorial experimental design was employed. The descriptive statistics were calculated with Microsoft Excel (Microsoft Corp., Seattle, W A ) . Regression coefficients and Wald test were computed using S T A T A 6.0 (Stata Corp L P . Austin, T X ) . M A T H E M A T I C A 5.1 (Wolfram Research Inc. Champaign, IL) was used to produce the two and three dimensional plots. 2.3 Results 2.3.1 Experiment #1: Significant genotype and treatment effects (P<0.00\), as well as a significant genotype x treatment interaction (F<0.001), were identified for direct shoot induction. The P G R treatment that induced the greatest response across genotypes was 2 mg 1"' B A + 0.1 mg f ' N A A . Genotype 01-37-61 produced the highest number of shoots (9.5 shoots per explant) followed by genotypes 01-47-49 (6.15 shoots per explant), 01-09-01 (3.75 shoots per explant), and 01-50-46 (3.4 shoots per explant) (Figure 2.1). However, a 24 significant genotype by treatment interaction was present. A t the two higher concentrations of T D Z (i.e. 0.05 and 0.1 mg l"1), shoot formation was completely inhibited for genotype 01-47-49, while the remaining three genotypes produced a few shoots. In addition, T D Z caused explant necrosis and tissue degradation on 01-47-49, while this effect was not observed for the other genotypes. Regardless of genotype, P G R treatment effects were also significant. A t W K 2 of culture on the B A + N A A supplemented medium, explants became swollen and at W K 4, shoot initiation was first identified (Figure 2.2). Shoots arose from fresh cut borders on all explants without an intervening callus stage. In this treatment, genotypes 01-37-61 and 01-47-49 produced larger shoots more quickly as compared to the other two genotypes. However, healthy calli with small shoots did develop on the remaining genotypes by W K 4. N o TDZ-based treatments induced similar responses within this time frame ( W K 4) on any genotype. B y W K 5, the two higher T D Z concentrations induced shoot organogenesis. However, the number of shoots on this treatment was significantly lower as compared to the B A + N A A treatment. Regardless, with all T D Z treatments, genotypes 01-37-61 and 01-09-01 displayed superior performance compared to the other two genotypes. 14 12 4-* c 10 J5 a X 8 CD B 6 o o -C w 4 0.01 TDZ 0.05 TDZ 0.1 TDZ Treatments mg I'1 2 BA+0.2NAA Genotypes • 01-37-61 s 01-47-49 S 01-50^6 • 01-09-01 Figure 2.1 Effects of PGR supplements on shoot regeneration from leaf explants of Exacum. Error bars are standard error of the mean. Figure 2.2 Direct shoot organogenesis from leaf explants of Exacum genotype 01-09-01 on MS medium supplemented with 2mg 11 BA and 0.2 mg I"' NAA. (WF 10x). 26 2.3.2 Experiment #2: Significant differences (/><().01) were identified among genotypes for the overall number of shoots per explant (Figures 2.3a and 2.3b). Regardless of P G R treatment, genotypes 01-37-61 and 01-09-01 produced more shoots per explant than either 01-42-3 or 01-37-37 (Figure 2.3a). The rank order of genotypes for shoots per explant based on overall means across treatments were 01-09-01 (10.2 shoots/explarit), 01-37-61 (6.6 shoots/explant), 01-37-37 (5.1 shoots/explant), and 01-42-3 (2.7 shoots/explant) with genotype 01-09-01 producing four times the number of shoots of genotype 01-42-3 (Figure 2.3a). In addition to genotype 01-09-01 producing greater number of shoots per explant, it also produced shoots more consistently across treatments than any other genotype (Table 2.1). The production of shoots was also evaluated for each P G R combination among the four genotypes. Genotypes 01-37-61 and 01-09-01 exhibited the best performance on M S medium supplemented with 1 mg l " 1 B A + 0.1 mg 1"' N A A , producing 30.7 and 25.3 shoots, respectively (Table 2.1). In addition, genotypes 01-42-3 and 01-37-37, which in general had low shoot organogenesis rates, performed better on medium supplemented with 2 mg l" 1 B A + 0.5 mg l " 1 N A A with shoot production of 16.3 and 21.7, respectively (Table 2.1). 27 a 15 c 10 -01-37-61 01-09-01 01-42-3 Exacum genotypes b Comparison of overall means (Wald test) 01 -37-61=01-09-1 ** 01 -37-61= 01-42-3 ** 01 -37-61=01-37-37 ** 01 -42-3=01-37-37 ** ' ** Significant difference at 1%. Figure 2.3 Overall mean comparisons (i.e. across media) of shoot regeneration of four Exacum genotypes, a) means of shoot production per explant for each genotype; and b) Wald test for evaluation of statistical differences among means of genotypes. Table 2.1 Effect of NAA and BA concentrations on shoot production from leaf explants of four Exacum genotypes after 6 weeks of culture Genotype 01-37-61 Genotype 01-09-01 Genotype 01-42-3 Genotype 01-37-37 PGR Response Mean # shoots/ Response Mean # shoots/ Response Mean # shoots/ Response Mean # shoots/ mgr1 (%) explant (%) explant (%) explant (%) explant NAA BA (Mean ± S.D.) (Mean ± S.D.) (Mean ± S.D.) (Mean ± S.D.) 0 0 0 0 19 1 ±0 0 0 0 0 0.1 60 1.8±0.6 52 0.7 ±0.7 33 0 40 0 0.5 95 3.2 + 0.6 76 9.3 ±2.5 46 0.2 ±0.3 60 1.1 ±0.9 1 80 5.9 ±2.5 61 12.8 ± 5.4 80 1 ± 1.4 60 1.9 ±2.4 2 100 5.5 ± 1.4 67 21.1 ± 3.8 50 0 20 1.5 ± 2.1 0.01 0 13 0 40 1.3 ±0.9 80 0.6 ±0.5 86 2.1 ± 0.4 0.1 81 2.4 ±0.7 65 4.9 ± 1.6 93 1.9 ± 0.1 93 4.9 ± 0.7 0.5 95 7 ±5.1 100 17.3 ±0.9 87 4 ± 1.8 93 9.1 ± 1.8 1 95 9.9 ±2.5 100 13.4 ±2.8 80 3.7 ± 1.8 100 4.2 ±2.3 2 100 16.6 ± 5.7 83 16.8 ±7.4 80 0.2 ±0.1 60 0 0.1 0 90 0 80 0.7 ±0.6 100 2.2 ±0.9 73 2.8 ±2 0.1 80 2.8 ±0.7 65 7.1 ±3.3 100 4.3 ± 1.4 86 11 ±2.2 0.5 79 6.1 ±6.1 85 12.8 ±5.6 100 5 + 1.4 93 11.8 ± 2.8 1 100 30.7 ± 1.3 100 25.3 ±2.1 93 6± 1.5 73 5.8 ±4.7 2 89 25.5 ± 1.7 100 22.8 ±4.5 60 0.2 ±0.2 0 0 0.5 0 75 0 39 0.7 ±0.6 100 1.5 ± 0.1 100 3.9± 1.5 0.1 100 0.1 ±0.1 65 1.2 ± 1.1 100 3.6 ± 2 100 7.8 ±2.8 0.5 100 2 ± 1.1 77 3.5 ± 0 100 7.4 ± 1.5 100 8.5 ±2.7 1 95 11.6 ± 4.9 87 8 ± 1.1 80 7.7 ±4.5 100 14.7 ±5 2 100 16.3 ± 0.7 100 21.7±4.2 100 16.3 ± 0.7 100 21.7±4.2 to 29 Due to a significance interaction between PGRs , A N O V A was not appropriate to analyze the individual effects of B A and N A A . Therefore, regression analyses were performed to explain the individual effects of these PGRs and their interaction. Genotype-specific regression analyses were performed that compared various P G R combinations to the control treatment (i.e. 0 mg 1"' N A A and B A ) . For example, results for genotype 01-09-1 are presented (Table 2.2) and showed that the regression coefficient increased as N A A concentration increased from 0.01 to 0.1 mg \ ' \ indicating a significant increase in shoot production between these concentrations. However, this response decreased significantly at the 0.5 mg l " 1 N A A level. For B A supplements, increased concentrations induced greater shoot production throughout the range evaluated. When main effects were found to be significant, an additional test (Wald test) was performed to identify the level of significance among all P G R concentrations evaluated. The results of this test showed that there were significant differences ( P O . O l ) between the N A A and B A concentrations evaluated (Table 2.2). These differences expanded the inferences possible from A N O V A analyses and identified which concentrations of PGRs (considering the interaction) were the most appropriate within the range tested. 30 Table 2.2 Regression coefficients for BA and NAA concentration and their interaction on shoot regeneration of Exacum genotype 01-09-01. Variable Regression coefficients Std. Err. Wald test N A A x BA 3.53 3.73 N A A 0.01 1.19 1.57 N A A 0.01=NAA 0.1* N A A 0.1 4.53** 1.59 N A A 0.1 = N A A 0.5 ** N A A 0.5 -2.61 2.10 N A A 0.5= N A A 0.01 . BA0.1 2.55 1.75 B A 0.1= BA 0.5 ** BA 0.5 9.37** 1.86 BA 0.5= BA 1 * BA 1 13.77** 1.83 BA 1= BA 2 * BA 2 18.66** 2.09 Adj. R-square 0.76 ** Significant difference at 1%. * Significant difference at 5%. 31 Table 2.3 Regression coefficients for the effects of BA, NAA and their interaction on shoot regeneration of Exacum; a) quadratic regression and b) linear regression. a Variable Regression coefficient Std. Err. NAA 49.25** 9.43 NAAxNAA -100.9** 17.87 BA 13.11** 1.52 BAxBA -4.11 ** 0.72 NAA x BA 6.40** 2.05 G 01-37-61 0.04 0.77 G 01-9-1 3.48** 0.82 G 01-42-3 -4.48** 0.85 G 01-37-37 -2.21** 0.84 ' Adj. R-square 0.77 b Variable Regression coefficient Std. Err. NAA -1.95 2.53 BA 4.83** 0.56 NAA x BA 5.71* 2.31 G 01-37-61 2.94** 0.76 G 01-9-1 6.41** 0.83 G 01 -42-3 -1.32 0.85 G 01-37-37 0.89 0.84 Adj. R-square 0.70 ** Significant difference at 1%. * Significant difference at 5%. 32 In addition, a regression analysis that included all genotypes was performed for all independent variables against the dependent variable 'number of shoots'. This regression identified significant effects (/>>0.01) for N A A , B A , and their interaction ( N A A x B A ) (Table 2.3a). This regression was modeled without including a constant variable in order to compare all genotypes against the overall shoot production mean. This was done because genotypes were coded into the database as categorical variables and by including an intercept component, this would have removed one genotype from the analysis and made comparison based on this genotype. The resulting genotype coefficients for the 'no constant' regression showed significant effects for the three genotypes 01-09-1, 01-42-3, and 01-37-37, but not for genotype 01-37-61 (Table 2.3a). This last genotype was not found significant because its mean value was equal to the overall mean and therefore, was used as a benchmark comparison point. Additionally, when the regression analysis compared the quadratic variables to the linear variables, the interaction effects were more fully explained (i.e. adjusted R-squared increase by 7%), showing that the interaction between PGRs was not linear but rather quadratic in nature (Tables 2.3a and 2.3b). To graphically evaluate the N A A x B A interaction, a 3-dimensional plot was prepared using the coefficients obtained by regressing 'shoots per explant' against B A , N A A , B A x N A A , quadratic B A , quadratic N A A , and genotype. This plot displays both the singular effects of each PGR, while also displaying the effects from their interaction across all genotypes. The variation observed for shoots per explant explained by this model was relatively low when the effects of the PGRs were evaluated singularly (Figures 2.4 a and b). However, when the N A A x B A interaction was included, the variation was more fully explained (Figure 2.5). Based on the predicted values generated by this analysis, an optimal P G R supplement would range between 0.2-0.3 mg l " 1 for N A A and between 1.5-2 mg 1"' for B A (Figure 2.5). Regardless of genotype, the production of new organs was dependent on both auxin and cytokinin concentration. A l l genotypes initiated shoots after W K 3 of culture and continued to grow normally; the shoots typically arose from cut borders. Through W K 6 on PGR-free medium, most genotypes produced no shoots, roots, or callus tissue (Figure 2.7a). In addition, when low concentrations of B A (0.1 or 0.5 mg l"1) were added to the medium together with 0.5 mg 1"' N A A , the majority of new growth was roots while only a few shoots were produced (Figure 2.6). This was especially noticeable on the medium containing 0.5 mg 1"' N A A and 0.1 mg 1"' B A (Figure 2.7b). Furthermore, all B A concentrations used in conjunction with 0.5 mg 1"' N A A produced roots alone (Figure 2.6) or with shoots (Figure 2.7c). Callus was induced on medium supplemented with 0.5 mg N A A l " 1 combined with both 0.5 and 1 mg 1"' B A . Finally, the presence o f low concentrations of N A A (i.e. 0.01 mg 1"') in combination with any concentration of B A increased shoot production compared to when B A was used alone. 34 Figure 2.4 Graphical representation of the effects of BA and NAA across genotypes on the expected values of shoot regeneration from leaf explants of Exacum. a) NAA effect, b) BA effect. NAA mg 1 Figure 2.5 Three dimensional representations of the overall effects of BA, NAA and their interaction on the mean of shoot regeneration from leaf explants of Exacum. 36 o 4 co o w 3 > •-75 3 1 T _ T T o < CD i^ j f j l ^ in y- CM o i— m T— CM O in T CM o T m ^ c i o < < o d < < < o d < < < d d < < < CD m CO < < oo DO co < < CD CO CO < < 00 CO m m CO 00 CO CO NAAO NAA0.01 NAA0.1 NAA0.5 Treatments mg I"1 Figure 2.6 Overall mean effects of different BA+NAA combinations on root organogenesis of Exacum across genotypes. Root production was measured on a scale 0 to 5, where 0 equaled no roots and 5 equaled many roots. 37 Figure 2.7 Effects of different BA+NAA combinations on Exacum organogenesis, a) Hormone free medium, no organogenesis; b) 0.5 mg I"1 NAA+ 0.1 mg I"1 BA, root induction; c) 0.5 mg 11 NAA+ 1 mg I* BA, callus, root and shoot induction; and d) 0.01 mg 11 NAA and 0.5 mg I"' BA, callus and shoot induction. 38 2.4 Discussion The results presented demonstrate that leaf explants of Exacum Styer Group offer great potential as source tissue for the induction of either direct shoot organogenesis or callus production. However, tissue response was highly dependant on P G R concentration, P G R combination, and genotype. The urea-based cytokinin T D Z has been reported to improve shoot regeneration in several plant species as compared to adenine-based cytokinins (Huetteman and Preece, 1993; Turk et al, 1994; Dubois and de Vries, 1995). Based on this information, T D Z was included for evaluation within Experiment 1. Unfortunately, T D Z did not produce the same effects on Exacum as those reported on other species, and to the contrary, induced explant necrosis without shoot initiation, under these specific conditions. On Exacum, the use of an adenine-based cytokinin ( B A ) was far more effective than any TDZ-based treatment. Similar conclusions were reported on some other herbaceous plants, including carnation (Dianthus caryophyllus L.) (Messeguer et al., 1993), gerbera (Gerbera jamesonii Bolus) (Nongmanee and Kanchanapoom, 1995), gloxinia (Gloxinia perennis (L.) Druce) (Wuttisit and Kanchanapoom, 1995), and African violet (Saintpaulia ionantha Wendland) (Sunpui and Kanchanapoom, 2002). Following broad evaluation of various P G R supplements on tissue performance, refinement and optimization for individual genotypes is required for maximum production. Both B A and N A A were shown to be the most suitable P G R s for organogenesis on Exacum. Further refinement of concentrations has shown that the generation of adventitious shoots was highly dependent on the ratio between cytokinin and auxin concentrations. Reports of auxin and cytokinins combinations supporting 39 organogenic differentiation have been well documented for several species (Lisowska and Wysonkinska, 2000; Pereira et al, 2000). Those studies are in agreement with the results presented here, where it was observed that higher concentrations of B A combined with lower concentrations of N A A produced greater numbers and higher quality shoots (best combination was 1 mg 1"' B A + 0.1 mg 1"' N A A ) . In addition, significant differences among genotypes were found, with genotypes 01-09-01 and 01-37-61 displaying the best performance and representing good choices for use in future research. This is the first successful report of direct organogenesis in Exacum with protocols for direct organogenesis from leaf explants established for four genotypes. These genotypes were found to be responsive under these specific conditions and are now positioned for use in micropropagation or genetic transformation system development. In view of the significant genotypic effects observed, the present protocols should be further optimized in order to obtain the maximum response possible across all genotypes. Overall, however, my results demonstrate that leaves of Exacum Styer Group have great organogenic potential and that their ability to support shoot and root regeneration are easily manipulated by P G R supplements. 40 2.5 Literature Cited 1. Cramer, L . 1981. Gentianaceae. In: M . D . Dassanayake, editor. A revised handbook to the flora of Ceylon, Balkema, Rotterdam vol . 3 pp. 55-78 2. Dubois, L . and de Vries, D . 1995. Preliminary report on the direct regeneration of adventitious buds on leaf explants of in vivo grown glasshouse rose cultivars. Gartenbauwissenschaft 68: 249-253. 3. Huetteman, C. and Preece, J. 1993. Thidiazuron: a potent cytokinins for woody plant tissue culture. Plant Cell Tiss, Org. Cult. 33: 105-119. 4. L in , S., De Jeu, M . , and Jacobsen, F. 1997. Direct shoot regeneration from excised leaf explants of in vitro grown seedling of Alstroemeria L . Plant Cel l Rep. 16:770-774. 5. Lisowska K and Wysokinska H . 2000. In vitro propagation of Catalpa ovata G . Don. Plant Cell Tiss. Org Cult. 60: 171-176. 6. Messeguer, J., Arconada, M . , and Mele, E . 1993. Adventitious shoot regeneration in carnation (Dianthus caryophyllus L.) . Scientia Hort. 54: 153-163. 7. Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant 15: 473-497. 8. Nongmanee, W., and Kanchanapoom, K . 1995. Plantlet production from young petiole and immature inflorescence of Gerbera jamesonii cultured in vitro. Songklanakarin Journal Sci. Tech. 17: 137-142. 9. Pereira A . , Bertoni, B . , Appezzato-da-Gloria, B . , Araujo, A . , Januario, A . , Loureco M . , and Franca S. 2000. Micropropagation of Pathomorphe umbellate via direct organogenesis from leaf explants. Plant Cell Tiss. Org. Cult. 60: 47-53. 10. Riseman, A . and Chennareddy, S. 2004. Genotypic Variation in the Micropropagation of Sri Lankan Exacum hybrids. J. Amer. Soc. Hort. Sci . 129: 698-703. 11. Riseman, A . , Sumanasinghe, V . , Justice, D. and Craig, R. 2005. New name 'Styer Group' proposed for interspecific hybrids of Exacum species native to Sri Lanka. Hortscience 40: 1580-1583. 12. Sunpui, W. and Kanchanapoom, K . 2002. Plant regeneration from petiole and leaf of African violet (Saintpaulia ionantha Wendl.) cultured in vitro. Songklanakarin Journal Sci . Tech. 24: 357-364. 13. Turk, B . , Swartz, H . , Zimmerman, R. 1994. Adventitious shoot regeneration from in v/7/'o-cultured leaves of Rubus genotypes. Plant Cell Tiss. Org. Cult. 38: 11-17. 14. Wutisit, M . and Kanchanapoom, K . 1995. Tissue culture propagation of Gloxinia. Suranaree Journal Sci. Tech. 3:63-67. 42 Chapter III Agrobacterium-mediated transformation of Exacum Styer Group using a thaumatin-like protein (TLP) construct 3.1 Introduction The development and introduction of Exacum as a new horticultural crop can be facilitated by the development of several in v/'/ro-based technologies. These protocols include meristem culture (for production of virus-free materials), isolation of somaclonal variants (for access to genetic variability within existing cultivars), haploid production (for the development of completely homozygous progeny and pure lines) and genetic engineering (for the introduction of novel genes). Depending on the objective, one or more of these technologies may be appropriate. However, when the objective is to introduce a trait not coded for within the species' genome, genetic engineering is the only available technology. The introduction of one or more genes of known function to Exacum should allow for the expression of novel traits, while not altering the desirable traits present. This study focuses on establishing genetic transformation protocols for Exacum using a novel thaumatin-like protein (TLP) construct that, when functioning, should confer broad-based resistance to fungal pathogens. A n y transformation system for a particular crop requires an effective method of gene delivery. Currently, the most widely used method for transferring genes into dicotyledonous plants is by Agrobacterium-mcdiated transformation, a system that involves co-cultivating explants with a suspension of Agrobacterium cells. 43 Agrobaclerium-mediated transformation exploits the unique biology of Agrobaclerium tumefaciens, a bacterium that causes 'crown gall disease' in dicotyledonous plants (Tinland, 1996). During transformation, the bacterium transfers a portion of its Ti plasmid into the host genome. To enable the use of this system in transformation studies, the T i plasmid is typically modified by inserting the gene of interest plus a marker gene to be used during selection, and "disarming" the plasmid by deleting the genes involved in the production of bacterial food and plant hormones. Co-cultivation is typically done with small excised portions of plant tissue which, once they are transformed, can be regenerated into a whole plant. Successful transformation, with transgene insertion and expression, is measured as a function of the number of plants recovered from culture on a selection medium. Several factors affect the efficiency of the Agrohaclerium-medxated co-cultivation method, including the compatibility between the bacterial strain and plant species, exposure time and conditions for infection, and the composition of the selection medium. At the time this work was undertaken, there were no published protocols for Exacum transformation. However, a closely related plant (i.e. a member of a sister genus in the Gentianaceae family) that had been successfully transformed is Eustoma grandiflorum Griseb. (Hecht et al, 1997 and Semeria et al., 1996). The transformation methods published for this plant evaluated three Agrobaclerium strains: EHA-105 (Hecht et al. 1997), C-58, and A-281 (Semeria el al, 1996). Based on these reports, both EHA-105 and C-58 strains were included for evaluation of transformation efficiency. One goal of the Exacum breeding/biotechnology program is to increase resistance to fungal pathogens. Exacum hybrids are very susceptible to fungal root pathogens, . 44 especially Fusarium solani (Carrasco, 2002), a pathogen that causes stem and root rot. This specific pathogen is responsible for significant losses during production, a situation that may inhibit industrial acceptance. However, Exacum are susceptible to other fungal pathogens as well , including powdery mildew and Bolrylis. Therefore, the development of a broad-based fungal resistance system is highly desirable. In response to phytopathogen attack from viruses, bacteria or fungi, plants typically accumulate a large number of proteins called pathogenesis-related (PR) proteins. These proteins have been divided into several 'families' based on serological properties, molecular weight, and/or amino acid sequence. Among these families, the PR5 family of proteins has been well characterized in a wide range of both dicotyledonous and monocotyledonous plant species (Frendo et al., 1992; Hajgaard et al., 1991; Hygnh et al, 1992; Malehorn et al, 1994; Rebmann el al, 1991; Reimmann el al, 1993; Vigers el al., 1991; and Woloshuk et al, 1991). Within this family, a subclass of proteins identified as T L P s (thaumatin-like proteins) are of interest because of their plant protective properties. Although the biochemical function of thaumatin-like proteins has not yet been established, members of this group are reported to have antifungal activity against a broad spectrum of fungal pathogens (Abad el al, 1996; Hajgaard et al, 1191; Huynh et al, 1992; Fu et al, 2004; and Monteiro et al, 2003). In rice, two T L P c D N A s (C22 and D34) were identified following infection with Rhizoclonia solani (Velazhahan el al, 1998). The rice TLP-D34 transcript was detected 2-6 days after inoculation, with peak expression occurring on the fourth day. Therefore, the pathogen-inducible nature of this clone makes it a promising candidate for enhancing plant defense against fungal 45 pathogens, and I therefore chose to exploit those properties in the development of fungal-resistant transgenic Exacum. When designing the construct, specific attention must be paid to the promoter that wi l l drive expression of the gene of interest. Promoter sequences determine tissue specificity, timing, and the level of gene expression. In the many different plant species that have been transformed, three different promoters have been widely used: the maize ubiquitin gene promoter, the rice actin gene promoter, and the cauliflower mosaic virus ( C a M V ) 35S promoter. Each of these promoters induces constitutive expression of the gene of interest. The C a M V 3 5 S promoter is the most common promoter used for dicotyledonous plant transformation because it generally mediates high levels of expression for heterologous genes in most plant tissues and organs. Therefore, it is a very useful tool for research and commercial applications where a broad-based expression of the transgene is desired (Scholthof el al., 1996). In addition, when two 35S promoters are used in tandem and the leader sequence from the alfalfa mosaic virus RNA4 ( A M V leader) is added, the resulting promoter display greatly increased expression of transgenes relative to single copy versions (Datla el al., 1993: Kay el al., 1987). A T L P gene (rice thaumatin-like protein (TLP-D34)) introduced into a pCambia-bar-ubi-TLP transformation vector under the control of the maize ubiquitin promoter was previously used to successfully transform several plant species, and the resulting transformants displayed increased fungal resistance (Chen el al. 1999 and Chen and Punja, 2002). The maize-ubiquitin promoter used in this construct has not been widely used on dicotyledonous plants, so its efficacy for transforming Exacum was unpredictable. For that reason, I decided to develop a second transformation vector (pSM-3) that 46 contained the alfalfa mosaic virus ( A M V ) leader sequence with a 2 x tandem of the C a M V : 3 5 S promoter. Another very important aspect in the design and development of a suitable construct is the type of selectable marker system employed. Several types of selectable markers have been used in plant transformation, including antibiotic resistance provided by the hygromycin phosphotransferase (hpl) or neomycin phosphotransferase II (nptU) genes, as well as herbicide resistance provided by the (bar) gene coding for PPT acetyltransferase. The hygromycin phosphotrasferase gene originally derived from Escherichia coli, codes for the enzyme that detoxifies the aminocyclitol antibiotic hygromycin B (Pettinger el al., 1953) which inhibits protein synthesis by interfering with ribosomal translocation and aminoacyl-tRNA recognition (Cabanas, 1978; Gonzales, 1978; and Singh, 1979). The neomycin phosphotransferase II gene codes for aminoglycoside 3'-phosphotransferase, an enzyme that inactivates a range of aminoglycoside antibiotics including kanamycin, neomycin, geneticin, and paromomycin. In general, both hygromycin B and kanamycin resistance are two of the more common systems used for the selection of transgenic plants. However, newer systems such as PPT acetyltransferase are now regularly used for transgenic selection by conferring resistance to the phosphinothricin-based herbicides bialaphos and Basta® (Aventis-CropSciences). In preliminary experiments, kanamycin was not sufficiently lethal to Exacum to allow for transgenic identification (Appendix II). However, Exacum explants were found to be sensitive to hygromycin (see below) and therefore, hygromycin phosphotransferase was chosen as a suitable selection system. 47 The objectives of this study were: .1) to develop a novel T L P construct with a hygromycin resistance sequence and a double 35S promoter; 2) to develop an Agrobacterium-mediated transformation system for Exacum using the novel T L P construct; and 3) to characterize expression patterns of the T L P transgene in transgenic plants. 3.2 Materials and Methods 3.2.1 Plant Material The plant material used for transformation studies included leaf discs, calli and seedlings of Exacum Styer Group (Riseman et al, 2005). Leaf discs were cut with a #4 cork borer (7 mm diameter) from young leaves of five-week-old plantlets grown in the U B C Center for Plant Research tissue culture laboratory. Plants were grown at 24° C with a 16 h photoperiod at 350 uW cm" 2. Stock cultures of genotypes 01-37-61 and 01-42-3 were cultured on full-strength M S salts + 2 g 1"' charcoal, 30 g 1"' sucrose, 3 g l " 1 agar, and l . l g 1"' Phytogel supplemented with 4 mg 1"' 2ip and 0.04 mg l " 1 N A A . Cal l i from genotypes 02-174-9, 01-37-50, and 01-42-3 were sub-cultured every four weeks on B5 medium (Gamborg et al., 1976) supplemented with 1 mg 1"' 2,4-D. The seedlings (families 04-5, 04-6, 04-7, 04-10, 04-12, 04-13 and 04-19) were germinated on half-strength M S O medium and grown either with light or in the dark for 3 weeks prior to use. Tobacco (Nicotiana tabacum var. Xanthi) plants were used as a positive control. They were maintained in vitro in Magenta boxes containing M S O medium and subcultured every four weeks. Leaf discs, excluding the mid-vein, were cut as described above. 48 3.2.2 Transformation vectors and bacterial strains The T L P transformation vector was constructed by excising the T L P gene from a pCambia-bar-ubi-TLP vector (kindly provided by Dr. Zamir Punja, Simon Fraser University, B C ) and ligating it into the pSM-3 vector (supplied by Dr. Shawn Mansfield, U B C ) . A l l molecular manipulations were completed in the Biotechnology laboratory, Wood Science Department (Dr. Shawn Mansfield, Faculty of Forestry, U B C ) . Initial modifications included P C R site-directed mutagenesis to introduce two restriction sites (BamU] and EcoRI) to the border sequences of the T L P gene, so that the T L P gene could be excised as a BamH]-EcoR\ fragment from the original pCambia-bar-ubi-TLP transformation vector. The P C R primers used for the site-directed mutagenesis were designed by changing three nucleotides from the original T L P sequence to create the desired restriction sites as detailed below: Forward primer Original: 5' C C G A A C A A C A A A G C C G T C C T C T 3' T L P B A M : 5' C C G A A C A A C A A A G G G A T C C T C T 3' Reverse primer: Gil^ flti Original: 5' T A G T T T C C A T A G T T A C A C G C A T 3' T L P E C O : 5' T A G T T T C T T A A G T T A C A C G C A T 3' P C R reactions for site-directed mutagenesis used Pfu D N A polymerase (Fermentas Canada, Inc. ONT) , and the primers T L P B A M and T L P E C O (described above). The reaction contained 1 ul (42 ng) pCambia-bar-ubi T L P plasmid D N A , 1 ul (10 pmol) of each primer, 2 ul (2 m M ) dNTP, 0.5 pi (2.5 u/pl) Pfu polymerase, 2 ul (1 Ox) 49 Pfu buffer, and 12.5 u.1 water. The P C R reaction was run on a PTC100 P C R machine with a pre-programmed thermal cycling regime as follows: initial 4 min at 94° C, 35 cycles of 94° C for 30 sec, 57° C for 40 seconds, 72° C for 1 min, followed by 1 cycle of 72° C for 7 min. P C R products were visualized by ethidium bromide staining and were extracted from the gel matrix using the QIAquick gel extraction kit (Qiagen Inc, Mississauga, ON) . D N A digestion was performed using EcoR\ and BamU 1 restriction endonucleases (New England Biolabs Inc., Ipswich, M A ) . The reaction mix contained 37 pi amplified D N A , 2 pi BamhU (20,000 units/ml), 1 pi EcoRI (20,000 units/ml), 5 pi EcoRI buffer and 5 pi water. The enzymatic reaction was run for 1 h followed by product purification with the QIAquick P C R purification kit (Qiagen Inc, Mississauga, ON) . Once purified, the product was cloned into the T O P O T A cloning vector (Invitrogen Canada Inc., Burlington, ON) and used to transform E. coli (Topo One-Shot® Invitrogen Canada Inc., Burlington, ON) . E. coli competent cells were thawed on ice and mixed gently with 2 pi cloning reaction. The mixture was kept on ice for 30 min followed by a 30 sec heat-shock treatment at 42°C without shaking. The mixture was then returned to ice and 250 pi room temperature S.O.C. medium was added and incubated at 37° C for 1 h with shaking. Thirty micro liters bacterial culture was then spread on a pre-warmed L B agar plate containing 50 ug ml" 1 kanamycin and incubated overnight at 37° C . E. coli colonies that grew on kanamycin medium were selected and grown individually in liquid L B medium + kanamycin (50 mg 1"'). One colony was confirmed transformed by P C R using the following primers: T L P F W D : 5 ' C G G G G C A G A C G T G G A C C A T C A A C 3' and T L P R E V : 5 ' G C G C G T G C G G G C T T A T T C T T A G G T 3'. These primers amplified a 600 bp fragment of the T L P gene that was then visualized by gel electrophoresis. This colony was used to produce plasmid D N A for sequencing, and all subsequent sequence comparisons were made through B L A S T (National Center of Biotechnology Information, N C B I ) searches. Plasmid D N A from this colony was isolated using the QIAprep Spin Miniprep Ki t (Qiagen Inc, Mississauga, ON) with the T L P gene excised using both EcoR\ and Bam\\\ restriction endonucleases. Following gel electrophoresis, the expected 1 kb fragment was extracted using the QIAquick gel extraction kit and then ligated into the Bam\\\IEcoR\ pre-digested pSM-3 vector using the Rapid D N A Ligation Ki t (Roche, Laval , QC) . A ratio of 100 ng plasmid D N A to 40 ng insert D N A was used in the ligation reaction, based on manufacturer's recommendations. After ligation, the binary vector + insert were used to transform T O P O One-Shot E. coli competent cells. Confirmation of plasmid incorporation into E. coli was done by P C R amplification using the following primers: 2Ca F W D 5' G G C T A T C G T T C A A G A T G C C T C T G 3' and T L P R E V 5 ' G C G C G T G C G G G C T T A T T C T T A G G T 3'. D N A from a transformed colony was sent to the Nucleic A c i d Protein Service Unit ( U B C ) for sequencing using the following primer: TLP003 5 ' C G A C G A G G A G G A A G A G A G C 3'. The new construct (pSM-3 + T L P gene) was isolated from E. coli and then used to transform the Agrobacterium strains C-58 and EHA-105 , using the 'freeze-thaw' method for direct Agrobacterium transformation (Cellfor Inc., Vancouver, B C ) . Colonies that grew on the selection medium (i.e. 50 mg 1"' rifamycin and 50 mg 1"' kanamycin) were confirmed as transformants by P C R with the T L P F W D and T L P R E V primers. For 51 future use as negative controls, both Agrobacterium strains were also transformed with the empty pSM-3 vector. Bacterial stock cultures of Agrobacterium strains C-58 and EHA-105 carrying the pSM-3 T L P construct and strain L B A 4 4 0 4 carrying the original pCambia-bar-ubi T L P transformation vector were grown individually overnight at 28° C on a gyratory shaker (200 rpm) in L B medium with rifamycin (50 mg l"1) and kanamycin (50 mg 1"'). Prior to co-cultivation, 1 ml of each bacterial culture was subcultured in M S O medium + 100 u M acetosyringone and grown at 28° C on a gyratory shaker (200 rpm). Bacterial concentrations were determined using a Lambda 45 U V / V i s spectrometer (Perkin Elmer Inc., Wellesley, M A ) with the resulting OD6oo values for each culture as follows: C-58 with pSM-3 T L P (OD = 0.6), EHA-105 with pSM-3 T L P (1.0), EHA-105 with empty vector (0.9) and L B A 4 4 0 4 with pCambia-bar-ubi T L P (0.6). For co-cultivation, 30 ml of each bacterial culture was placed in 50 ml Falcon tubes with 25 leaf explants and incubated at 28° C in a gyratory shaker (60 rpm) in the light. 3.2.3 Hygromycin sensitivity To determine whether Exacum leaf tissue is sensitive to hygromycin selection, leaf explants of genotypes 01-09-1 and 01-37-61 were cultured on full-strength M S medium supplemented with 2 mg 1"' B A , 0.1 mg 1"' N A A , 30 g 1"' sucrose, 3 g 1"1 agar, 1.1 g 1"' Phytogel plus either 15, 20, 25 or 30 mg l " 1 hygromycin. Petri-plates containing 30 ml test medium were cultured with 7 explants/plate at 24° C with a 16 h photoperiod at 350 uW cm" . Three replicates of each treatment were evaluated for surviving explants after 6, 15 and 26 days of culture. 52 3.2.4 Agrobacterium-mediated transformations Exacum transformation protocols were based on procedures developed in the Biotechnology Laboratory Wood Science, Faculty of Forestry, U B C (Dr. Shawn Mansfield) that had proven effective on both poplar (P39-Populus alba x grandidentata, 7)7-Populus Iremula x alba) and tobacco (Nicotiana tabaccum var. Xanthi). The protocol used was as follows: 25 leaf discs (7 mm diameter) per genotype were co-cultivated with 30 ml bacterial culture in 50 ml Falcon tubes for either 35 minutes or 1 hour at 28° C in a gyratory shaker (60 rpm) with light. Following co-cultivation, the explants were blotted dry on sterile filter paper and placed abaxial side up in deep petri plates filled with 60 ml medium. The culture medium for Exacum contained full-strength M S medium, 30 g l" 1 sucrose, 3 g l" 1 agar, 1.1 g 1"' Phytogel, pH adjusted to 5.8 and supplemented with 2 mg l" 1 B A and 0.1 mg f ' N A A . For tobacco culture, the medium was also composed of full-strength M S medium, 30 g 1"' sucrose, 3 g 1"' agar, 1.1 g l " 1 Phytogel, pH adjusted to 5.8 and supplemented with 0.02 mg 1"' o f N A A , B A and T D Z . The petri plates were cultured in the dark for two days at room temperature. On the third day, residual Agrobacterium was killed,by transferring the leaf discs to species-specific medium (for Exacum: full-strength M S salts, 30 g 1"' sucrose, 3 g f 1 agar, 1.1 g l " 1 Phytogel, pH adjusted to 5.8 supplemented with 2 mg 1"'BA + 0.1 mg 1"' N A A ; for tobacco: full-strength M S salts, 30 g 1"' sucrose, 3 g 1"' agar, 1.1 g 1"' Phytogel, pH adjusted to 5.8 supplemented with 0.02 mg 1"1 N A A , B A and T D Z ) containing 250 mg 1"1 cefotoxine and 500 mg l " 1 carbenicillin. A l l plates were kept in the dark for an additional two days. Following this period, explants were transferred to selection medium (i.e. same media as described above for both species plus 25 mg 1"' hygromycin). After 7 53 weeks, explants were transferred to fresh medium with the same composition as described above. Various modifications to this protocol were evaluated for improved transformation efficiency; these included: 1) leaf discs were pre-cultured on regeneration medium (MS medium, 30 g l " 1 sucrose, 3 g l " 1 agar, 1.1 g l " 1 Phytogel, pH adjusted to 5.8 supplemented with 2 mg 1"' B A + 0.1 mg 1"' N A A ) for two days prior to co-cultivation with Agrobacterium; 2) following co-cultivation, leaf discs were cultured on M S medium supplemented with 1 mg I"1 2,4-D for two days in the dark and then transferred to regeneration medium supplemented with carbenicillin and cefotoxine as described above. Wounded seedlings (i.e. abraded using sterile sand) were also used as explants and co-cultivated as described for leaf explants. In addition, a modified Agrobacterium-mediated transformation protocol for conifers (Klimaszewska et al., 2005) was evaluated on Exacum callus. The protocol included incubation of approximately 100 mg callus tissue in an Agrobacterium suspension containing liquid M S (culture density ODfioonm 0.7), and 50 u M acetosyringone. The tissue was co-cultivated on a gyratory shaker set at 100 rpm for one hour at 28° C. The excess bacterial culture was drained off using a Buchner funnel lined with sterile filter paper. The filter paper with callus tissue was then placed on petri plates containing shoot induction medium (full-strength M S salts, 30 g 1"' sucrose, 3 g 1"' agar, 1.1 g 1"' Phytogel, pH adjusted to 5.8 supplemented with 2 mg 1"'BA + 0.1 mg l " 1 N A A ) for two days in the dark. After this period, tissue was transferred from the filter paper and rinsed with M S O liquid medium containing carbenicillin (500 mg f ' ) and cefotoxine (250 mg l"1) to inhibit further Agrobacterium growth. The rinse medium was drained through a Buchner funnel with the filter paper and the calli placed on shoot 54 induction medium with carbenicillin (500 mg 1"') and cefotoxine (250 mg 1"') for two days in the dark. Finally, calli were transferred to selection medium containing carbenicillin (500 mg 1"'), cefotoxine (250 mg 1"'), and hygromycin (25 mg l"1). Subcultures of each clump of callus were completed every 3 weeks to maintain selection pressure by the antibiotic. 3.2.5 Molecular analysis 3.2.5.1 Genomic DNA extraction The C T A B D N A (Sigma-Aldrich Co., St. Louis, M O ) extraction method was used for both Exacum and tobacco tissue. Using sterile technique, 100 mg tissue cut from young leaves was placed into sterile 1.5 ml microfuge tubes. The tissue was frozen with liquid nitrogen, ground into a powder and mixed with 1 ml extraction buffer (2% w/v C T A B , 1.4 M N a C l , 20 m M E D T A , 100 m M Tr i s -HCl , 1% P V P , and 0.2%, v/v |3-mercaptoethanol). The tubes were then incubated for 60 min at 65° C and centrifuged for 10 min at 13000 rpm. The supernatant was transferred to a sterile 2 ml microfuge tube and 1 volume of phenol / chloroform / isoamyl alcohol (24:1:1) was added to the solution. The tubes were inverted a few times, followed by centrifugation for 10 min at 13000 rpm. After this, the supernatant was transferred to a new microfuge tube and the D N A was precipitated by addition of !/2 volume of isopropanol. Tubes were lightly mixed a few times and then centrifuged for 10 min at 13000 rpm. The resulting D N A pellet was washed with 70% ethanol, air dried for 30-60 min and re-suspended in nanopure water (NANOPure Infinity 1 M ultrapure water systems, Dubuque, IA). The D N A concentration 55 was measured with a GeneQuant pro machine (BioChrom Labs. Inc., Terre Haute, IN) and the solution was then stored at -20° C until use. 3.2.5.2 PCR analysis of putative transformants A l l tissues (Exacum and tobacco) that survived the antibiotic selection were evaluated for the presence of the transgene. Transgene detection was carried out by P C R using the T L P F W D and T L P R E V primers (described previously) to amplify the 600 bp T L P fragment. In addition, hygromycin primers H Y G F W D 5' G G C A A A C T G T G A T G G A C G A C 3' and H Y G R E V 5' G C T G G G G C G T C G G T T T C 3'were used to amplify a 518 bp fragment of the hygromycin phosphotransferase gene (hpl). The P C R mixture (total volume 20 pi) contained 1 ul (200-400 ng) genomic D N A , 1 pi (10 pmol) of each primer, 2 pi (2 m M ) dNTP, 0.5 pi (5000 units/ml) of Taq polymerase (New England Biolabs, Ipswich, M A ) , 2 ul ( l O x ) P C R buffer, and 12.5 pi water. The P C R reaction was run on a P T C I 0 0 P C R machine with a pre-programmed thermal cycling regime as follows: initial 4 min at 94° C, 35 cycles of 94° C for 30 seconds, 57° C for 40 seconds, 72° C for I min, followed by 1 cycle of 72° C for 7 minutes. The P C R products were then visualized in 1 % agarose gels. Electrophoresis was carried out using the /Mupid system (Helixx Technology Inc., Scarborough, ON) for 25 min at 100V. The ethidium bromide-stained D N A bands were visualized using the Alphalmager™ 2200 Imaging system (Alpha Innotech Co., San Leandro, C A ) . 56 3.2.5.3 RNA isolation For Exacum and tobacco plants, the T R I Z O L R N A isolation kit (Invitrogen Canada Inc., Burlington, ON) was used. Leaf tissue (200 mg) was cut under sterile conditions and placed in a microfuge tube in liquid nitrogen. The tissue was then ground into a powder in the presence of liquid nitrogen and 1 ml T R I Z O L reagent. After this, the mixture was vortexed and incubated at room temperature for 5 minutes followed by the addition of 0.2 ml chloroform; this was then vigorously vortexed for 15 sec. The tubes were then incubated at room temperature for 3 min and centrifuged for 15 minutes at 12000 g at 4° C. The resulting supernatant was transferred to a fresh tube and the R N A precipitated with 0.5 ml isopropanol by incubation at room temperature for 10 min, followed by centrifugation as described above. The resulting pellets were washed with 75% ethanol, air dried, re-suspended in RNase-free water, and incubated for 10 min at 60° C. The R N A concentration was then calculated with the GeneQuant. DNase I D I G E S T kit (Ambion Inc, Austin, T X ) was used to eliminate contamination by D N A as follows: 5 pi Turbo 10x buffer were added to 10 pg R N A in 44 pi DEPC-treated (diethylpyrocarbonate) water. Turbo-DNase (2U/pl) (1 pi) was added to the microfuge tube containing the mixture and incubated for 30 min at 37° C. After incubation, the enzymatic reaction was stopped by adding 5 pi inactivation solution and centrifuged for 2 min at 10000 g at 4 ° C . The supernatant was then transferred to a fresh 1.5 ml tube, and then quantified using the GeneQuant and stored at -80° C until use. c D N A was generated from R N A using the Superscript II RT kit (Invitrogen Canada Inc. Burlington, ON) . In a P C R tube, DNA-free R N A (1 pg) was brought up to 10 pi with DEPC-treated water with 1 pi oligo (dT18 500 pg ml"1) solution and 1 pi 57 dNTP (10 m M ) added and then heated for 5 min at 65° C. Then, 4 ul of 5x First-Strand buffer and 2 pi of 0.1 M D T T were added to the P C R tube and incubated for 2 min at 42° C. Additionally, 1 ul (200 units) Superscript II R T was added, mixed by pipetting, and incubated for 50 min at 42° C. The enzymatic reaction was stopped by heating the tube for 15 min at 70° C. The resulting c D N A was stored at -20° C until use. 3.2.5.4 Real time- reverse transcriptase PCR analysis Real time R T - P C R reactions consisted of 12.5 pi of Brilliant® SYBR® Green Q R T - P C R master mix (Stratagene, L a Jolla, C A ) , 20 pmol of primers TLP004 ( 5 ' G G G A T C C T C T T C T G A C C A A A 3 ' ) and TLP005 (5' A C C G T G T A C T G G C A C C T G T T 3 ' ) , l p l of c D N A , and water to a total volume of 25 pi . R T - P C R was performed in the Mx3000P® Q P C R System (Stratagene, La Jolla, C A ) . The following thermal cycling regime was used to amplify a 142 bp fragment of the T L P transcript: 1 cycle of 5 min at 95° C, 40 cycles of 95° C for 30 seconds, 55° C for 1 min, and 72° C for 30 seconds, followed by 1 cycle of 95° C for 1 min, 55° C for 30 sec, and 95° C for 30 sec. Plasmid D N A from the transformation vectors was used to quantify gene expression. The quantification method compared the amplification of the T L P transcript in the sample against a regression curve based on known concentrations of the same T L P transcript from the plasmid used for transformation. The regression curves were generated by using 10-fold serial dilutions (10"', 10"2, 10"3, 10"4, 10"5and 10"6 ug ml" 1) of each plasmid concentration (pSM-3 T L P and pCambia-bar-ubi T L P ) and quantified in 58 three replicates by real time R T - P C R . The copy number of the initial template in the plasmid's sample was calculated using the following formula: mass amplicon (g) copy number average mol wt of bp x Avogadro's number x amplicon length bp 3.2.5.5 Total protein extraction Exacum leaf tissue (0.2 g) was initially ground into a powder with liquid nitrogen, and then re-ground in the presence of an extraction buffer (Lin et al., 1995) containing 1.5 ml 0.1 M K H 2 P 0 4 , 0.5M P S M F (phenylmethylsulfonylfluoride) p H 6.7. The tissue slurry was centrifuged for 20 min at 13000 rpm and the supernatant collected. Using this supernatant, total protein concentrations were calculated using the Bradford protein assays (BioRad Laboratories, Hercules, C A ) with bovine serum albumin as the standard. The protein solutions were then concentrated using ultra filtration columns (Millipore Ultrafree®-MC Centrifugal Filter Units with U F Membrane, Billerica, M A ) and concentration recalculation. Protein was stored at -80° C until use. For tobacco, 0.3 g of small fresh leaves were placed in a 'Matrix D ' tube (QBiogene, Montreal, QC) with 1.2 ml extraction buffer (50 m M Tr i s -HCl , pH 6.5, 1 m M N a 2 E D T A . 2 H 2 0 , 100 m M N a C l , and 0.1% Triton X-100). The tubes were placed on the high-speed benchtop device for disruption of cell membranes FastPrep FP120 (Qbiogene, Montreal, QC) on setting 6.0 for 40 sec. The tubes were then centrifuged for 5 min at 7500 rpm at 4° C with the supernatant transferred to a fresh 1.5 ml microfuge tube. Samples were concentrated using ultrafiltration columns as described above and stored at -80° C until use. 59 3.2.5.6 Immunoblot assays For Exacum and tobacco samples, proteins were separated by polyacrylamide gel electrophoresis ( P A G E ) in the presence of sodium dodecyl sulfate (SDS) (Laemmli, 1970). Briefly, the protein solutions previously concentrated by ultrafiltration were mixed with 5x sample buffer (60 m M Tr i s -HCl , 25% glycerol, 2% SDS, 14.4 m M p-mercaptoethanol, and 0.1% bromophenol blue) and heated for 10 min at 65° C. Two protein gels were run together with the BioRad protein standards (BioRad, Hercules, C A ) at 140 V for 60 min. The first gel was stained for 1 h with Coomassie Blue R-250 Stain (0.1% Coomassie Brillant Blue R-250, 45% methanol, and 45% glacial acetic acid in water) and distained overnight with a distaining solution (10% methanol and 10% glacial acetic acid in water). Proteins from the second gel were transferred to a nitrocellulose membrane (BioRad, Hercules, C A ) using the semi-dried transfer method as follows: the S D S - P A G E gel and the membrane were pre-treated with the semi-dry transfer buffer (0.04 M glycine, 0.05 M TrisBase, 20% methanol and 1 m M SDS) and then placed in a Trans-Blot SD semi dry transfer cell (BioRad 170-3940) set at 10 V for 30 min. Afterward, the membrane was allowed to dry for 10 min and then probed with a T L P antibody. Detection of the T L P protein was done using a rabbit polyclonal antibody that is reported to bind to the rice T L P (Velazhahan et al., 1998). The antibody used was kindly provided by Dr. Muthukrishnan (Biochemistry Department, Kansas State University). The serum was diluted to 1:500 in T B S (Tris-Base, sodium chloride, pH 7.5) and 3% milk powder before use. The nitrocellulose membrane was re-equilibrated for 5 min in T B S and blocked for 1 h with T B S 3% milk powder. Later, the membrane was incubated 60 on a rocking shaker for 1 h with the primary antibody followed by the goat anti-rabbit antibody (gamma chain- Sigma A2556). Reactions were then visualized in a developing buffer ( B C I P / N B T - Sigma B1911, St. Louis, M O ) . 3.3 Results 3.3.1 Transformation vector and bacterial strains The fragment obtained following site directed mutagenesis was cloned into the T O P O cloning vector and inserted into E. coli. Gel electrophoresis following P C R amplification of the 600 bp fragment of the T L P gene confirmed that five of the ten E. coli colonies contained the desired insert. Plasmid D N A from these positive colonies was purified and sequenced. A B L A S T search (NCBI) of the sequence identified a 99% match with the m R N A sequence from Oryza saliva L . (subsp. japonica) that encodes the thaumatin-like protein. As expected, three nucleotides were mismatched and correlated to the same nucleotides mutated to create the newly inserted restriction sites. A B L A S T -x search confirmed that the sequence would code for the T L P protein reported from rice. Plasmid D N A was digested with EcoRI and BamHl restriction endonucleases to release the T L P gene from the vector. A s expected, the digested samples produced two bands representing the 1 kb insert and the 3.5 kb vector. The 1 kb band (i.e. T L P gene) was extracted from the gel and ligated into the pSM-3 vector that was previously digested with the endonucleases BamHl and E c o R I . Following ligation, the binary vector + insert were used to transform E. coli competent cells. P C R amplification followed by gel electrophoresis confirmed the incorporation of the insert into E. coli by amplifying a 1.2 kb fragment (i.e. 264 bp of the 35S promoter (2x) + 1 kb T L P gene). A single positive 61 colony yielded plasmid D N A that was then sequenced. The TLP003 primer used for the sequencing reactions was a reverse primer that amplifies the junction between the promoter and the insert. The sequences results confirmed that the insert had ligated to the vector pSM-3 in the correct orientation. This construct was then used to transform Agrobacterium strains C-58 and EHA-105 . Colonies that grew on the selection medium (i.e. rifamycin 50 mg l " 1 and kanamycin 50 mg 1"') were confirmed as transformants by P C R , as described above. Gel electrophoresis displayed the expected band (600 bp) that corresponded to the T L P gene fragment in C-58 (Figure 3.1) and EHA-105 Agrobacterium strains. Together, these results confirmed the incorporation of the plasmid pSM-3 T L P containing a hygromycin resistance gene and the two tandem 35S promoters in the bacteria (Figure 3.2). Figure 3.1 Gel electrophoresis of P C R amplicons obtained from the C-58 Agrobacterium strain transformed with the pSM-3 T L P transformation vector. LB 35S Hyg 35S 2x 35S TLP PolyA Figure 3.2 Schematic representation of transformation vector pSM-3 T L P . 63 3.3.2 Hygromycin sensitivity Survival of plants cultured on hygromycin medium was assessed after various periods of culture, with significant genotypic differences observed. For example, wild type leaf explants of genotype 01-09-1 were completely necrotic after 26 days on medium containing 20 mg 1"' of hygromycin (Figure 3.3a). However, after 15 days on medium containing 25 mg 1"' of hygromycin, only 5% tissue survived, thus proving to be a more time efficient method. Furthermore, no antibiotic concentration evaluated allowed for shoot development from this genotype, whereas shoots developed on antibiotic-free medium after 22-24 days of culture. Compared to genotype 01-09-01, genotype 01-37-61 displayed much greater tolerance to this antibiotic. At the highest concentration of hygromycin tested (i.e. 30 mg l"1), 01-37-61 still had 11.1% survivorship after 26 days (Figure 3.3b). Leaf explants of genotype 01-37-61 started to produce shoots on antibiotic free-medium after 21 days of culture. These results suggested that once effective genotype-specific concentrations are determined, hygromycin resistance is a suitable selectable marker for Exacum. Although effective antibiotic concentrations varied slightly between genotypes, none of the explants exposed to any concentration of hygromycin induced shoot organogenesis. Based on these data, medium containing 25 mg 1"' of hygromycin was used as the initial selection pressure to identify putative transformants. 64 Hygromycin (mg I"1) Figure 3.3 Hygromycin sensitivity results a) genotype 01-09-01, b) genotype 01-37-61 Error bars equal standard deviation. 65 3.3.3 Generation of transgenic Exacum Transformation of Exacum was carried out with Agrobacterium strains EHA-105 and C-58, each carrying the binary vector pSM-3 T L P (Figure 3.2) and strain LBA-4404 carrying the binary vector pCambia-bar-ubi T L P , with selection on hygromycin supplemented medium. From leaf explants, adventitious shoots formed along the cut borders within 5-6 weeks (Figure 3.4a) while callus and seedling cultures formed shoots following 9 and 6 weeks, respectively (Figure 3.4b and c). None of the non-transfonned controls, from leaf, callus or seedling, developed shoots but rather turned necrotic and died (Figure 3.4d). Putative transformed shoots were sub-cultured every three weeks on elongation medium (2 mg 1"' 2ip and 0.01 mg 1 N A A ) supplemented with the selection antibiotics. One healthy shoot per explant was cultured on the selection medium containing charcoal in an attempt to avoid chlorosis of the younger leaves typical of this germplasm (Figure 3.5). Once healthy tissue was produced, P C R analyses on genomic D N A were performed on a total of 15 putative transformants (Figure 3.6). Twelve Exacum lines were confirmed as transgenic (9 lines from the C-58 strains and 3 lines from the LBA-4404 strain). The plants were maintained under constant selection pressure to reduce the possibility of false positives and 'escapes'. A l l transgenic lines obtained with the C-58 strain were part of the standard treatment. Modifications to the standard protocol described in the Materials and Methods section did not increase transformation efficiency in Exacum. Agrobacterium C-58 was the most effective strain, yielding 9 of the 12 transgenic lines produced (Figure 3.6). Despite numerous attempts with bacterial strain EHA-105 , 66 no transformed tissue was recovered. The transgenic lines were produced from leaf discs (4), calli (2), and seedlings (3) among different Exacum genotypes (Table 3.1). Overall transformation efficiency was considered low (< 1%) as compared to other dicotyledonous plants while transformation efficiency was not different among explant sources (i.e. leaf, callus or seedling) (Table 3.1). Leaf discs were also co-cultivated with Agrobacterium containing the empty pSM-3 vector; unfortunately, no shoots were obtained from these treatments. Figure 3.4 Shoot organogenesis from Agrobacterium-infected tissue on hygromycin medium a) leaf discs, b) callus, c) seedling, and d) non-transformed control. Figure 3.5 Transgenic Exacum shoot on charcoal elongation medium with antibiotics for selection. 68 Table 3.1 Comparison of transformation efficiency among Agrobacterium strains and Exacum explant source. Agrobacterium strain (plasmid) Explant type Number of Explants Co-cultivated Number of Transgenic Lines Produced Percent Transgenic Plants EHA-105 (PSM-3TLP) leaf-discs 725 0 0 EHA-105 (empty vector) leaf-discs 275 0 0 C-58 (pSM-3 TLP) leaf-discs 425 4 0.94 seedling (individuals) 323 3 0.93 callus (petri-plates) 27 2 -C-58 (empty vector) leaf-discs 75 0 0 Non-infected control leaf-discs 150 0 0 seedling (individuals) 80 0 0 callus (petri-plates) 3 0 0 69 Figure 3.6 PCR products from transgenic Exacum lines transformed with pSM-3 TLP construct. C (-) non transformed control and C (+) plasmid DNA containing the TLP gene, a) transgenic lines from callus and leaf disc, b) transgenic lines from seedlings. 70 The bacterial strain LBA-4404 carrying the plasmid pCambia-bar ubi-TLP was also used for co-cultivation of Exacum leaf discs. This construct differed from the p S M -3 construct in that it harbored the ubiquitin promoter to drive the T L P expression and has two selectable marker genes (i.e. hph and bar genes). The transformation methods used were basically the same as described for the other strains. However, two modifications were tested as described in the Materials and Methods section. The modifications were pre-treating the leaf discs on shoot induction medium and co-cultivating the explants with the bacteria on 2,4-D medium for two days. Three transgenic lines, confirmed by P C R analyses (Figure 3.7) were produced by this vector and construct. In addition, among the 3 transgenic lines, the highest transformation efficiency was when leaf discs were first pre-treated for two days on induction medium prior to co-cultivation (Table 3.2). Despite the low transformation efficiency, it was the highest observed from any experiment. Unfortunately, it is not clear whether the increased efficacy was based on bacterial strain, the promoter driving the T L P gene, or some other factor. 71 Table 3.2 Transformation efficiency of Exacum transformed with the Agrobacterium strain LBA4404 carrying the pCambia-bar-ubi TLP plasmid. Agrobacterium Treatment Number of Transgenic % strain explants co- lines transgenic (plasmid) cultivated plants L B A 4 4 0 4 (pCambia-bar-ubi-75 1 1.3 TLP) Standard Pre-treatment N A A + B A 100 2 2 Co-cultivation 2,4-D 75 0 0 0 7> S00 bp Figure 3.7 PCR products from transgenic Exacum lines transformed with the pCambia-bar-ubi TLP construct. C (-) non transformed, C (+) plasmid DNA containing TLP gene. 72 3.3.4 Generation of transgenic tobacco T L P transgenic tobacco plants were produced as transformation controls for this experiment. Leaf discs of Nicoliana tabaccum var. Xanthi were co-cultivated with Agrobacterium C-58 carrying the p S M - 3 - T L P construct. Adventitious shoots arose from cut borders of leaf discs within 2-3 weeks, while no shoots were produced on control explants. Putatively transformed shoots were isolated and maintained on the same medium for 7-8 weeks, subculturing them every 3 weeks to maintain the antibiotic selection pressure (Figure 3.8a). Once the plants had elongated (i.e. 3-4 cm), they were transferred to M S O medium with antibiotics for rooting (Figure 3.8b). Four lines were confirmed transgene-positive by P C R analyses using the T L P F W D and T L P R E V primers (Table 3.3). Additionally, 2 hygromycin-positive lines transformed with the bacteria carrying the empty pSM-3 vector were confirmed by P C R using H Y G F W D and H Y G R E V primers designed to amplify a 518 bp fragment of the htp gene. N o gene specific P C R products were amplified from non-transformed tobacco controls (Table 3.3). 73 Table 3.3 Transformation efficiency of tobacco transformed with the Agrobacterium strain C-58 carrying the pSM-3 TLP plasmid. Agrobacterium Number of Transgenic 0/ strain explants co- lines transgenic (plasmid) cultivated plants C-58 (pSM-3 TLP) 175 4 2.3 C-58 (empty vector) 75 2 2.7 Control 75 0 0 Figure 3.8 Transgenic tobacco plants on hygromycin medium a) shoot elongation b) development of complete plant on MSO medium with antibiotics. 74 3.3.5 Molecular analysis of gene expression in transgenic lines Transgene expression levels in Exacum (nine pSM-3 T L P lines, three pCambia-bar-ubi T L P lines) and tobacco (two pSM-3 T L P lines) transgenic lines were analyzed by real time R T - P C R . The reverse transcriptase reactions were performed with 1 ug of total R N A and the c D N A product was evaluated with real time P C R . The TLP004 and TLP005 primers successfully amplified the expected 142 bp fragment from the transgenic lines. Serial dilutions of the pSM-3 T L P and pCambia-bar-ubi T L P plasmids (62 ng and 108 ng, respectively) were used to generate the regression curves. The copy number of the amplicon in each plasmid preparation (see Materials and Methods section) was regressed against the Ct value calculated by the thermocycler. Since the Ct value is a logarithmic value, logio(copy number) was used to create the regression curve. For the pSM-3 T L P plasmid, the regression equation was x= (Ct-32.40)/(-2.48). This formula was then applied to back-calculate the number of amplicons per ug of R N A of each Exacum and tobacco transgenic line. The amplification plots from these P C R reactions (Figure 3.9a) showed that the fluorescence signal became greater than the background noise beginning at cycle 20, indicating that the fluorescence signal started to be measured at this point. The dissociation curve (Figure 3.9b) indicates that only one product was generated during P C R amplification. This graphical display allowed me to verify that no primer dimers formed or any other unexpected amplification products were generated during these reactions. In Exacum, gene expression was observed in 8 out of 9 putative transgenic lines transformed with the pSM-3 T L P construct (Figure 3.10), and they were classified as 75 having either high, medium or low expression levels. Transgenic lines E 37-50.TLP2 showed high expression, five lines (E 01-42-3.TLP2, E 01-42-3.TLP3, E 01-42-3.TLP4, E 04-13.TLP1, and E 04-5.TLP1) showed medium levels and 2 lines (E 37-50.TLP1 and E 4-13. TLP2) showed low gene expression. Line E 01-42-3.TLP 1 had no detectable expression of the transgene. In addition, no signal was detected from any non-transformed control plants. Overall, transgenic tobacco lines showed significantly lower expression levels compared to Exacum. For example, the Exacum line with the lowest expression level was 1.9 and 2.4 times higher when compared to the expression levels of the two tobacco lines (Figure 3.11). In addition, no signal was detected from the non-transformed tobacco control. Three transgenic lines of genotype 01-37-61 that were transformed with the pCambia-bar-ubi T L P construct were evaluated with a specific regression curve created for this plasmid. The regression equation used was x= Ci- 32.10/ (-2.67). Compared to the transcripts levels of the pSM-3 T L P plasmid, these three lines displayed low transgene expression (Figure 3.12). N o signal was obtained from the non-transformed control plants. 76 Figure 3.9 Graphical outputs of gene expression of transgenic Exacum and tobacco lines obtained by real time RT-PCR a) amplification plots and b) dissociation curve. 77 2500 -j < z 2000 • cc at 1500 • la c o o 1000 • Q. E re 500 • 0 • I X I <§>• <§> ,2> <0-' A' >>' & M V v v v c v , x> , ' Exacum lines Fig 3.10 TLP expression levels from transgenic Exacum lines transformed with Agrobacterium strain C58 carrying the pSM-3 TLP. Error bars equal standard deviation. 150 < z a. o> 100 H o o 50 E re mi i l l Ik Tobacco and Exacum lines Fig 3.11 TLP expression levels from transgenic tobacco and Exacum lines transformed with Agrobacterium strain C-58 carrying the pSM-3 TLP. Error bars equal standard deviation. 78 Fig 3.12 TLP expression levels from transgenic Exacum lines transformed with Agrobacterium strain LBA4404 carrying the pCambia-bar-ubi TLP. Error bars equal standard deviation. 79 3.3.6 TLP protein expression Total soluble proteins from Exacum and tobacco were separated on 12% SDS-polyacrylamide gels (10-15 ug of protein per well), then electro-blotted onto a nitro-cellulose membrane and probed with a rabbit antibody raised against a rice T L P . Based on the original methods described, the expected protein could not be detected from the transgenic plants produced. Therefore, various modifications to the original methods were evaluated in an attempt to increase detection sensitivity, including concentrating the protein from 0.5-0.75 mg ml" 1 to 1-1.25 mg ml" 1 (20-25 ug per well) using ultra-filtration columns; evaluating various gel loading buffers to test their effect on protein mobility in the SDS gel; and proteins were separated under non-denaturing conditions to assess i f the protein was affected by the denaturation process. Total protein extracts from transgenic lines of Exacum (transformed with the pSM-3 T L P construct) displayed a unique band of approximately 14 kDa which was absent from non-transformed control tissue (Figure 3.13). in addition, one Exacum line transformed with the other construct, pCambia-bar-ubi T L P , showed a band of similar size (i.e. 14 kDa) (Figure 3.14). Proteins from the transgenic tobacco lines did not differ from the wi ld type control at the 14 kDa location. However, one transgenic tobacco displayed a unique band of approximately 21 kDa that was not observed in the non-transformed tobacco control (Figures 3.14; 3.15). so Figure 3.13 Protein separation by SDS PAGE of Exacum transgenic lines transformed with Agrobacterium strain C-58 carrying the pSM-3 TLP construct displaying the 14 kDa band. Figure 3.14 Protein separations by SDS PAGE of Exacum and tobacco transgenic lines transformed with the Agrobacterium strains C-58 and LBA-4404 displaying the 14 kDa band on Exacum and the 21 kDa band on tobacco. Figure 3.15 Protein separation by SDS PAGE of Exacum and tobacco transgenic lines transformed with the C-58 Agrobacterium strain containing the pSM-3 TLP displaying the 21 kDa band on tobacco lines. 82 Western blot analyses failed to produce a signal from the T L P protein from any transgenic line, regardless of genotype or construct. To identify possible problems with the procedures used, the efficiency of protein transfer from SDS polyacrylamide gels to the nitrocellulose membranes was tested by using a pre-stained kaleidoscope protein standard. In addition, the SDS gels were stained with Coomassie blue following the transfer to evaluate the amount of protein remaining in the gel. Based on these observations, several adjustments to the protocol were made including: 1) testing two concentrations of the primary antibody (i.e. 1:250 and 1:500); 2) testing various times of exposure to the antibody (i.e. 1, 2, and 8 hrs); 3) testing two temperatures of exposure (i.e. 4 ° C and 18° C) ; 4) testing two concentrations of the secondary antibody (i.e. 1:100000 and 1:50000); and 5) evaluating two detection systems (i.e. colorimetric B C 1 P / N B T and chemiluminescence). Unfortunately, none of these attempts permitted for effective hybridization with this rabbit polyclonal antibody In a last attempt to work through the problems with the Western blots, a second T L P polyclonal antibody was acquired that was originally used to test the induction of T L P protein in Douglas-fir (Pseudotsuga menziesii [Mirb,] Franco) infected with Phellinus weirii (Murrill) R .L . (Zamani et al., 2004). In addition, a known positive control of this protein was included in this experiment. Colorimetric analysis detected a signal from the positive control, but unfortunately, not from any of the Exacum or tobacco transgenic lines (data not shown). This confirmed that the methods employed were correct despite a lack of signal from my transgenic plants. 83 3.4 Discussion This study is the first report of Agrobacterium-mediated transformation of Exacum Styer Group. The results from this research enable this technology to be used not only for the enhancement of fungal resistance but also for the insertion of other novel genes into this germplasm. Agrobacterium-mediated transformation is composed of multiple interacting factors, each able to affect the overall efficiency of the system. Important factors that affect the ability of Agrobacterium to transfer D N A to the plant genome include plant genotype, bacterium genotype, physiological condition of both organisms, and the physical bacterium x plant interaction (Godwin el al., 1992). The transformation percentages obtained for Exacum are considered very low compared to other plants that have been transformed with similar T L P constructs (Chen and Punja, 2002; Fu et ah, 2004). A t this point, it is unclear as to why this germplasm responded so differently. It could be due to differences in bacterial strain, co-cultivation methods, or some inherent quality in Exacum that makes it more difficult to transform. The most closely related species to Exacum that is reported to have been transformed is lisianthus (Eusloma grandiflorum Griseb.), also a member of the Gentianaceae family. Ledger et al. (1997) also reported low transformation percentages using G U S as the reporter gene. The report suggested that the regeneration of plants from transgenic tissue was the limiting factor to obtain higher number of transgenic lisianthus. This possibility may also exist for Exacum. To achieve higher transformation rates, leaf explants were exposed to a pre-treatment to stimulate cell division, a common requirement for successful Agrobacterium transformation (Binns and Thomashow, 1988). In addition, Agrobacterium cultures were 84 pre-conditioned and co-cultivated with acetosyringone, a compound reported to induce vir genes, extend the host range of some Agrobacterium strains, and deemed essential for transformation in some plant species (Hiei et al, 1994; Godwin et al, 1991). The length of the co-cultivation period is also critical for transformation success. Here, explants were co-cultivated with the bacterium for two days, which should have allowed sufficient time for the transfer of the T i plasmid without adversely affecting the plant with over growth of Agrobaclerium. In this study, none of these modifications increased the transformation efficiency. Promoter sequences are critical components to transgene expression. Two promoters were evaluated within this system, a 2x tandem C a M V : 3 5 S and the maize ubiquitin promoter. Both are reported to promote constitutive expression of transgenes when successfully incorporated into host genomes. The 2x tandem C a M V : 3 5 S promoter with the A M V leader sequence, a variant of the cauliflower mosaic virus 35S promoter, is reported to increase the activity of the transferred D N A (Kay et al., 1987; Datla el al., 1993). The second, the maize ubiquitin (Ubi-1) promoter has been widely used to drive constitutive transgene expression in monocotyledonous species (Rooke et al., 2000), but much less frequently in dicotyledoneous species. In the present study, the gene expression patterns showed higher transcript levels on plants transformed with the double 35S promoter /AMV leader sequence (pSM-3 plasmid) compared to the lines transformed with the ubiquitin promoter (pCambia-bar-ubi). However, it is not clear i f this difference resulted only from the promoter used. The selectable marker system employed is also a critical component to the transformation process. Often, a selection system that works on one species is ineffective 85 on another. During preliminary experiments, use of the nplU gene as the selectable marker was shown to be unsuitable based on Exacum's tolerance to kanamycin (Appendix II). However, subsequent evaluations of other antibiotics revealed Exacunfs sensitivity to hygromycin and supported using the hpl gene in the selection system. The hygromycin phosphotransferase enzyme has been used as a selectable marker when npl\ I was found to be ineffective (Twyman et al., 2002). Once the transgene has been incorporated into a host genome, transcript expression needs to be confirmed and characterized before analysis of protein production. Results from the real time R T - P C R demonstrated that the expression levels of the T L P gene varied significantly among transgenic lines of both Exacum and tobacco. This variation may be caused by the promoter or the bacterial strains used, by repeated transgene insertions, and/or by post-transcriptional gene silencing. Despite the thought that transgenic plants obtained from Agrobacterium-mediaXed transformation are considered 'less complex' (i.e. transgene integration on low copy number) than those obtained after direct gene transfer (i.e. particle bombardment), the integration of multiple T - D N A copies into direct and inverted repeats is fairly common (De Block and Debrouwer, 1991). In plants, these repeated transgene insertions have been correlated with variation in transgene expression and induction of transgene silencing (Hobb el al., 1993; Jorgensen, 1996; and Vain et al., 2002). The induction of post-transcriptional gene silencing is a common situation when using the strong viral 35S promoter which can lead to over-production of the transgene R N A above a threshold level, triggering auto-degradation (Vaucheret et al., 1997). These phenomena could explain why expression 86 levels ranged from high to undetectable among transgenic Exacum lines despite P C R confirmation of gene insertion. Analysis of T L P protein production from these transgenic lines was problematic. Although S D S - P A G E allowed for the identification of a small protein (~14 kDa) differentially produced between Exacum transgenic lines and non-transformed controls, it is difficult to confirm that this protein is T L P . It's smaller than expected size suggests that the protein could be degradated or that the protein has an abnormal mobility during electrophoresis. Unfortunately, the use of tobacco as an internal control did not help resolve this protein's identity. Despite using identical constructs, transformed tobacco produced yet another novel protein band as compared to either non-transformed tobacco controls or transformed Exacum lines (Fig 3.15). However, this new band was much closer in size to the expected T L P protein. The identities of either protein remain inconclusive. Confirmation of the production of specific novel proteins is best accomplished by immunoblot analyses using protein-specific antibody probes. Although modifications to protein extraction, protein visualization, and antibody source were attempted, as well as the inclusion of internal controls that confirmed methodology, Western blots displayed no signal from either Exacum or tobacco transgenic lines. Previous research showed that the leader A M V sequence combined with a duplicated-enhancer C a M V : 3 5 S promoter should have been adequate to express high levels of the transgene (Datla et al. 1993). This, together with the fact that the T L P transcripts were shown by real time R T - P C R , indicate that the lack of a protein signal could be explained firstly by loss of activity from the antibody (i.e. loss of antigen-binding capacity) or secondly, by a post-translational 87 modification of the protein antigens. This modification is usually just a problem that affects a subset of proteins in the sample and normally only decreases the strength of the signal, not prevent it entirely. However, in some cases, post-translational modification of protein antigens can result in significant changes to the antigen structure, thereby completely blocking access of the antibody (Harlow, 1999). Based on the results from Western blots, a second detection system (chemiluminescent) was used to evaluate whether the lack of signal was related to low levels of protein or to poor binding with the antibody. Despite using the more powerful chemiluminescent detection system, which has made it increasingly more possible to identify proteins that were previously undetectable, T L P protein was not detected, supporting the idea that there was a problem with antibody binding efficiency. To test this hypothesis, another antibody was employed, the T L P polyclonal antibody from Douglas-fir. Rice and Douglas-fir T L P s possessed 50.2% sequence similarity, therefore may not have sufficient homology to bind. Subsequent immunoblots with this antibody produced signal only from the positive control (from Douglas-fir), but not from our transgenic lines. In summary, T L P insertion was achieved in 9 lines of Exacum using the pSM-3 T L P plasmid with the Agrobaclerium strain C-58 and in 3 lines using the pCambia-bar-ubi T L P plasmid with the Agrobacterium strain LBA4404 . Also, 4 tobacco transgenic lines were produced and used as positive controls. Transcripts of the T L P gene were shown to be present but varied significantly among transgenic lines. Unfortunately, no T L P protein production was confirmed and may have been caused by the loss of activity of the antibody. Future studies should focus on the development of a new antibody that 88 would have higher specificity to this T L P . The results from this study also suggested that the protocols need to be further optimized primarily in terms of explant source and physiological conditions for regeneration to increase the transformation efficiency. The transgenic plants produced in this study are currently being transferred to greenhouse conditions for use in fungal pathogen bioassays. With this research, a first step has been taken toward the establishment of transformation protocols for this ornamental plant. 89 3.5 Literature Cited 1. Abad, L . , D 'Urzo , M . , L iu , D. , Narasimhan, M . , Reuveni, M . , Zhu, J., N i u , X . , Singh, N . , Hasegawa, P., and Bressan, R. 1996. Antifungal activity of tobacco osmotin has specificity and involves membrane permeabilization. Plant Sci. 118: 11-23. 2. Binns, A . , and Thomashow, M . 1988. Cel l biology of Agrobaclerium infection and transformation of plants. Annu. Rev. Microbiol . 42:575-606. 3. Bradford, M . 1976. A rapid and sensitive method for the quantization of microgram quantities of protein utilizing the principle of protein-dye binding. Annals Biochem. 72: 148-254. 4. Cabanas, M . , Vazquez, D. , and Modolel l , J. 1978. Dual interference of hygromycin B with ribosomal translocation and with aminoacyl-tRNA recognition. Eur. J. Biochem. 87: 21-27. 5. Chen, W. and Punja, Z . 2002. Transgenic herbicide-and disease tolerant carrot (Daucus carota L.) plants obtained through Agrobacierium-medmted transformation. Plant Cel l Rept. 20: 929-935. 6. Chen, W. , Chen, P., L i u , D. , Kynast, R., Friebe, B . , Velazhahan, R., Muthukrishnan, S., G i l l , B . 1999. Development of wheat scab symptoms is delayed in transgenic wheat plants that constitutively express a rice thaumatin-like protein gene. Theor. App l . Genet. 99: 755-760. 7. Datla, R., Bekkaoui, F., Hammerlindl, J., Pilate, G . , Dunstan, D. and Crosby, W. 1993. Improved high level constitutive foreign gene expression in plants using an A M V R N A 4 untranslated leader sequence. Plant Sci. 94: 139-149. 8. De Block, M . and Debrouwer, D. 1991. Two T - D N A ' s co-transformed into Brassica napus by a double Agrobacterium tumefaciens infection are mainly-integrated at the same locus. Theor. App l . Genet. 82: 257-263. 9. Frendo, P., Didierjean, J., Passelegue, E. , Burkard, G. 1992. Abiotic stresses induce a thaumatin-like protein in maize; c D N A isolation and sequence analysis. Plant Sci . 85: 61-69. 10. Fu, D. , Tisserat, N . , Xiao, Y . , Settle, D . , Muthukrishnan, S., Liang, G . 2005. Overexpression of rice TLPD34 enhances dollar-spot resistance in transgenic bentgrass. Plant Sci. 168: 671-680. 11. Gamborg, O., Murashige, T., Thorpe, T., and Vas i l , I. 1976. Plant tissue culture media. In Vitro J. Tiss. Culture Assoc. 12: 473-478. 12. Godwin, I.,Todd, G. , Ford-Lloyd, B . , Newbury, H . 1991. The effects of acetosyringone and pH on Agrobacterium-mediated transformation vary according to plant species. Plant Cel l Rept. 9: 671-675. 13. Godwin, I., Fordlloyd, B. and Newbury, H . 1992. In vitro approaches to extending the host-range of Agrobacterium for plant transformation. Austr. J. Bot. 40: 751-763. 14. Gonzalez, A . , Jimenez, A . , Vazquez, D. , Davies, J., Schindler, D. 1978. Studies on the mode of action of hygromycin B , an inhibitor of translocation in eukaryotes. Bioch. Biophys. Acta 521: 459-69. 15. Hajgaard, J., Jacobsen, S., Svendsen, I. 1991. Two antifungal thaumatin-like proteins from barley grain. F E B S Lett. 291: 127-131. 91 16. Harlow, E . and Lane, D . 1999. Choosing antibodies. In Harlow, E. and Lane, D. (Ed.). Using antibodies: a laboratory manual. Cold Spring Harbor Lab. Press, pp: 47-48. 17. Hecht, M . , Ecker, R., Stav, R., Watad, A . , Hecht, M , Hirschberg, J. and Mann, V . 1997. Agrobaclerium- mediated transformation of Lisianthus (Eustoma grandiflorum). Acta Hort. 447: 339-340. 18. Hiei , Y . , Ohta, S., Komari , T., and Kumashiro, T. 1994. Efficient transformation of rice (Oryza saliva L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T - D N A . Plant J. 6: 271-282. 19. Hobbs, S., Warkentin, T., and DeLong, C. 1993. Transgene copy number can be positively or negatively associated with transgene expression. Plant M o l . B io l . 21: 17-26. 20. Huynh, Q., Borgmeyer, J., Zobel, J. 1992. Isolation and characterization of a 22 kDa protein with antifungal properties from maize seeds. Bioch. Biophys. Res. Commun. 182: 1-5. 21. Jorgensen, R., Cluster, P., English, J., Que, Q. and Napoli , C . 1996. Chalcone synthase co-suppression phenotypes in petunia flowers: comparison of sense vs. antisense constructs and single-copy vs. complex T - D N A sequences. Plant M o l . B io l . 31: 957-973. 22. Kay, R., Chan, A . , Daly, M . and McPherson, J. .1987. Duplication of C a M V 35S promoter sequences creates a strong enhancer for plant genes. Science 236: 1299-1302. 23. Klimaszewska, K . , Rutledge, R., Seguin, A . 2005. Genetic transformation of conifers utilizing somatic embryogenesis. Methods M o l . B io l . 286: 151-164. 24. Laemmli, U . 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685. 25. Malehorn, D. , Borgmeyer, J., Smith, C , Shah, D . 1994. Characterization and expression of an antifungal zeamatin-like protein (Zip) gene from Zea mays. Plant Physiol. 106: 1471-1481. 26. Monteiro, S., Barakat, M . , Picarra-Pereira, M . , Teixeira, A . , Ferreira, R. 2003. Osmotin and thaumatin from grape: A putative general defense mechanism against pathogenic fungi. Phytopathology 93: 1505-1512. 27. Pettinger, R., Wolfe, R., Hoehn, M . , Marks, P., Dailey, W., and McGuire , J. 1953. Hygromycin 1. Preliminary studies on the production and biological activity of a new antibiotic. Antibiot. Chemother. 3: 1268-1278. 28. Rebmann, G . , Mauch, F. , Dudler, R. 1991. Sequence of a wheat c D N A encoding a pathogen-induced fhaumatin-like protein. Plant M o l . B i o l . 17: 283-285. 29. Reimmann, C , Dudler, R. 1993. c D N A cloning and sequence analysis of a pathogen-induced thaumatin-like protein from rice (Oryza saliva). Plant Physiol. 101:1113-1114. 30. Rooke, L . , Byrne, D. , Salgueiro, S. 2000. Marker gene expression driven by the maize ubiquitin promoter in transgenic wheat. Ann . App l . B io l . 136: 167-172. 31. Semeria, L . , Ruffoni, B . , Rabaglio, M . , Genga, A . , Vaira, A . , Accotto, G . and Allavena, A . 1996. Genetic transformation of Eusloma grandiflorum by Agrobaclerium lumefaciens. Plant Cel l Tiss. Organ Cult. 47: 67-72. 93 32. Scholthof, H . , Scholthof, K . , and Jackson, A . 1996. Plant virus gene vectors for transient expression of foreign proteins in plants. Annu. Rev. Phytopathol. 34: 299-323. 33. Singh, A . , Ursic, D . and Davies, J. 1979. Phenotypic suppression and misreading in Saccharomyces cerevisiae. Nature 277: 146-148. 34. Tinland, B . 1996. The integration of T - D N A into plant genomes. Trends Plant Sci. 1: 178-184. 35. Twyman, R . M . , Stoger, E . , Koh l i , A . , Capell, T., Christou, P., 2002. Selectable and screenable markers for rice transformation. M o l . Methods Plant Analysis. 22: 1-17. 36. Vain , P., James, V . , Worland, B . and Snape, J. 2002. Transgene behavior across two generations in a large random population of transgenic rice plants produced by particle bombardment. Theor. App l . Genet. 105: 878-889. 37. Vaucheret, H . , Nussaume, L . , Palauqui, J., Quillere, I. and Elmayan, T. 1997. A transcriptionally active state is required for post-transcriptional silencing (co-suppression) of nitrate reductase host genes and transgenes. Plant Cel l 9: 1495-1504. 38. Velazhahan, R., Cole, K . , Anuratha, C , and Muthukrishnan, S. 1998. Induction of thaumatin-like proteins (TLPs) in Rhizoclonia so/am-infected rice and characterization of two new c D N A clones. Physiol. Plant. 102: 21-29. 39. Vigers, A . , Roberts, W., Selitrennikoff, C . 1991. A new family of plant antifungal proteins. M o l . Plant-Microbe Interact. 4: 315- 323. 94 40. Vigers, A . , Wiedemann, S., Roberts, W., Legrand, M . , Selitrennikoff, C , and Fritig, B . 1992. Thaumatin-like pathogenesis related proteins are antifungal. Plant Sci. 83: 155-161. 41. Woloshuk, C , Meulenhoff, E . , Sela-Buurlage, M . , van den Elzen, P., Cornelissen, B . 1991. Pathogen-induced proteins with inhibitory activity toward Phyluphthora infestans. Plant Cell 3: 619-628. 42. Zamani, A . , Sturrock, R., Ekramoddoullah, A . , L i u , J., Y u , X . 2004. Gene cloning and tissue expression analysis of a PR-5 thaumatin-like protein in Phellinus we/V/z'-infected Douglas-fir. Phytopathology 94: 1235-1243. 43. Zhan, Y . , Zeng, F., X i n , Y . 2005. Progress on molecular mechanism of T - D N A transport and integration Acta Genetica Sinica 32: 655-665. 95 Chapter IV General Conclusions 4.1 Concluding remarks and future research The main objectives of this research were to establish direct shoot organogenesis protocols for the Exacum Styer Group and to develop methods required for genetic transformation of this germplasm with a designed T L P construct. The introduction of Exacum, or any species, as a new ornamental plant can be assisted by the development of these and related in vitro techniques. Specifically, micropropagation and organogenesis techniques are important as they can facilitate the production of numerous true-to-type plants in shorter time periods as compared to traditional propagation systems. With this in mind, the first part of this thesis addressed protocol development for direct shoot organogenesis which not only has application in micropropagation, but also for the introduction of novel genes (e.g. disease resistance genes) via Agrobacterium-medvdted transformation. The second part of this thesis addressed genetic transformation techniques using a T L P construct with the hypothesis that expression of the T L P protein could enhance Exacum's resistance to fungal diseases. The work presented in this thesis is a step forward toward the goal of developing genotypes with desirable trait combinations that wi l l promote the commercial potential of this plant. Specifically, Exacum's high susceptibility to fungal root pathogens, including Fusarium spp., is thought to potentially inhibit commercial acceptance once Exacum cultivars are released to the public. 96 From this research, protocols for direct organogenesis have been established for four genotypes of Exacum. Although significant genotypic variation was identified, all genotypes evaluated were found to be highly responsive to the tissue culture conditions tested in these experiments, indicating they are amenable for use in micropropagation and genetic transformation studies. The results obtained in the first part of the thesis (Chapter II) demonstrated that despite the significant interaction between genotypes and P G R treatments, higher concentrations of B A combined with lower concentrations of N A A were optimal to produce the greatest number and highest quality of shoots from leaf explants. Comparing genotypes, it was found that 01-09-01 and 01-37-61 produced the highest number of organogenetic shoots, suggesting that they are the most suitable genotypes for transformation research. However, i f these genotypes prove recalcitrant, additional genotypes should be evaluated. In addition, due to the significant genotypic differences found in response to P G R combinations, these protocols should continue to be Optimized for each genotype. The results in Chapter II also showed that leaf explants of Exacum are very responsive to the balance between P G R s where various combinations induce the production of roots, shoots, or calli . Conventional plant breeding has played a key role in the development of crop cultivars with desirable characteristics. However, with the development of Agrobacterium-mediaXed gene transfer and other novel biotechnological tools for direct gene transfer, researchers and breeders are now able to greatly expand the range of traits and genes available for use in their plant improvement programs. Chapter III presents results on the establishment of Agrobacterium-mediaXed transformation protocols for Exacum together with the construction of a novel T L P transformation vector. This is the 97 first report on Agrobacterium-mediated transformation for the Exacum Styer Group and it is likely that the protocols established here wi l l enable this technology to be used for the enhancement of fungal resistance and also for the insertion of other novel characteristics into this germplasm. The preliminary research involved identifying the closest relatives of Exacum that have been successfully transformed and basing the initial parameters on this research. Accordingly, an Agrobacterium co-cultivation method on leaf explants using two Agrobacterium strains was employed. In addition, results from preliminary experiments showed that hygromycin resistance could be used as the selectable marker system on this germplasm. The transformation percentages obtained when using either the C-58 or L B A 4 4 0 4 Agrobaclerium strains are considered low. Also, I was not able to recover any putative transformants incubated with the EHA-105 Agrobacterium strain, even though around 1000 explants were co-cultivated. One possible reason for these low transformation efficiencies is the significant sensitivity of Exacum to hygromycin, indicating that an alternative selection system may be desirable and should be studied in future research. The low number of transgenic plants recovered did not allow for any conclusions regarding transformation efficiency across explant source. Further studies on optimizing transformation protocols should help clarify whether transformation efficiencies are different among explant sources and genotypes. In addition, protocol optimization should focus on tissue culture conditions that produce proliferating and regenerable cells into which foreign D N A can be introduced without interfering with their ability to regenerate and undergo organogenesis. 98 The A. tumefaciens-mediated transformation system is usually preferred over direct gene transfer techniques because it allows for the transfer of pieces of D N A with defined ends and with minimal rearrangement, as well as for the integration of low copy numbers. However, because of the low transformation efficiencies achieved with Agrobacterium in this work, future studies could explore direct D N A transfer methods to compare their efficiency and functionality. Transcript levels of the T L P gene in transgenic lines were demonstrated by real time R T - P C R . The expression patterns of the T L P gene in Exacum and tobacco showed differences both within and among constructs and between species. These differences could be induced by the different promoters used (2x 35S or ubiquitin), by the strain of Agrobacterium used, or by post-transcriptional gene silencing which may be a consequence of using the strong viral 35S promoter. Immunoblot analyses were unable to confirm the production of the T L P protein. Although several modifications were performed, the presence of the T L P protein with the antibodies obtained from outside sources was unsuccessful. However, a unique protein band was present from all T L P transgenic Exacum plants but absent from all non-transformed controls. Unfortunately, this band could not be confirmed as the T L P protein due to the lack of antibody binding. The results from this experiment suggest that the lack of signal from the blot could be caused primarily by the loss of activity of the antibody or secondarily by post-translational modification to the protein antigen. Once transgenic plants are produced and T L P expression confirmed, future research should evaluate phenotypic changes associated with this modification via fungal bioassays. Whether or not this gene can be considered an effective fungal resistance gene 99 wil l depend on the expression and durability of the induced resistance under field conditions and wi l l require additional research. Nevertheless, studies using the T L P protein have shown positive results on other plants; thus, the future looks promising for T L P enhanced resistance in this germplasm with its subsequent incorporation into the breeding program. This research has developed the required foundation for future improvement of Exacum by establishing tissue culture protocols and transformation methodologies. With this information, the current genetic limitations in Exacum can be addressed and improved, thereby allowing for broader appeal of this germplasm. Appendix I. 100 Evaluation of Zinc and Charcoal Supplements AI.l Introduction The development of efficient micropropagation protocols suitable for Exacum wil l help to produce enough experimental tissue for use in this and future research. Plants produced on the medium described by Riseman and Chennareddy (2004) developed chlorosis and eventual necrosis on younger leaves and was thus deemed unsuitable. Initial diagnosis of the symptoms was zinc deficiency so the first medium modifications were in zinc content. A1.2 Materials and Methods AI.2.1 Plant material The studies were based on four genotypes of Exacum currently established in culture using published protocols developed at U B C (Riseman and Chennareddy, 2004). The genotypes were 01-50-46, 01-47-49, 01-09-01 and 01-37-61, all o f which were developed as part of the U B C Exacum breeding program. The base medium consisted of full-strength M S salts (Murashige and Skoog, 1962) + 2mg 1"' 2ip + 0.01 mg 1"' N A A , 30 g 1"' sucrose, and 7g l" 1 agar. The medium pH was adjusted to 5.8 and autoclaved for 21 min at 15 psi. 101 AI.2.2 Treatments Double (17.2 mg 1"') and triple (25.8 mg 1"' 1) basal zinc concentration were evaluated using Z n S C V In addition, the addition of activated charcoal (4 g 1"') to the base medium was tested. AI.3 Results and Discussion Neither zinc concentration alleviated the symptoms described above. Based on these observations, we concluded that the chlorosis and necrosis of younger leaves were not caused by zinc deficiency. Incorporation of charcoal into the base medium produced positive effects. Within two weeks of explant transfer to medium with charcoal, new growth was green and without any signs of chlorosis or necrosis (Figure A . l ) . However, while alleviating stress symptoms, charcoal incorporation significantly reduced multiplication rates. We concluded that charcoal may have bound both undesirable compounds (chlorosis causing) and desirable compounds (growth regulators). Therefore, with the addition of charcoal, a doubling of the plant growth regulators is required in order to maintain multiplication rates as compared to the base medium. 102 b Figure A.l Evaluation of M S medium (2mg I"1 2ip + 0.01 mg I1 NAA) on Exacum genotype 01-37-61, a) medium without charcoal, chlorosis of the leaves and b) charcoal added to the medium, healthy plants. Appendix II 103 Agrobacterium-mediated transformation of Exacum (previous experiments) The original objective of my research was to evaluate and develop transformation protocols based on an Agrobacterium-mediatQd system (i.e. co-culture and floral dip) using a G U S construct driven by a single 35S promoter. To address this objective, two experiments were performed using either the floral dip or co-cultivation method to introduce the G U S construct. The availability of flowering Exacum plants allowed me to evaluate the floral-dip transformation system and then later, when suitable tissue-cultured materials were available, the co-cultivation method. AIM Floral dip All.1.1 Introduction The Agrobacterium-mediated transformation method referred to as 'floral dip' allows plant transformation without the use of tissue culture. The floral dip method first exposes pre-fertilized ovules to an Agrobacterium suspension followed by pollination, fertilization, and seed ripening. Once mature seed are collected, they are grown on a selection medium to identify transformants. During the summer of 2003, a floral dip experiment was set up to see i f the published protocol (Clough, 1998) was suitable for Exacum. The experiment was designed to determine the appropriate flower-developmental stage for infection, as well as suitability of the Agrobacterium strain and the selectable marker system. 104 AII.t.2 Material and Methods AII.1.2.1. Plant Material Three plants of each Exacum genotype were used; they were grown at the U B C Horticultural Greenhouse. A l l buds and flowers were classified as one of four developmental stages: Stage 1, bud was closed with the petals still green; Stage 2, bud was semi-open with the petal color changed to light blue; Stage 3, the corolla was completely open but the anthers were immature; Stage 4, the corolla was completely open and anthers mature. A total of 68 genotype 01-9-1 flowers were used, 161 genotype 01-37-61 flowers, 51 genotype 01-50-46 flowers and 54 genotype 01-47-49 flowers. A l l flowers were emasculated, and petals removed (to open wounds for bacterial infection) prior to treatment. Flowers were dipped into the suspension for 10 seconds followed by enclosure in plastic bags for one day to maintain high humidity. A l l flowers were pollinated with a common pollen source (genotype 01-09-01) when the stigmatic surface was receptive (approximately 3 days following first dip for flowers in stage # 4). The plants were re-treated with a fresh Agrobacterium suspension five days after the first treatment. The last pollination was done 15 days after the first flower dip. Fruit harvest started 25 days after the first pollination. Genotype 01-50-46 did not produce seeds, suggesting either some kind of incompatibility system with the pollen parent 01-9-1, ovule non-viability, or early embryo abortion. The seeds from all other genotypes were disinfected and plated on either kanamycin selection medium or medium without antibiotic as a control. 105 AII.1.2.2 Treatments The experiment was conducted using a construct that contained the G U S reporter gene, a kanamycin resistance gene (obtained from the pRD400 binary vector), with both genes being driven by the 35S promoter. The Agrobacterium strain used was GV3101 that had gentamycin and rifamycin resistance genes. A n Agrobacterium suspension was produced by growing the bacterium overnight in L B medium + 25mg 1"' o f rifamycin, 25mg l " 1 gentamycin and 50mg 1"' of kanamycin. The culture was then centrifuged at 4000 rpm for 20 min (Beckman Centrifuge Model g2-21) and the resulting pellet resuspended in 5% sucrose solution plus 2 ml 1"1 of Tween 20. The solution had an OD600 of 0.8. AII.1.3 Results and Discussion Germination percentages are shown in Table A . l for all genotypes. The code (example 3.3.2) used represents the stage of the flower, the plant from which the seed was collected and the different dates of pollination, respectively. Although seeds from all genotypes germinated on a kanamycin containing medium, they did not develop root systems. In addition, these seedlings required three transfers to fresh medium before detectable differences in leaf chlorosis between the controls and treated seedlings were observed. Five weeks after germination, putative transformants were tested for G U S activity, however no G U S staining was observed. The seedlings were maintained on antibiotic medium for an additional three months and retested for G U S activity. Again, no G U S activity was detected. 106 Table A.l Percentages of germination of Exacum genotypes on kanamycin medium. Genotype Code % Germination Genotype Code % Germination 01-37-61 3.3.2 65 01-47-49 4.1.1 43 01-37-61 3.3.2 53 01-47-49 1.3.4 37 01-37-61 1.3.3 63 01-47-49 4.1.1 51 01-37-61 3.3.1 39 01-47-49 3.2.1 42 01-37-61 1.3.3 41 01-47-49 Control 92 01-37-61 4.3.1 46 01-09-01 1.2.3 28 01-37-61 2.1.2 46 01-09-01 2.1.3 44 01-37-61 2.1.2 64 01-09-01 3.2.1 46 01-37-61 1.3.3 32 01-09-01 4.3.2 39 01-37-61 Control 92 01-09-01 4.1.1 41 01-47-49 3.2.1 39 01-09-01 Control 90 Based on these preliminary results, the floral dip method does not appear suitable for Exacum. However, several factors may have prevented accurate evaluation of this method and require further investigation. These include: 1) the use of a non-suitable strain of Agrobacterium, 2) G U S staining was too low for detection, and 3) kanamycin pressure did not allow for clear delineation between transgenic and wild type seedlings. AII.2 Co-cultivation AII.2.1 Introduction Currently, the most widely used method for transferring genes into dicotyledonous plants is by Agrobacterium-mediated transformation through co-cultivation of explants with an Agrobacterium suspension. This method is typically done 107 on small excised portions of plant that once transformed, are regenerated into a whole plant through organogenesis of callus or through direct shoot formation. The success of transgene insertion is measured by growing on a selectable medium. The co-cultivation method is only suitable i f we can establish the bacterial strains specificity for the system, the appropriate time and conditions for infection, and the optimum composition of the selection medium. AII.2.2 Materials and Methods AII.2.2.1 Plant Material Petunia specific transformation protocols and materials (seed and vector) were used for this experiment (modified from Jorgensen et al. 1996), and were supplied by Dr. David Clark (University of Florida). However, due to logistic difficulties, the construct was lost. Therefore, I used the same Agrobacterium strain and construct used for the floral dip method. A l l plants were produced in the U B C Centre for Plant Research tissue culture laboratory. Genotype 01-37-37 and the petunia explants were used with 4-5 mm diameter pieces co-cultivated in the Agrobacterium suspension. Two co-cultivation times were evaluated 10 and 15 min. Then Exacum explants were blotted dry with sterile filter paper and immediately placed on M S medium containing 2 mg 1"' B A and 0.1 mg 1"' N A A and placed in the dark for two days at 23° C. Petunia explants were plated on M S medium containing T D Z (1 mg 1"'). Following two days of dark co-cultivation, the explants were transferred to M S medium containing kanamycin (150 mg 1"') + carbenicillin (500 mg l"1) with either 2mg l " 1 B A + 0.1 mg l " 1 N A A for Exacum, or 1 mg 1" 1 T D Z for petunia. Explants were incubated in the dark for an additional two weeks. 108 Following these two weeks of culture, explants were transferred a second time to the same selection medium but without growth regulators. AII.2.2.2 Treatments The construct used contained the reporter gene G U S , a kanamycin resistance gene (obtained from the pRD400 binary vector), with both genes being driven by the 35S promoter. The Agrobaclerium strain used was GV3101 that is gentamycin and rifamycin resistant. Bacteria were grown overnight in L B medium + 25mg 1"' of rifamycin, 25mg 1" 1 gentamycin and 50mg 1"' of kanamycin. The OD6oo of the solution was adjusted to 0.1 with L B medium. AII.2.3 Results and Discussion After three weeks in culture, petunia explants had initiated calli formation but no shoot formation was observed. However, after 4 weeks, petunia explants became necrotic and died. Within the same time frame, all Exacum explants became necrotic and did not produce any calli or shoots. The failure to recover transgenic plants could be due to several factors including inappropriate protocols for Exacum, the Agrobacterium strain was not suitable for the system or plant, or the construct employed did not contain an efficient selectable marker for use with Exacum. AI.4 Literature Cited 1. Clough, S. and Bent, A . 1998. Floral dip: a simplified method for Agr abaci erium-mediated transformation of Arabidopsis thaliana. Plant J. 16:735-43. 2. Jorgensen, R. and Napoli , C. 1996. A responsive regulatory is revealed by sense suppression of pigment genes in Petunia flower. In Gustafson, J. and Flavell , R. (Ed.). Genomes of Plants and Animals: Proceedings of the 22nd Stadler Genetics Symposium. New York: Plenium Press, pp. 159-176. 3. Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15: 473-497. 4. Riseman, A . and Chennareddy, S. 2004. Genotypic variation in the micropropagation of Sri Lankan Exacum hybrids. J. Amer. Soc. Hort. Sci. 129: 698-703.