SELECTIVE PLANT GROWTH USING D-AMINO ACIDS By MASSIMO BOSACCHI A thesis submitted to the Graduate School - New Brunswick Rutgers, The State University of New Jersey in partial fulfillment of the requirements for the degree of Master of Science Graduate Program in Plant Biology Written under the direction of Dr. Randall Kerstetter and approved by ___________________________________ ___________________________________ __________________________________ New Brunswick, New Jersey May 2008 ABSTRACT OF THE THESIS Selective Plant Growth Using D-Amino Acids By Massimo Bosacchi Thesis Director: Dr. Randall Kerstetter Selectable marker genes are essential for the efficient selection of transgenic plants. Heterologous genes used as markers typically encode enzymes that neutralize a toxic compound, allowing for positive selection of plants containing the marker. We report on the efficacy of a novel marker gene system that exploits the varying phytotoxicity of D-amino acids and their oxidative deamination products. Our investigation of a putative DAAO gene from Schizosaccharomyces pombe reveals sufficient oxidative deaminase activity to confer D-alanine and D-serine tolerance to transgenic Arabidopsis thaliana plants carrying the heterologous spDAAO gene. We have demonstrated that the spDAAO, when used in conjunction with D-alanine, allows for the positive selection of primary transformants at levels comparable to hygromycin. Additionally, our selection scheme carries the potential for negative selection in the presence of different selection substrates, such as D- valine or D-isoleucine. This attribute, known as conditional selection would provide particular utility to applications involving site-specific recombinase mediated marker gene removal. ii Table of Contents Abstract ............................................................................................................... ii List of Figures ...................................................................................................... iv Introduction ............................................................................................................... 1 Results .................................................................................................................. 10 Discussion ................................................................................................................. 14 Materials and Methods ......................................................................................... 18 Figures ................................................................................................................. 21 Table ................................................................................................................. 28 Appendix 1 - Plant Transformation Protocol ..................................................... 29 Appendix 2 - Cloning Steps ............................................................................. 32 Appendix 3 - Mammalian and Microbial DAAO ................................................ 36 Appendix 4 - Static Test of Lemna minor .............................................................. 38 References ................................................................................................................. 42 iii List of Figures and Tables Figure 1 The oxidative deamination of D-amino acids ................................ 21 Figure 2 Characterization of the S.pombe DAAO ....................................... 22 Figure 3 Establishing D-alanine and D-serine as selective ubstrates ............. 23 Figure 4 Vectors ....................................................................................... 24 Figure 5 Selection Analysis ........................................................................... 25 Figure 6 GUS staining ........................................................................... 26 Figure 7 Visible color change in titanium test solutions ............................... 27 Table 1 Quantifying peroxide levels in DAAO plants ............................... 28 iv 1 Introduction Selectable Marker Genes In Plant Transformation. Selectable markers in plant transformation confer antibiotic or herbicide resistance to transformed cells, which develop into readily identifiable transgenic plants. Selection pressure is generally applied in regeneration media. The antibiotics and herbicides used in conjunction with marker genes typically impart negative selection, resulting in the abnormal or arrested development of untransformed plants. As part of any plant transformation cassette, selectable marker genes are integrated, along with the transgene of interest, into the native DNA of the plant cell. (1) Marker genes used in positive selection express heterologous proteins that inactivate or degrade the antibiotic or herbicidal selective agent. A limited number of bacterial genes are well established as effective selectable markers for the positive selection of transgenic plants. A 2002 evaluation of several peer-reviewed plant journals found that over 90% of the studies employed selection on kanamycin, hygromycin, or phosphinothricin. (1) Kanamycin and hygromycin B are antibiotics that target prokaryotic mechanisms, giving them a dual role in medicine and biotechnology. (2) They are both therapeutic antibiotics used to combat bacterial infection, and their effect on the translation machinery of chloroplasts and mitochondria make them valuable selection agents against plants. Kanamycin was first isolated from the soil bacterium Streptomyces kanamyceticus. It is one of several aminoglycoside compounds which block prokaryotic protein synthesis by binding the 30S ribosomal subunit. Aminoglycoside activity is 2 deactivated via phosphorylation by neomycin phosphotransferase II (NPTII). NPTII activity is expressed in plants from the nptII gene native to Escherichia coli. (3) In addition to kanamycin, other commonly used aminoglycoside antibiotics include gentamycin, neomycin, and streptomycin. Hygromycin B is an antibiotic of the aminocyclitol group, and severely inhibits cell expansion in developing plants, resulting in the reduced hypocotyl of wild-type seedlings relative to unselected controls. (4) Hygromycin phosphotransferase activity from the E.coli gene aphIV restores hypocotyl growth, enabling selection in plant transformation. (5) Hygromycin B has been shown to induce apoptosis in human cancer cells, and has some limited clinical use as a chemotherapeutic agent (6). Phosphinothricin (PPT) is the ammonium salt of glufosinate, a glutamine analog. PPT has a broad-spectrum herbicidal effect, targeting nitrogen assimilation in plants through its inhibition of glutamine synthase. Acetylation of PPT converts it into a non- toxic form. PPT acetyltransferase activity is conferred to the plant from the bar gene of Streptomyces hygroscopicus. (7) Negative selection schemes allow for the positive selection of cells lacking a particular gene. They are facilitated by the expression of counter-selectable marker genes, which results in cell death or abnormal plant morphology. The cytosine deaminase gene codA from E.coli provides negative selection when used in conjunction with 5-fluorocytosine (5-FC), a non-toxic substrate. The functional expression of codA from the plant converts 5-FC into 5-fluorouracil, whose toxicity is sufficiently lethal to provide counter-selection. (8) Efficient negative selection using the codA counter- selectable gene has been demonstrated in monocots and dicots (9). 3 Issues With Plant Selectable Markers. The most commonly used selectable marker systems have proven to be reliable and efficient indicators of transformation events. However, there are theoretical problems related to the safety and environmental impact of the selective agents used, and concerns associated with marker genes remaining in the transgenic plant. (1) The propagation of resistant bacterial strains is an inherent risk coupled with any antibiotic usage. The antibiotics commonly used in a laboratory setting to identify transformation events also play therapeutic roles in combating human bacterial infection. (10) Hygromycin B requires particular caution owing to its demonstrated apoptotic effects on mammalian cells. (11) The presence of selectable marker genes in the ecosystem carry some risk of the antibiotic or herbicide resistance trait spreading beyond the host. Transgenic pollen may escape to other fields, contaminating their wild-type counterparts. Interspecific hybridization may result in the expression of broad-spectrum herbicide resistance in weed species. (12) Antibiotic resistance could also be transferred to microbial or mammalian cells by means of horizontal gene transfer. In a study using transgenic tobacco, the relocation to bacteria of an antibiotic resistance gene was accomplished, albeit with homologous recombination facilitated. (13) In another study, plasmid DNA fed to mice turned out to be detectable within their cells. (14) Therefore, the possibility of genetic material moving from the plant into soil-borne pathogens or intestinal microorganisms cannot be excluded. Moreover, the protein products expressed from selectable marker genes may in themselves cause an allergic reaction to the consumer. (15) 4 D-Amino Acids. With the exception of glycine, each amino acid exists in one of two stereoisomeric forms based on the position of side groups relative to the chiral carbon. The two forms have different optical properties, and are accordingly designated levorotatory (L) or dextrorotatory (D). L-amino acids play the principal role in metabolism and protein composition of all organisms. The biochemical pathways responsible for the synthesis of L-amino acids involve the transfer of an amino group onto a 2-oxocarboxylic intermediate. This reaction, catalyzed by aminotransferases, exists in all kingdoms and is highly stereospecific. (16) D-amino acids have a much more limited occurrence in nature than their L stereoisomers. D-amino acids appear to have some structural and signaling functions in both prokaryotes and eukaryotes. D-alanine and D-glutamate are major components of the peptidoglycan layer in bacterial cell walls. (17) D-serine is present at high levels in the human brain, where it functions as a neurotransmitter. (18) Metabolic activity involving D-amino acids has not been observed in the plant kingdom. (19) However, their presence has been detected in higher plants like peas, barley, hops, and tobacco, where they exist in amounts 0.5% to 3% relative to L-amino acids. (20) The presence of D-amino acids in plants is attributable to racemase activity or the uptake of microbial D-amino acids from the soil. (21) Amino acid racemases catalyze the inversion of L-amino acids to form the D enantiomer, or vice versa. An alanine racemase detected in Medicago sativa (Alfalfa) seedlings was the first such enzyme to be established in plants. (22) A serine racemase gene was recently identified and characterized in Arabidopsis thaliana. (23) The biological function of amino acid 5 racemases has not been determined in plants. Some bacterial racemases have been determined to be involved in the synthesis of peptidoglycans or poly-γ-glutamate, neither compound having an identified role in plants. (24) In plants, it is unclear why a biosynthetic pathway for D-amino acids exists. Little is known about what, if any function D-amino acids serve. They are unusable as protein building blocks, and are not subject to any known enzymatic activity that could release the amine group as a usable form of nitrogen. (25) Plant cells conjugate limited amounts of free D-amino acids into N-malonyl and N-acetyl derivatives, which are then compartmentalized in the vacuole. (26) Since some D-amino acids are known to bring about negative effects on plant growth and development (27), compartmentalization may be a means to offset toxicity. In any case, the existence of a mechanism to isolate D- amino acids suggests to us that plants are unable to make use of any D enantiomer. It has been shown that Arabidopsis thaliana seeds sown on media containing 3mM D-serine or 3mM D-alanine arrest development shortly after germination. (19) Several other plant species have demonstrated D-serine sensitivity at similar concentrations. (28) D-amino acids may interfere with endogenous protein synthesis, but the mechanism of their toxicity has not yet been elucidated. It has been hypothesized that D-serine competes with endogenous β-alanine in the synthesis of pantothenic acid (29) In an earlier study, however, incubation with exogenous β-alanine and pantothenic acid failed to rescue wild type A. thaliana seedlings from D-serine pressure. (19) D-Amino Acid Oxidase. The oxidative deamination of D-amino acids is a FAD- dependent substitution of the amine group for a ketone group. In a DAAO catalyzed reaction, the D-amino acid is converted to an imino acid intermediate, reducing FAD in 6 the process. FAD is re-oxidized by molecular oxygen, releasing hydrogen peroxide. Subsequent hydrolysis of the imino acid forms an α-keto acid along with ammonia. (30) (Fig.1) The enzymatic activity of the DAAO flavoenzyme is highly specific to the D- configuration, and favors neutral amino acids. All non-plant organisms exhibit DAAO activity, which may play a role in detoxifying endogenous or environmental D-amino acids. (31) Absent DAAO activity, plants may not be able to metabolize D-amino acids beyond a certain threshold. The basis for the varying phytotoxicity of different D-amino acids and their oxidative deamination products is unclear. (32) Catastrophic effects on plant development are observed in the presence of excess D-alanine and D-serine, and not their corresponding α-keto acids. (19) In contrast, D-valine and D-isoleucine have minimal effects on plant development, but are oxidatively deaminated into phytotoxic compounds. These compounds, 3-methyl-2-oxopentanoate and 3-methyl-2- oxobutanoate, completely inhibit plant growth at respective concentrations of 1mM and 5mM. (19) This dichotomy forms the basis of conditional selection using a plant oxidative deaminase transgene. D-amino acid oxidase activity was first conferred to plants by the dao1 gene from Rhodotorula gracilis. D-alanine and D-serine tolerance was observed in Arabidopsis thaliana plants expressing dao1 from the CaMV 35S promoter. Enzymatic rgDAAO activity was quantified by incubating crude protein extracts from transgenic plants with D-alanine, and assaying for increased absorbance associated with its conversion to pyruvate. Although a range of dao1 expression levels was established, phenotypes of the corresponding plants were uniform, indicating complete D-alanine and D-serine tolerance