Investigating the Roles of Diterpenoids in Rice-Xanthomonas Oryzae Interactions" (2015)

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2015 Investigating the roles of diterpenoids in rice- Xanthomonas oryzae interactions Xuan Lu Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Agriculture Commons, and the Plant Sciences Commons

Recommended Citation Lu, Xuan, "Investigating the roles of diterpenoids in rice-Xanthomonas oryzae interactions" (2015). Graduate Theses and Dissertations. 14912. https://lib.dr.iastate.edu/etd/14912

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

Investigating the roles of diterpenoids in rice-Xanthomonas oryzae interactions

Xuan Lu

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Plant Biology

Program of Study Committee: Reuben J. Peters, Major Professor Gwyn Beattie Gustavo MacIntosh Alan Myers Bing Yang

Iowa State University

Ames, Iowa

2015

TABLE OF CONTENTS

Page

ABSTRACT ...... iv

CHAPTER 1 GENERAL INTRODUCTION ...... 1 Rice–Xanthomonas oryzae Interactions ...... 1 Ecological Function of Secondary Metabolites ...... 2 Terpenoids ...... 2 Gibberellin ...... 4 Rice Phytoalexins ...... 5 Organization of Dissertation and Author Contribution ...... 7 References ...... 8 Figures ...... 12

CHAPTER 2 AN ENT-KAURENE DERIVED DITERPENOID VIRULENCE FACTOR FROM XANTHOMONAS ORYZAE PV. ORYZICOLA ..... 15 Abstract ...... 15 Introduction ...... 16 Results ...... 18 Discussion ...... 24 Materials and Methods ...... 26 Acknowledgements ...... 32 References ...... 32 Figures ...... 35 Supplementary Results ...... 39 Supplementary Table ...... 41

CHAPTER 3 FUNCTIONAL CONSERVATION OF THE CAPACITY FOR ENT-KAURENE BIOSYNTHESIS AND AN ASSOCIATED OPERON IN CERTAIN RHIZOBIA ...... 42 Abstract ...... 42 Introduction ...... 43 Results ...... 45 Discussion ...... 52 Materials and Methods ...... 53 Acknowledgements ...... 57 References ...... 58 Table ...... 62 Figures ...... 62

iii

CHAPTER 4 PHYSIOLOGICAL FUNCTION AND METABOLITES LEVEL OF DITERPENOIDS IN RICE DEFENSE ...... 65 Abstract ...... 65 Introduction ...... 66 Results ...... 70 Discussion ...... 79 Materials and Methods ...... 80 References ...... 85 Figures ...... 89 Supplemental Results ...... 96

CHAPTER 5 INVESTIGATING THE BIOLOGICAL FUNCTION OF PHYTOCASSANES IN RICE DEFENSE ...... 99 Abstract ...... 99 Introduction ...... 99 Results ...... 101 Discussion ...... 103 Materials and Methods ...... 104 Acknowledgements ...... 106 References ...... 106 Figures ...... 107

CHAPTER 6 EXPLORATION OF ORYZALIDE-RELATED DITERPENOIDS IN RICE DEFENSE WITH GENOME EDITING TECHNOLOGY ...... 111 Abstract ...... 111 Introduction ...... 111 Results ...... 113 Discussion ...... 116 Materials and Methods ...... 117 Acknowledgements ...... 118 References ...... 119 Figures ...... 120

CHAPTER 7 CONCLUSIONS AND FUTURE DIRECTIONS ...... 124 Conclusions ...... 124 Future Directions ...... 125 References ...... 128 Figures ...... 129

ACKNOWLEDGMENTS ...... 130

ABSTRACT

Secondary metabolites play a significant role in mediating plant-microbe interactions. Rice (Oryza sativa) is one of the most important staple crops. Once rice is attacked by microbial pathogens, it will produce phytoalexins antibiotics to defend against enemies such as the bacterial pathogen Xanthomonas oryzae. Diterpenoids provide a rich source of metabolites mediating rice-X. oryzae interaction. First, gibberellin (GA), as complex diterpenoid phytohormone, besides the profound effects on plant growth and development, it has been shown to negatively regulate rice defense.

Interestingly, we found a putative GA biosynthetic gene operon in rice bacterial leaf streak pathogen Xanthomonas oryzae pv. oryzicola (Xoc). Based on this, we biochemically characterized Xoc for its capacity to produce the precursor of GA. With genome mining, we found three other genera of rhizobia contained homologous GA biosynthetic gene operon and shared the identical biochemical function for production of ent-kaurene. As for Xoc, we detected the putative production of GA was relevant to virulence ability in Xoc through antagonism to jasmonic acid. The role of putative GA diterpenoid regulated Xoc’s association with rice. Second, rice produces an arsenal of phytoalexins as responses to fungal and bacterial infection. Most of the known rice phytoalexins are diterpenoids. The relevance of these diterpenoids to defense was evaluated based on their antibiotic activity in vitro; the physiological role in planta remains undefined. Rice diterpenoids phytoalexins biosynthesis proceeds ent-copalyl diphosphate synthase (OsCPS2) or syn-copalyl diphosphate synthase (OsCPS4) from diterpenoid precursor geranylgeranyl diphosphate (GGPP). With reverse genetics approach we found the relevance of rice diterpenoids to defense depends on OsCPS2

v pathway instead of OsCPS4 pathway. Furthermore, with application of CRISPR/Cas9 genome editing technology, I not only explored the relevance of the specific diterpenoid pathway downstream of OsCPS2 to rice plant defense but identified the redirection of metabolites levels in rice specialized metabolism as well. These results will not only elucidate some insight into the metabolites balance in rice-X. oryzae interaction but also provide the potential agricultural application for molecular breeding.

CHAPTER 1: GENERAL INTRODUCTION

Rice–Xanthomonas oryzae Interactions

Rice, as a staple model crop, feeds around two-thirds of the whole population in the world. Moreover, rice contains around 20% caloric intake for human (FAO, 2005).

However, its yields are susceptible to microbial pathogens both above and below the ground. The most devastating fungal disease of rice is leaf blast disease caused pathogen

Magnaporthe oryzae (M. oryzae), which was estimated to reduce the global rice harvest by 10-30% (Talbot, 2003). The most devastating bacterial disease is leaf blight disease caused by Xanthomonas oryzae pv. oryzae (Xoo) (Nino-Liu et al., 2006). Xanthomonas oryzae pv. oryzicola (Xoc) is another devastating rice bacterial pathogen that causes bacterial leaf streak disease. After entering plant leaves, Xoc colonizes parenchyma tissue, causing lesions that appear water soaked initially and then develop into translucent, bacterial leaf streaks (Wang et al., 2007). One of the major disease differences between Xoo and Xoc is that Xoo invades the plants through the vascular tissue while Xoc invades through parenchyma tissue.

Rice produces a variety of diterpenoids phytochemicals as arsenals to defense against microbial pathogens. These compounds were originally identified by their activity against M. oryzae and Xoo (Peters, 2006). On the other hand, bacteria produce secondary metabolites to help infect plants, these compounds play critical roles in toxins, antibiotics, pigments (Demain, 1998). As for X. oryzae, there are very few reports about the secondary metabolites from it. We elucidate the diterpenoid gibberellin produced from

Xoc.

Ecological Functions of Secondary Metabolites

Due to the sessile nature of plants, they need to produce large amounts of metabolites to respond to biotic and abiotic stresses. These metabolites can be classified as primary and secondary metabolites. Primary metabolites are the essential metabolites required for life. Not just carbohydrates, lipids, proteins, and nucleic acid, they also include other compounds such as plant hormone gibberellin, which functions in fundamental physiological processes in plant growth and development. Unlike primary metabolites, secondary metabolites are the metabolites not directly involved in normal growth, development or reproduction (Fraenkel, 1959), but they play major roles in the ecology, including plant-microbe and plant-herbivore interactions. From the structural classification, secondary metabolites include terpenoids, alkaloids, phenylpropanoids, and polyketides. During plant-microbe symbiotic process, leguminous plants secrete flavonoids to initiate nodule formation and nitrogen fixation (Straight and Kolter, 2009).

Maize produces terpenoids to exhibit both anti-fungal and anti-insect activities (Schmelz et al., 2011). Besides the direct defense, maize uses an indirect strategy of defense, such as producing terpenoids to attract enemies of herbivores. The insect-induced terpenoid

(E)-β-caryophyllene emitted from maize roots to attracts entomopathogenic nematodes, which prey on the attacking insect larvae (Rasmann et al., 2005; Degenhardt et al., 2009).

Terpenoids

Terpenoids (also called isoprenoids) are the largest class in secondary metabolites. They play a significant role in plant-insect, plant-pathogen, and plant-plant interactions. Because of the structural diversity, the biological functions of terpenoids

3 vary tremendously. For example, carotenoids function in photosynthetic pigments, which are rich in carrots for bright color. Perfume industry takes advantage of limonene for the use of its odor of oranges. Moreover, terpenoids can be used for pharmaceuticals, such as star molecule artemisinin (anti-malaria) extracted from Artemisia annua and taxol (anti- cancer) from the bark of Pacific yew.

Biosynthesis of terpenoids is derived from five-carbon isoprene unit isopentenyl diphosphate (IPP) and its allylic isomer dimethylallyl diphosphate (DMAPP) (Moss,

1995). Mevalonate (MEV) pathway and 2C-methyl-D-erythritol 4-phosphate (MEP) pathway are the two primary carbon sources from primary metabolites for terpenoid biosynthesis. MEV pathway initiates in the cytosol of eukaryotes while MEP pathway is in prokaryotes and plastid of plants (Dudareva et al., 2013).

Through the prenyltransferase, isoprene units are provided by IPP and DMAPP.

These two units are condensed in head to tail fashion. Two isoprene units can link together for the formation of monoterpene (10C), three for sesquiterpene (15C) and four for diterpene (20C) geranylgeranyl diphosphate (GGPP) (Fig. 1). Notably, triterpene

(30C) and tetraterpene (40C) are condensed with two FPPs and GGPPs respectively in head to head fashion. Followed by the condensation of linear terpenes, terpene synthases catalyze to generate cyclic carbon skeleton. Thousands of terpenoids are decorated on the carbon skeleton by redox reactions catalyzed by cytochromes P450 (CYPs) or other oxygenases for the production of final bioactive terpenoids. My research focuses on the biological functions of diterpenoids.

Gibberellin

One of the remarkable compounds in diterpenoids is the phytohormone gibberellins. They are produced by plants and associated microbes. Bioactive GA1, GA3,

GA4 and GA7 are required for plant growth and development, including seed germination, stem elongation and flower induction. Furthermore, GA has pushed forwarded the Green Revolution (Hedden, 2003).

Interestingly, GAs are also relevant to plant defense. GAs actually were first isolated from the fungal pathogen Gibberella fujikuroi (Fusarium moniliforme), which causes the foolish seedling disease in rice, with its characteristic excessive growth phenotype (Yabuta and Sumiki, 1938). In rice, GA has been shown to negatively regulate resistance to both the bacterial blight pathogen Xoo and fungal blast pathogen M. oryzae

(Yang et al., 2008; Qin et al., 2013). Moreover, the production of gibberellin by G. fujikuroi has been recently reported to play a role in the virulence of this rice plant pathogen as well (Wiemann et al., 2013). While best known for their role in promoting plant growth, GAs have more recently been recognized as antagonists of the plant defense response mediated by jasmonic acid (JA) and ethylene (Bari and Jones, 2009).

Not only found in plants and fungi, GA is also produced by plant-associated bacteria (Bottini et al., 2004), particularly nitrogen fixing rhizobacteria. Bradyrhizobium japonicum (B. japonicum) is a legume-nodulating and nitrogen-fixing bacterium species.

It has been shown that B. japonicum bacteroids have the capacity to convert GA precursors to GA9 (Mendez et al., 2014). In the genome of B. japonicum USDA110, an operon consisting of genes putatively encoding a ferredoxin, a short chain dehydrogenase, three P450s, a GGPP synthase, and two diterpene synthases, have been

5 proposed to be involved in GA biosynthesis (Tully et al., 1998). Our group cloned and characterized diterpene synthase genes encoding ent-copalyl diphosphate synthase (CPS) and ent-kaurene synthase (KS), respectively from B. japonicum (Morrone et al., 2009)

(Fig. 2).

Rice Phytoalexin

Phytoalexins are the small molecular weight antimicrobial compounds, and they are inducible as responses to attack from pathogens. Those constitutively present are termed phytoanticipans (VanEtten et al., 1994). Besides sakuranetin classified as a flavonoid, all other known rice phytoalexins are diterpenoids (Schmelz et al., 2014) (Fig.

3). From the chemical structure, these diterpenoids are in the subfamily of labdane- related diterpenoids (LRD), featuring two cyclization reactions derived from GGPP

(Peters, 2010). Three groups of LRD phytoalexins were isolated and identified by their activity against the fungus M. oryzae: momilactones A&B, phytocassanes A-F and oryzalexins A-F (Peters, 2006), and GA is actually the founding member of rice LRDs.

The remaining LRD phytoalexins include oryzalide A and B, oryzalic acid A and B, and oryzadione I to III (Toyomasu, 2008). All of these compounds, termed oryzalide-related here, were isolated and identified by their activity against the bacterial leaf blight pathogen Xoo. Oryzalide-related LRD seems to act as phytoanticipans because they are only weakly induced by Xoo infection (Watanabe et al., 1996). Besides microbial attack, phytoalexins can be induced by abiotic stresses such as UV irradiation, jasmonic acid and

CuCl2 treatment (Yamane, 2013).

GA biosynthetic pathway serves as a genetic reservoir for the evolution of more specialized LRD biosynthesis. Firstly, from the diterpenoid common precursor GGPP, it is catalyzed by ent-copalyl diphosphate (CPP) synthases (Os-CPS). Os-CPS homologous gene Os-CPS2 and Os-CPS4 catalyze GGPP to cyclized ent-CPP and its diastereomer syn-CPP respectively (Otomo et al., 2004; Prisic et al., 2004; Xu et al., 2004). In GA biosynthetic pathway, ent-kaurene synthase (Os-KS) converts ent-CPP to ent-kaurene.

Derived from this, kaurene synthase-like (Os-KSL) catalyzes the further cyclization to diterpene olefins, which are the precursors of LRDs (Fig. 4). Followed by that, downstream cytochrome P450s and short-chain dehydrogenases act in decorating for phytoalexin biosynthesis.

Some of LRD biosynthetic gene Os-CPSs, Os-KSLs, and CYP450s are co- clustered together in the genome (Fig. 5). One gene cluster involved in momilactone biosynthesis is located on chromosome 4, which contains Os-CPS4, Os-KSL4, two soluble short-chain alcohol dehydrogenases and two P450s (Shimura et al., 2007). The other gene cluster is located on chromosome 2, which includes Os-CPS2, Os-KSL5-7 and six P450s (Swaminathan et al., 2009), which presumably initiate phytocassanes biosynthesis.

Although these diterpenoid phytoalexins have been isolated, the relevance of rice

LRD phytoalexins to plant defense was identified on the basis of their antibiotic activity in vitro. For instance, the inhibitory activity of momilactones against M. oryzae depends on the concentration of momilactones (Hasegawa et al., 2010). Although momilactones are induced specifically at the site of Xoo infection in rice leaves (Klein et al., 2015), their role in anti-microbial activity was uncertain due to their chemical complexity in

7 planta. To elucidate the direct role for the momilactones in planta, we took reverse genetics approach using an Os-CPS4 knock-out (Os-cps4). Unexpectedly, Os-cps4 plants are not more susceptible to rice M. oryzae than parental wild type plants. Furthermore, the additional analysis indicates that momilactones act as allelochemicals in planta (Xu et al., 2012).

Organization of Dissertation and Author Contribution

After the introduction, Chapter 2 elucidates the biochemical and physiological functions of putative GA production in Xoc. We found putative GA biosynthetic operon in the genome of the bacterial pathogenic bacteria Xoc BLS256. Interestingly, putative

GA functions as the virulence factor in Xoc. This project was in cooperation with Dr.

Adam Bogdanove and has been published on New Phytologist. Chapter 3 was released in

Journal of Bacteriology, which was initiated by the previous lab member David Hershey, and I served the co-first author to finish up this project. Inspired by that GA biosynthetic operon was found in B. japonicum, homologous gene cluster GA biosynthetic operons were found in three other plant-associated in other species of rhizobia and shared the same biochemical function encoding diterpene synthases.

The rest chapters present the biological function and metabolites of diterpenoids from the side of rice. Chapter 4 focuses on the relevance of Os-CPS2 dependent diterpenoids pathways to rice plant defense and provides insight into their metabolic fluxes. Followed by the Os-CPS2 pathway, the next two Chapters are incomplete stories about more specific pathways downstream the Os-CPS2 pathway. Chapter 5 explores the relevance of phytocassanes to rice defense against pathogenic fungi and bacteria. With

8 the advent of genome editing CRISPR/Cas9 technology, Chapter 6 emphasizes on the biological activity of oryzalide/ oryzadione related LRDs. Moreover, this story indicates the additive effect on the production of phytoalexins in rice. I mentored two rotation students Hao Jiang, Tibebe Teklemariam and one undergraduate student Ben Brown, who contributed to identification the Os-KSL5 and Os-KSL6 CRISPR/Cas9 mutant plants in this Chapter. The last Chapter discusses the conclusions and future directions for this dissertation.

References:

Bari, R., and Jones, J.D. (2009). Role of plant hormones in plant defence responses. Plant Molecular Biology 69, 473-488.

Bottini, R., Cassan, F., and Piccoli, P. (2004). Gibberellin production by bacteria and its involvement in plant growth promotion and yield increase. Applied microbiology and biotechnology 65, 497-503.

Degenhardt, J., Hiltpold, I., Kollner, T.G., Frey, M., Gierl, A., Gershenzon, J., Hibbard, B.E., Ellersieck, M.R., and Turlings, T.C. (2009). Restoring a maize root signal that attracts insect-killing nematodes to control a major pest. Proceedings of the National Academy of Sciences of the United States of America 106, 13213-13218.

Demain, A.L. (1998). Induction of microbial secondary metabolism. Int Microbiol 1, 259-264.

Dudareva, N., Klempien, A., Muhlemann, J.K., and Kaplan, I. (2013). Biosynthesis, function and metabolic engineering of plant volatile organic compounds. The New phytologist 198, 16-32.

FAO. (2005). Summary of world food and agricultural statistics 2005. Rome: Food and Agriculture Organization of the United Nations,, 64.

Fraenkel, G.S. (1959). The raison d'etre of secondary plant substances; these odd chemicals arose as a means of protecting plants from insects and now guide insects to food. Science 129, 1466-1470.

Hasegawa, M., Mitsuhara, I., Seo, S., Imai, T., Koga, J., Okada, K., Yamane, H., and Ohashi, Y. (2010). Phytoalexin accumulation in the interaction between rice and the blast fungus. Molecular Plant-Microbe Interactions 23, 1000-1011.

Hedden, P. (2003). The genes of the Green Revolution. Trends in genetics : TIG 19, 5-9.

Klein, A.T., Yagnik, G.B., Hohenstein, J.D., Ji, Z., Zi, J., Reichert, M.D., MacIntosh, G.C., Yang, B., Peters, R.J., Vela, J., and Lee, Y.J. (2015). Investigation of the Chemical Interface in the Soybean-Aphid and Rice-Bacteria Interactions Using MALDI-Mass Spectrometry Imaging. Analytical chemistry 87, 5294-5301.

Mendez, C., Baginsky, C., Hedden, P., Gong, F., Caru, M., and Rojas, M.C. (2014). Gibberellin oxidase activities in Bradyrhizobium japonicum bacteroids. Phytochemistry 98, 101-109.

Morrone, D., Chambers, J., Lowry, L., Kim, G., Anterola, A., Bender, K., and Peters, R.J. (2009). Gibberellin biosynthesis in bacteria: separate ent-copalyl diphosphate and ent-kaurene synthases in Bradyrhizobium japonicum. FEBS letters 583, 475-480.

Moss, G.P., Smith, P.A.S., and Tavernier, D. . (1995). Glossary of Class Names of Organic Compounds and Reactive Intermediates Based on Structure. Pure and Applied Chemistry 67, 1307-1375.

Nino-Liu, D.O., Ronald, P.C., and Bogdanove, A.J. (2006). Xanthomonas oryzae pathovars: model pathogens of a model crop. Molecular plant pathology 7, 303- 324.

Otomo, K., Kenmoku, H., Oikawa, H., Konig, W.A., Toshima, H., Mitsuhashi, W., Yamane, H., Sassa, T., and Toyomasu, T. (2004). Biological functions of ent- and syn-copalyl diphosphate synthases in rice: key enzymes for the branch point of gibberellin and phytoalexin biosynthesis. Plant Journal 39, 886-893.

Peters, R.J. (2006). Uncovering the complex metabolic network underlying diterpenoid phytoalexin biosynthesis in rice and other cereal crop plants. Phytochemistry 67, 2307-2317.

Peters, R.J. (2010). Two rings in them all: The labdane-related diterpenoids. Natural Product Reports 27, 1521-1530.

Prisic, S., Xu, M., Wilderman, P.R., and Peters, R.J. (2004). Rice contains two disparate ent-copalyl diphosphate synthases with distinct metabolic functions. Plant Physiology 136, 4228-4236.

Qin, X., Liu, J.H., Zhao, W.S., Chen, X.J., Guo, Z.J., and Peng, Y.L. (2013). Gibberellin 20-oxidase gene OsGA20ox3 regulates plant stature and disease development in rice. Molecular Plant-Microbe Interactions 26, 227-239.

Rasmann, S., Kollner, T.G., Degenhardt, J., Hiltpold, I., Toepfer, S., Kuhlmann, U., Gershenzon, J., and Turlings, T.C. (2005). Recruitment of entomopathogenic nematodes by insect-damaged maize roots. Nature 434, 732-737.

Schmelz, E.A., Huffaker, A., Sims, J.W., Christensen, S.A., Lu, X., Okada, K., and Peters, R.J. (2014). Biosynthesis, elicitation and roles of monocot terpenoid phytoalexins. Plant Journal 79, 659-678.

Schmelz, E.A., Kaplan, F., Huffaker, A., Dafoe, N.J., Vaughan, M.M., Ni, X., Rocca, J.R., Alborn, H.T., and Teal, P.E. (2011). Identity, regulation, and activity of inducible diterpenoid phytoalexins in maize. Proceedings of the National Academy of Sciences of the United States of America 108, 5455-5460.

Shimura, K., Okada, A., Okada, K., Jikumaru, Y., Ko, K.W., Toyomasu, T., Sassa, T., Hasegawa, M., Kodama, O., Shibuya, N., Koga, J., Nojiri, H., and Yamane, H. (2007). Identification of a biosynthetic gene cluster in rice for momilactones. The Journal of biological chemistry 282, 34013-34018.

Straight, P.D., and Kolter, R. (2009). Interspecies chemical communication in bacterial development. Annual review of microbiology 63, 99-118.

Swaminathan, S., Morrone, D., Wang, Q., Fulton, D.B., and Peters, R.J. (2009). CYP76M7 is an ent-cassadiene C11alpha-hydroxylase defining a second multifunctional diterpenoid biosynthetic gene cluster in rice. The Plant cell 21, 3315-3325.

Talbot, N.J. (2003). On the trail of a cereal killer: Exploring the biology of Magnaporthe grisea. Annual review of microbiology 57, 177-202.

Toyomasu, T. (2008). Recent Advances Regarding Diterpene Cyclase Genes in Higher Plants and Fungi. Bioscience, Biotechnology, and Biochemistry 72, 1168-1175.

Tully, R.E., van Berkum, P., Lovins, K.W., and Keister, D.L. (1998). Identification and sequencing of a cytochrome P450 gene cluster from Bradyrhizobium japonicum. Biochimica et biophysica acta 1398, 243-255.

VanEtten, H.D., Mansfield, J.W., Bailey, J.A., and Farmer, E.E. (1994). Two Classes of Plant Antibiotics: Phytoalexins versus "Phytoanticipins". The Plant cell 6, 1191-1192.

Wang, L., Makino, S., Subedee, A., and Bogdanove, A.J. (2007). Novel candidate virulence factors in rice pathogen Xanthomonas oryzae pv. oryzicola as revealed by mutational analysis. Applied and environmental microbiology 73, 8023-8027.

Watanabe, M., Kono, Y., Esumi, Y., Teraoka, T., Hosokawa, D., Suzuki, Y., Sakurai, A., and Watanabe, M. (1996). Studies on a Quantitative Analysis of Oryzalides and Oryzalic Acids in Rice Plants by GC-SIM. Bioscience, Biotechnology, and Biochemistry 60, 1460-1463.

Wiemann, P., Sieber, C.M., von Bargen, K.W., Studt, L., Niehaus, E.M., Espino, J.J., Huss, K., Michielse, C.B., Albermann, S., Wagner, D., Bergner, S.V., Connolly, L.R., Fischer, A., Reuter, G., Kleigrewe, K., Bald, T., Wingfield, B.D., Ophir, R., Freeman, S., Hippler, M., Smith, K.M., Brown, D.W., Proctor, R.H., Munsterkotter, M., Freitag, M., Humpf, H.U., Guldener, U., and Tudzynski, B. (2013). Deciphering the cryptic genome: genome-wide analyses of the rice pathogen Fusarium fujikuroi reveal complex regulation of secondary metabolism and novel metabolites. PLoS Pathogens 9, e1003475.

Xu, M., Hillwig, M.L., Prisic, S., Coates, R.M., and Peters, R.J. (2004). Functional identification of rice syn-copalyl diphosphate synthase and its role in initiating biosynthesis of diterpenoid phytoalexin/allelopathic natural products. Plant Journal 39, 309-318.

Xu, M., Galhano, R., Wiemann, P., Bueno, E., Tiernan, M., Wu, W., Chung, I.M., Gershenzon, J., Tudzynski, B., Sesma, A., and Peters, R.J. (2012). Genetic evidence for natural product-mediated plant-plant allelopathy in rice (Oryza sativa). The New phytologist 193, 570-575.

Yabuta, T., and Sumiki, Y. (1938). On the crystal of gibberellin, a substance to promote plant growth. Journal of Agricultural and Food Chemistry Soc. Japan 14, 1526.

Yamane, H. (2013). Biosynthesis of phytoalexins and regulatory mechanisms of it in rice. Bioscience, biotechnology, and biochemistry 77, 1141-1148.

Yang, D.L., Li, Q., Deng, Y.W., Lou, Y.G., Wang, M.Y., Zhou, G.X., Zhang, Y.Y., and He, Z.H. (2008). Altered disease development in the eui mutants and Eui overexpressors indicates that gibberellins negatively regulate rice basal disease resistance. Molecular Plant 1, 528-537.

Figures

MEV Pathway MEP Pathway

C 5 IPP isomerase OPP OPP IPP (C ) 5 DMAPP (C5)

OPP

Monoterpenoid C10 GPP

FPP C 15 Sesquiterpenoid

PPO

OPP Diterpenoid C GGPP 20

Figure 1. Biosynthetic pathway of monoterpenoid, sesquiterpenoid and diterpenoid from Mevalonate (MEV) and 2C-methyl-D-erythritol 4-phosphate (MEP) pathways. IPP, isopentenyl diphosphate; DMAPP, dimethylallyl diphosphate; GPP, geranyl diphosphate; GGPP, geranylgeranyl diphosphate.

CYP CYP CYP Fd SDR GGPS CPS KS 112 114 117

Figure 2. Putative GA biosynthesis operon identified from Bradyrhizobium japonicum. CYP, cytochrome P450; Fd, ferredoxin; SDR, short-chain alcohol dehydrogenase/ reductase; GGPS, geranylgeranyl diphosphate synthase; CPS, copalyl diphosphate synthase; KS, ent-kaurene synthase.

O O HO HO HO O O HO O HO O O O O Phytocassane A Phytocassane B Momilactone A Momilactone B O O HO O

HO HO HO O O OH Phytocassane D Oryzalexin A Oryzalexin B Phytocassane C O HO HO

O O HO O OH OH Oryzalexin C Oryzalexin D Phytocassane E Oryzalexin S

O OH O O O HO HO OH O O Oryzalexin E Oryzalexin F Oryzadione A Oryzalides

Figure 3. The known Labdane-Related Diterpenoids phytoalexins from Rice.

OPP OsCPS4 OsCPS1 OPP OPP GGPP OsCPS2

H H syn-CPP OsKSL10 OsKSL5 ent-CPP OsKS1 OsKSL11 OsKSL6 OsKSL8 OsKSL7 OsKSL4 H OsKSL10 syn-labda-8(17), H 12E,14-triene ent-pimara-8(14), H 15-diene H H H ent-isokaurene H H H H H H syn-stemod- syn-pimara- ent-cassa-12, ent-kaurene syn-stemar- 13(17)-ene ent-sandaraco- 15-diene 13-ene 7,15-diene pimaradiene

Momilactones A & B ??? ??? ??? Oryzalides A-C Phytocassanes A-F GA Oryzalexin S Oryzalexins A-F

Figure 4. Rice labdane related diterpenoids biosynthetic pathway. Solid arrows indicate gibberellin biosynthetic pathway; dashed arrows indicate multiple enzymatic reactions.

Figure 5. Scheme of two Labdane related diterpenoid gene clusters from rice genome

CHAPTER 2: AN ENT-KAURENE DERIVED DITERPENOID VIRULENCE

FACTOR FROM XANTHOMONAS ORYZAE PV. ORYZICOLA

A paper published in New Phytologist 2015, 206 (1), 295-302

Xuan Lu1, David M. Hershey1,‡, Li Wang2,‡, Adam J. Bogdanove2,‡, and Reuben J.

Peters1,*

1Department of Biochemistry, Biophysics and Molecular Biology, 2Department of Plant

Pathology and Microbiology, Iowa State University, Ames, IA 50011, USA

Abstract

Both plants and fungi produce ent-kaurene as a precursor to the gibberellin plant hormones. A number of rhizobia contain functionally conserved, sequentially acting ent- copalyl diphosphate and ent-kaurene synthases (CPS and KS, respectively), which are found within a well-conserved operon that may lead to the production of gibberellins.

Intriguingly, the rice bacterial leaf streak pathogen Xanthomonas oryzae pv oryzicola

(Xoc) contains a homologous operon. Here is reported biochemical characterization of the encoded CPS and KS, and the impact of insertional mutagenesis on virulence and the plant defense response for these genes, as well as that for one of the cytochromes P450

(CYP112) found in the operon. Activity of the CPS and KS found in this phytopathogen was verified – i.e., Xoc is capable of producing ent-kaurene. Moreover, knocking-out

CPS, KS, or CYP112 led to mutant Xoc that exhibited reduced virulence. Investigation of the effect on marker gene transcript levels suggests that the Xoc diterpenoid affects the plant defense response, most directly that mediated by jasmonic acid (JA). Xoc produces

16 an ent-kaurene derived diterpenoid as a virulence factor, potentially a gibberellin phytohormone, which is antagonistic to JA, consistent with recent recognition of opposing effects for these phytohormones on the plant microbial defense response.

Introduction

It has long been evident that microbes produce phytohormones to manipulate their interaction with plant hosts (Bari and Jones, 2009). Indeed, the gibberellins were first isolated from the fungal pathogen Gibberella fujikuroi (Fusarium moniliforme), which causes the bakanae or foolish seedling disease in rice (Oryza sativa), with its characteristic excessive growth phenotype (Yabuta and Sumiki, 1938). While best known for their role in promoting plant growth, gibberellins have more recently been recognized as antagonists of the plant defense response mediated by jasmonic acid (JA) and ethylene

(Bari and Jones, 2009). For example, in rice the gibberellins have been shown to negatively regulate resistance to both the bacterial blight pathogen Xanthomonas oryzae pv. oryzae and fungal blast pathogen Magnaporthe oryzae (Yang et al., 2008; Qin et al.,

2013). Moreover, the production of gibberellin by G. fujikuroi has been recently reported to play a role in the virulence of this rice plant pathogen as well (Wiemann et al., 2013).

In addition to G. fujikuroi, it has been found that certain (rhizo)bacteria also produce gibberellins (Bottini et al., 2004). This includes Bradyrhizobium japonicum

USDA 110 (Mendez et al., 2014), whose genome contains an operon consisting of three cytochromes P450 (CYPs), a ferredoxin (Fd), a short-chain alcohol dehydrogenase/reductase (SDR), a putative prenyl transferase, and two putative prenyl cyclases, that have been hypothesized to be involved in the production of gibberellin

(Tully et al., 1998a). Both plants and fungi produce ent-kaurene as a key precursor in

17 gibberellin biosynthesis (Hedden et al., 2001), and it has been shown that the putative prenyl cyclases from B. japonicum act sequentially to produce ent-kaurene via ent- copalyl diphosphate (CPP) from (E,E,E)-geranylgeranyl diphosphate (GGPP), and these are then an ent-CPP synthase (CPS) and ent-kaurene synthase (KS), respectively

(Morrone et al., 2009a). Moreover, homologs of this operon are found in a number of other rhizobia, several of which are known to produce gibberellins, and representative examples from the three other major genera of rhizobia (Mesorhizobium loti,

Sinorhizobium fredii, and Rhizobium etli) were shown to at least retain the capacity for production of ent-kaurene (Hershey et al., 2014). In particular, the CPS and KS homologs also produce ent-kaurene from GGPP, which was further shown to be the product of the associated prenyl transferase, which is then a GGPP synthase (GGPS). These results strengthen the association of this operon with gibberellin biosynthesis (Figure 1).

Intriguingly, there is a homolog of this putative gibberellin biosynthesis operon in the genome sequence of the rice bacterial leaf streak pathogen Xanthomonas oryzae pv. oryzicola (Xoc) strain BLS256 (Bogdanove et al., 2011). Here it is demonstrated that the putative CPS and KS from Xoc similarly produce ent-kaurene from GGPP. In addition, insertional disruption of the genes for not only the diterpene synthases, but also one of the

CYPs, led to Xoc strains with impaired pathogenicity, indicating that an ent-kaurene derived diterpenoid, potentially a gibberellin, acts as a virulence factor in the Xoc-rice interaction. Expression profiling of phytohormone marker genes indicates the ent-kaurene derived diterpenoid produced from Xoc down-regulates the rice plant defense response, most directly that mediated by JA.

Results

It has been previously noted that the KS from B. japonicum exhibited a distinct sequence relative to other characterized bacterial diterpene synthases (Morrone et al.,

2009a). This distinctive sequence has been used to identify other rhizobia that contain homologs to KS, and an associated diterpenoid biosynthetic operon, with functional conservation of at least the capacity for production of ent-kaurene (Hershey et al., 2014).

With the recent discovery of an ent-kaurene synthase from Streptomyces platensis, albeit for production of the antibiotic platensimycin rather than gibberellin phytohormones

(Smanski et al., 2011), we became interested in the wider distribution of such biosynthesis. While the ent-kaurene synthase from S. platensis is not closely related to the KSs found in rhizobia, sharing < 26% amino acid (aa) sequence identity with these

(which share > 67% aa sequence identity with each other), broader BLAST searches (i.e., outside of the previously investigated rhizobiales order), revealed the presence of homologs in the Xanthomonas genus. The initial example found was from the genome sequence of X. oryzae pv. oryzicola strain BLS256 (Bogdanove et al., 2011). Moreover, examination of the genomic context of this XocKS (Xoc_0077) revealed that it was situated in an operon exhibiting strong similarity to the putative gibberellin biosynthetic gene cluster found in the rhizobia. Specifically, it contains all eight genes found in B. japonicum that have been defined as the core operon in the rhizobia more generally (i.e.,

CYP112, CYP114, Fd, SDR, CYP117, GGPS, CPS and KS), with those from Xoc sharing

70-78% nucleotide (nt) sequence identity with their reported rhizobial homologs, close to the 80-92% nt sequence identity these share with each other. In addition, there are two other genes that appear to be part of the Xoc operon (Figure 1C). As also found in R. etli

19 and M. loti, on the 3′ end of the Xoc operon is an isopentenyl diphosphate isomerase

(IDI), which shares ~77% nt sequence identity with the IDIs from these two rhizobia

(again close to their 83% nt sequence identity to each other). There also is an additional

CYP, CYP115, on the 5′ end of the operon in Xoc. This shares some similarity to a partial

CYP pseudo-gene located in a similar 5′ position in the B. japonicum operon (Tully et al.,

1998a), but appears to be full-length.

Despite the strong similarity, the greater divergence of the Xoc operon from the previously characterized rhizobial homologs prompted us to investigate if the encoded enzymes were functionally conserved. Accordingly, we biochemically characterized the

CPS and KS from Xoc. These were cloned from Xoc genomic DNA and investigated using a previously described modular metabolic engineering system (Cyr et al., 2007), much as described for the rhizobial enzymes. Briefly, co-expression of the CPS and KS from Xoc, in recombinant E. coli also expressing a GGPS, led to the production of kaurene (Figure 2). Production of the ent-kaurene stereoisomer consistent with production of the gibberellin phytohormones was demonstrated by co-expressing the Xoc diterpene synthases with plant counterparts whose stereospecificity has already been established (again, along with GGPS co-expression in each case). In particular, the KS from Xoc was co-expressed with the ent-CPP producing CPS from Arabidopsis thaliana

(Sun and Kamiya, 1994), while the CPS from Xoc was co-expressed with the ent-CPP specific KS from Cucurbita maxima (Yamaguchi et al., 1996), leading to production of ent-kaurene in both cases. Moreover, analogous co-expression studies with plant diterpene synthases exhibiting alternative stereochemical specificity further demonstrated that XocCPS and XocKS were specific for the enantiomeric (9R,10R) isomer of CPP

(data not shown). Accordingly, the CPS and KS from Xoc cooperatively produce ent- kaurene, exhibiting functional conservation with their homologs from rhizobia, which is consistent with broader functional conservation of the entire operon.

Given that ent-kaurene is the olefin intermediate in gibberellin biosynthesis in both plants and fungi (Hedden et al., 2001), we hypothesized that Xoc might produce gibberellins. Unfortunately, we could not detect any of the common bioactive gibberellins (e.g., GA1, GA3, GA4 or GA7), or even ent-kaurene, from Xoc cultures. It has been similarly difficult to detect gibberellins (and also ent-kaurene) from rhizobial cultures (Hershey et al., 2014), although gibberellin metabolism can be observed with bacteroids isolated from root nodules (Mendez et al., 2014). Indeed, in the rhizobia it is clear that the operon is strongly induced upon interaction with the host plant during nodulation (Pessi et al., 2007). Accordingly, we hypothesized that the Xoc operon is similarly induced upon interaction with its host plant during infection. This was examined by comparing the expression level of several genes from the Xoc operon in infected leaves relative to that observed in bacterial culture over a 5-day time course.

Increases of up to 4-fold in transcript levels were observed, with the highest levels consistently observed after the first day (Figures 3 and S1). Thus, similar to the rhizobial expression pattern, the Xoc operon is induced during plant-microbe interactions.

To investigate the role of ent-kaurene production in Xoc, we generated knockout mutants of its CPS and KS. In particular, these were disrupted via single-crossover homologous recombination using pZeroBlunt based constructs (Figure S1). The mutant cps, ks and cyp112 could be amplified with the pZeroBlunt specific primer. Somewhat surprisingly, these were non-polar insertions, as expression of the downstream KS could

21 be detected in the resulting cps (CPS::pZeroBlunt) strain, and expression of the downstream IDI could be detected in the resulting ks (KS::pZeroBlunt) strain as well, although the targeted gene was clearly no longer expressed in both strains (Figure S2).

The effect of these mutations was investigated using a previously described paired infection assay that enables comparison of relative virulence (Wang et al., 2007). Briefly, the mutant and parental (wild-type) strains were inoculated, via infiltration with a needleless syringe, in a pairwise fashion on opposite sides of the mid-vein of the same leaf. Lesion length was then measured ten days after inoculation. With both cps and ks the resulting lesions were significantly shorter than those produced by the paired wild- type Xoc in paired student’s t-test (Figure 4A). To verify that the observed reduction in lesion length reflects decreased bacterial growth (i.e., virulence), the numbers of viable bacteria following inoculation with equivalent titers of each of the Xoc strains (i.e., wild- type, cps and ks) were compared. With this assay there is a clear decrease in virulence, with significantly reduced numbers of viable Xoc from inoculation with both cps and ks relative to wild-type that are observed within one day (Figure 4B).

To confirm that this reduction in virulence was caused by the loss of CPS and KS, we generated complementation strains. These strains (cps+CPS and ks+KS) were then tested in paired infection assays against either the parental mutant strains or the original wild-type strain. The results demonstrated that complementation effectively restored the virulence of both the cps and ks strains to levels equivalent to the wild-type in paired student’s t-test (Figure 5), confirming that these diterpene synthases contribute to the virulence of Xoc in rice, presumably via production of ent-kaurene as an intermediate in biosynthesis of a more elaborated diterpenoid virulence factor.

To investigate if it is the ent-kaurene product of CPS and KS or a further oxidized diterpenoid that acts as the virulence factor, the cytochrome P450 CYP112, presumably involved in such further elaboration, was targeted by insertional mutagenesis, creating the knock-out strain cyp112 (CYP112::pZeroBlunt), which also was found to be non-polar, with the continued expression of CPS and KS suggesting continued production of ent- kaurene (Figure S2). In paired infection tests, cyp112 also was found to exhibit significantly reduced lesion length relative to wild-type Xoc in paired student’s t-test

(Figure 6). Thus, it does appear to be an ent-kaurene derived diterpenoid, potentially a gibberellin, rather than ent-kaurene itself that acts as a virulence factor for Xoc.

Particularly given the reported antagonism between gibberellins and the JA mediated plant defense response (Bari and Jones, 2009), we hypothesized that the Xoc diterpenoid virulence factor might act through effects on phytohormone signaling. This was investigated by analysis of the expression profile of plant defense related marker genes in rice following infection with either cps or wild-type strains of Xoc. Transcript levels of lipoxygenase (OsLOX) and allene oxide synthase (OsAOS2) are increased by JA mediated signaling, while pathogenesis-related genes such as OsPR1a and OsPR1b are increased by salicylic acid (SA) mediated signaling (Navarro et al., 2008; Yoshii et al.,

2010; Qin et al., 2013). Hence, normalized transcript levels for these four rice genes were measured by qPCR over a 5-day time course following infection with either wild-type or cps Xoc, with significant differences evident for each. The most immediate effect was a greater increase in transcript levels of the JA signaling marker genes OsLOX and

OsAOS2, which is observed on the first day post-inoculation for leaves infected with cps relative to those infected with wild-type Xoc (Figure 7A&B). By contrast, no significant

23 difference over this time is observed between leaves infected with either wild-type or cps

Xoc for the SA signaling marker genes OsPR1a and OsPR1b (Figure 7C&D). Thus, the

Xoc ent-kaurene derived diterpenoid virulence factor does appear to act, at least in part, through suppression of the JA mediated plant defense response.

Notably, this diterpenoid biosynthetic operon does not appear to be present in any of the sequenced strains of the other pathovar of X. oryzae, pv. oryzae (Xoo), or other isolates that group separately from Xoc and Xoo (Lee et al., 2005; Ochiai et al., 2005;

Salzberg et al., 2008; Triplett et al., 2011). On the other hand, BLAST searches revealed the presence of a homologous operon in the recently reported genome sequence of

Xanthomonas translucens pv graminis ART-Xtg29 (Wichmann et al., 2013). Moreover, homologous sequence also can be found in the genome sequences for the two strains of

X. translucens pv. translucens, DSM 18974 and DAR61454, that are currently available at NCBI (Genome accession 14066). The operons from these two strains are essentially identical to each other, sharing ~99% nt sequence identity. In addition to the eight core operon genes found in the rhizobia (CYP112, CYP114, Fd, SDR, CYP117, GGPS, CPS and KS), these examples from X. translucens also all include the accessory IDI on the 3′ end and additional full-length CYP115 on the 5′ end. The shared presence of the unique full-length CYP115 suggests a homologous origin for the operons found in the X. translucens genomes and that from Xoc. However, between X. translucens and Xoc, these operons exhibit quite distinct sequences, sharing only ~70% identity at the nt level, indicating that any such common origin is quite distant and these may represent separate horizontal gene transfer events. By contrast, the operons from the two distinct pathovars of X. translucens are more closely related to each other (~89% nt sequence identity) and

24 presumably share a much more recent common origin, presumably via vertical descent rather than horizontal gene transfer (Figure S3).

Discussion

Horizontal gene transfer of this diterpenoid biosynthetic operon among the rhizobia has been previously demonstrated (Hershey et al., 2014). However, its presence in the Xanthomonas genus, which falls into the separate gamma-proteobacteria class, represents a surprisingly wide distribution. Nevertheless, these bacteria share an obvious common association with plants, albeit in strikingly different contexts, as symbionts and pathogens, respectively. Intriguingly, it appears that the operon may have been separately acquired by Xoc and X. translucens via horizontal gene transfer, presumably from the rhizobia.

The operon has been associated with gibberellin production in the rhizobia, and such production in the Xanthomonas phytopathogens would be consistent with the recent discovery that gibberellins are antagonistic to JA mediated plant defense (Robert-

Seilaniantz et al., 2007). However, it should be noted that the presence of the additional, unique cytochrome P450 CYP115 in the Xanthomonas associated operons indicates that these may produce a further elaborated diterpenoid than the rhizobia. For example, by analogy to the production of the JA-Ile mimic coronatine by the phytopathogen

Pseudomonas syringae (Fonseca et al., 2009), the Xanthomonas operon might lead to production of a gibberellin mimic resistant to degradation by plant catabolism.

Alternatively, the resulting diterpenoid might inhibit catabolism of endogenously produced gibberellins.

The observed rice host marker gene expression analysis is consistent with the potential production of gibberellin. First, the lack of such production by cps Xoc most immediately leads to a more robust JA mediated defense response, consistent with antagonism between the ent-kaurene derived diterpenoid and JA signaling, much as has been reported for gibberellin due to interactions between the DELLA and JAZ proteins involved in the gibberellin and JA response mechanisms, respectively (Hou et al., 2010;

Wild et al., 2012; Yang et al., 2012). Regardless of the exact identity of the ent-kaurene derived diterpenoid product of this operon, the results presented here demonstrate that this compound acts as a virulence factor promoting Xoc phytopathogen growth, ostensibly by suppressing the defense response of its host plant rice via effects on JA signaling. This mechanism also may underlie the use of gibberellins as virulence factors by the rice fungal pathogen G. fujikuroi (Wiemann et al., 2013), and is consistent with the emerging paradigm that a key pathogen virulence strategy is such modulation of plant hormone signaling (Robert-Seilaniantz et al., 2011).

Much as found in the rhizobia, this diterpenoid biosynthesis operon exhibits a scattered distribution within the Xanthomonas genus. The operon characterized here from the X. oryzae pathovar Xoc is not found in the Xoo pathovar. As Xoo and Xoc are otherwise very closely related (Nino-Liu et al., 2006; Lu et al., 2008; Bogdanove et al.,

2011), this operon may contribute to the differentiation between the rice leaf streak and blight diseases caused by Xoc and Xoo, respectively. Alternatively, production of the resulting ent-kaurene derived diterpenoid virulence factor may only provide a selective advantage under the conditions encountered by X. oryzae in the context of rice leaf streak rather than blight (e.g., infection of the mesophyll versus the xylem, respectively).

However, the presence of homologous sequence in various isolates of X. translucens suggests a broader role for the resulting ent-kaurene derived diterpenoid virulence factor.

In particular, together the sequenced X. translucens collectively cause leaf diseases on other small grain cereal crops such as wheat and barley, as well as forage grasses such as ryegrass, with some infecting the vasculature and others the mesophyll. Thus, this ent- kaurene derived diterpenoid may serve as a virulence factor for Xanthomonas phytopathogens of the Poaceae/grass plant family more generally. Building on the role in pathogenicity shown here for the Xoc-rice interaction, future investigation of the exact role of this ent-kaurene derived diterpenoid in the other Xanthomonas in which it is present will help elucidate its role in tissue-specificity versus Xanthomonas-Poaceae interactions more broadly.

Materials and Methods

Unless otherwise noted, all molecular biology reagents were purchased from

Invitrogen and all other chemicals from Fischer Scientific. Escherichia coli was grown at

37 °C or 16 °C using either NZY (for cloning) or TB medium (for expression). Xoc was cultured on GYE (2% glucose, 1% yeast extract) medium at 28 °C. When necessary,

1.8% (v/v) agar was added to the relevant medium to pour plates. Where applicable, antibiotics were used at the following concentrations: chloramphenicol – 30 µg/mL, carbenicillin – 50 µg/mL, spectinomycin – 50 µg/mL, and/or kanamycin – 50 µg/mL.

Liquid cultures were grown with vigorous shaking (200 rpm), generally in 250 mL

Erlenmeyer flasks containing 50 mL media.

The presence of the diterpenoid biosynthetic operon in Xoc was first detected by

BLAST searches of the non-redundant nucleotide database at GenBank in 2010 using the distinctive rhizobial KS gene as the query sequence (specifically, that from B. japonicum). Identification of the KS homolog was followed by analysis of the surrounding sequence, which proved to have the entire core operon defined from our studies in rhizobia (Hershey et al., 2014). The presence of homologous sequence in the various pathovars/strains of X. translucens was only more recently discovered by similar analysis carried out in 2013. Sequence analysis/alignments were carried out using the

CLC Main Workbench program version 6.8.4 (CLC bio).

Biochemical characterization of the CPS and KS from the Xoc operon was carried out as previously described (Morrone et al., 2009a). Briefly, the XocCPS and XocKS genes were amplified via PCR from Xoc BLS256 genomic DNA, prepared as previously described (Bogdanove et al., 2011), using gene specific primers (Table S1), cloned by topoisomerization into pENTR/SD/D-TOPO (Invitrogen), and verified by complete gene sequencing. For co-expression, XocCPS and XocKS were sub-cloned by directional recombination into pDest14 (Invitrogen) and a previously described GGPS containing pGG-DEST vector (Cyr et al., 2007), respectively. The resulting pDest14::XocCPS and pGG-DEST::XocKS constructs were co-transformed into E. coli strain C41 (Lucigen) along with pIRS, which overexpresses genes from the endogenous methylerythritol phosphate dependent isoprenoid precursor supply pathway and has been shown to increase flux towards terpenoid metabolism (Morrone et al., 2010). The resulting recombinant strains were cultured in 1 L of liquid media at 37 °C until they reached an optical density of ~ 0.6 at 600 nm, the pH adjusted to 7.1, the cultures shifted to 16 °C for

28 an hour prior to induction with IPTG added to a final concentration of 0.5 mM, followed by further fermentation at 16 °C for ~72 more hours. The cultures were then extracted with an equal volume of hexanes, which was separated out, dried in a rotary evaporator, and the residue resuspended in 100 µL hexanes. This concentrated extract was analyzed by gas chromatography (GC) carried out on a Varian (Palo Alto, CA) 3900 GC with

Saturn 2100 ion trap mass spectrometer (MS) in electron ionization (70 eV) mode.

Samples (1 µL) were injected in splitless mode at 50 °C and, after holding for 3 min. at

50 °C, the oven temperature was raised at a rate of 14 °C/min. to 300 °C, where it was held for an additional 3 min. MS data from 90 to 600 m/z were collected starting 12 min. after injection until the end of the run. The production of ent-kaurene was verified by comparison of mass spectra and retention time to an authentic standard (enzymatically produced by the characterized CPS and KS from A. thaliana).

For analysis of GA formation, Xoc was cultured in 500 mL flasks containing 300 mL liquid GYE media. The cultures were incubated for 3, 6, 9 and 12 days respectively on a rotary shaker (200 rpm) at 28 °C. Cells were separated from the spent media by centrifugation (15 min. at 4000 × g). For analysis of gibberellin content, the supernatant was acidified to pH 2.5 with acetic acid, and then extracted with an equal volume of ethyl acetate saturated with acetic acid (1% v/v). This organic extract was separated and passed over a 1 mL HP-20 resin column, which was then eluted with 3 mL each of 1% acetic acid in dH2O, then 40% and 80% methanol (v/v in dH2O with 1% acetic acid). Each of these fractions was dried in a rotary evaporator and the residue resuspended in 100 µL acetic acid-saturated ethyl acetate for analysis by GC-MS as described above. For analysis of the ent-kaurene intermediate, the total culture was directly extracted with an

29 equal volume of hexanes, which was separated out and passed over a 1 mL silica gel column to remove contaminating polar compounds, with the resulting organic extract dried under a gentle stream of N2, and the residue resuspended in 100 µL of hexanes, for analysis by GC-MS, again as described above.

To generate insertional mutant strains of Xoc, internal fragments from the CPS,

KS and CYP112 genes (700, 400 or 500 bp in length, respectively) were amplified from genomic DNA with gene specific primers. These PCR fragments were cloned into the pZeroblunt plasmid (carries kanamycin resistance) to create suicide vectors for insertional inactivation of CPS, KS and CYP112. The resulting plasmids were electroporated into Xoc for intragenic integration. Kanamycin-resistant colonies appearing at 28 °C were taken to be integrating mutants, in which a single crossover homologous recombination event had taken place. This was confirmed by PCR from genomic DNA [prepared using the Wizard Genomic DNA Purification Kit (Promega)] from each mutant, using the following primer pairs; for cps, M13-F and CPS200-R; for ks, KS50-F and M13-R; and for cyp112, CYP112.50-F and M13-R; generating PCR fragments of 1000, 700 or 900 bp length, respectively, if the relevant suicide vector has been inserted. In each case, the lack of such a PCR product with wild-type Xoc also was confirmed.

To investigate the polarity of the various mutant strains, RT-PCR analysis was carried out on these strains and compared to wild-type Xoc. Single colonies of Xoc strains were inoculated into 5 mL GYE and grown at 28°C to OD600 ~ 0.5. Bacterial cells were collected by centrifugation (15 min. at 4000 × g), and total RNA extracted using the

RNeasy Kit (Qiagen). The extracted RNA was treated with DNase I (Invitrogen) and

30 purified. cDNAs were obtained from RT reactions containing 500 ng total RNA using the iScript Reverse Transcription kit (BioRad), following the manufactures’ instructions.

This was used as the template for PCR with gene specific primers (for CPS, CPS430-R and CPS1103-R; for KS, KS249-F and KS649-R; for CYP112, CYP112.351-F and

CYP112.850-R; for CYP114, CYP114.106-F and CYP114.459-R; for IDI, IDI1-F and

IDI301-R), and 16S rRNA primers as a positive internal control.

To complement the cps and ks mutant strains, XocCPS and XocKS were transferred by directional recombination from the pENTR/SD/D-TOPO derived constructs described above to the Xanthomonas expression vector pKEB31 that contains the necessary DEST cassette (Cermak et al., 2011). The resulting pKEB31/DEST::XocCPS and pKEB31/DEST::XocKS vectors were transformed by electroporation into the Xoc cps and ks mutant strains, respectively. These two complementation strains (cps+CPS and ks+KS, respectively) were confirmed by sequencing PCR products of the introduced CPS and KS, respectively.

Quantitative real-time PCR (qPCR) was carried out with total RNA, extracted using the RNeasy Plant Mini Kit, from infected wild-type Nipponbare rice leaves collected either quickly (~30 min.) or 1, 2, 3, 4 or 5 days post-inoculation. Template cDNA was generated from 500ng DNAse treated RNA as described above. This was diluted 8-fold for use in ABsolute qPCR SYBR Green Mix (Thermo) reactions (25 µL), with the relevant gene specific pairs of primers (Table S1), carried out using a Stratagene

Mx4000 instrument located in the Iowa State University Genomics Technologies

Facility. Transcription FactorIIAγ5 (LOC_Os05g01710) was used as internal control for rice genes, and the Xoc reference gene was the ribosomal 16S subunit. Each data point

31 represents the average of two independent biological replicates, each measured with two technical replicates. For Xoc gene analysis, transcript levels in bacterial culture also were measured using DNAse treated total RNA isolated from wild-type Xoc as described above.

Relative virulence was assessed using paired-infection assays carried out with

Xoc cells suspended in 10 mM Magnesium chloride to an optical density at 600 nm of

0.5, which were used to spot infiltrate leaves of 8-week-old rice plants with a needleless syringe. Leaves of rice plants were inoculated with one strain on each side of the midrib

(e.g., mutant on one side and the wild type on the other). After 10 days, lesion lengths were measured for each paired inoculation, and a paired, two-tailed Student’s t-test was performed across all three replicates. Each experiment was carried out three separate times, each time with 10 leaves.

For bacterial colony counting, three leaves each infected with only one strain of

Xoc, were cut and ground in a sterile mortar and pestle with 2 ml sterile water. This suspension was then serially diluted (10-fold) and 5 µL drops from the 10-1 to 10-6 dilutions placed on peptone sucrose plates with cephalexin antibiotics, with each measurement (i.e., plate) carried out in triplicate. These plates were incubated at 28 °C until individual colonies were apparent for at least some of the spots, which were then counted, allowing calculation of the original titer of viable bacteria from each leaf.

Acknowledgements

The authors thank Shan Qi (Cornell University) for her help with the inoculation assays and helpful discussion. We also thank Aaron Hummel for productive discussions.

We are thankful to Brice Floyd for the technical assistance with the qRT-PCR analyses.

This work was supported by grants from the National Science Foundation

(MCB0919735) and U.S. Dept. of Agriculture (USDA-AFRI-NIFA 2014-67013-21720) to R.J.P.

References

Bari R, Jones JD. 2009. Role of plant hormones in plant defence responses. Plant Mol. Biol. 69(4): 473-488.

Bogdanove AJ, Koebnik R, Lu H, Furutani A, Angiuoli SV, Patil PB, Van Sluys MA, Ryan RP, Meyer DF, Han SW, et al. 2011. Two new complete genome sequences offer insight into host and tissue specificity of plant pathogenic Xanthomonas spp. J Bacteriol 193(19): 5450-5464.

Cermak T, Doyle EL, Christian M, Wang L, Zhang Y, Schmidt C, Baller JA, Somia NV, Bogdanove AJ, Voytas DF. 2011. Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic acids research 39(12): e82.

Cyr A, Wilderman PR, Determan M, Peters RJ. 2007. A Modular Approach for Facile Biosynthesis of Labdane-Related Diterpenes. J. Am. Chem. Soc. 129: 6684-6685.

Fonseca S, Chico JM, Solano R. 2009. The jasmonate pathway: the ligand, the receptor and the core signalling module. Curr Opin Plant Biol 12(5): 539-547.

Hedden P, Phillips AL, Rojas MC, Carrera E, Tudzynski B. 2001. Gibberellin Biosynthesis in Plants and Fungi: A Case of Convergent Evolution? J. Plant Growth Regul. 20: 319-331.

Hershey DM, Lu X, Zi J, Peters RJ. 2014. Functional conservation of the capacity for ent-kaurene biosynthesis and an associated operon in certain rhizobia. J. Bact. 196: 100-106.

Hou X, Lee LY, Xia K, Yan Y, Yu H. 2010. DELLAs modulate jasmonate signaling via competitive binding to JAZs. Dev Cell 19(6): 884-894.

Lee BM, Park YJ, Park DS, Kang HW, Kim JG, Song ES, Park IC, Yoon UH, Hahn JH, Koo BS, et al. 2005. The genome sequence of Xanthomonas oryzae pathovar oryzae KACC10331, the bacterial blight pathogen of rice. Nucleic acids research 33(2): 577-586.

Lu H, Patil P, Van Sluys MA, White FF, Ryan RP, Dow JM, Rabinowicz P, Salzberg SL, Leach JE, Sonti R, et al. 2008. Acquisition and evolution of plant pathogenesis-associated gene clusters and candidate determinants of tissue- specificity in xanthomonas. PloS one 3(11): e3828.

Mendez C, Baginsky C, Hedden P, Gong F, Caru M, Rojas MC. 2014. Gibberellin oxidase activities in Bradyrhizobium japonicum bacteroids. Phytochemistry 98: 101-109.

Morrone D, Chambers J, Lowry L, Kim G, Anterola A, Bender K, Peters RJ. 2009. Gibberellin biosynthesis in bacteria: Separate ent-copalyl diphosphate and ent- kaurene synthases in Bradyrhizobium japonicum. FEBS Letter 583(2): 475-480.

Morrone D, Lowry L, Determan MK, Hershey DM, Xu M, Peters RJ. 2010. Increasing diterpene yield with a modular metabolic engineering system in E. coli: comparison of MEV and MEP isoprenoid precursor pathway engineering. Applied Microbiology and Biotechnology 85: 1893-1906.

Navarro L, Bari R, Achard P, Lison P, Nemri A, Harberd NP, Jones JD. 2008. DELLAs control plant immune responses by modulating the balance of jasmonic acid and salicylic acid signaling. Current Biology 18(9): 650-655.

Nino-Liu DO, Ronald PC, Bogdanove AJ. 2006. Xanthomonas oryzae pathovars: model pathogens of a model crop. Molecular Plant Pathology 7(5): 303-324.

Ochiai H, Inoue V, Takeya M, Sasaki A, Kaku H. 2005. Genome sequence of Xanthomonas oryzae pv. oryzae suggests contribution of large numbers of effector genes and insertion sequences to its race diversity. Jarq-Japan Agricultural Research Quarterly 39(4): 275-287.

Pessi G, Ahrens CA, Rehrauer H, Lindemann A, Hauser F, Fischer H-M, Hennecke H. 2007. Genome-wide transcript analysis of Bradyrhizobium japonicum bacteroids in soybean root nodules. Molecular Plant-Microbe Interactions 20: 1353-1363.

Qin X, Liu JH, Zhao WS, Chen XJ, Guo ZJ, Peng YL. 2013. Gibberellin 20-oxidase gene OsGA20ox3 regulates plant stature and disease development in rice. Molecular Plant-Microbe Interactions 26(2): 227-239.

Robert-Seilaniantz A, Grant M, Jones JD. 2011. Hormone crosstalk in plant disease and defense: more than just jasmonate-salicylate antagonism. Annual Review of Phytopathology 49: 317-343.

Robert-Seilaniantz A, Navarro L, Bari R, Jones JD. 2007. Pathological hormone imbalances. Current Opinion in Plant Biology 10(4): 372-379.

Salzberg SL, Sommer DD, Schatz MC, Phillippy AM, Rabinowicz PD, Tsuge S, Furutani A, Ochiai H, Delcher AL, Kelley D, et al. 2008. Genome sequence and rapid evolution of the rice pathogen Xanthomonas oryzae pv. oryzae PXO99A. BMC Genomics epub.

Smanski MJ, Yu Z, Casper J, Lin S, Peterson RM, Chen Y, Wendt-Pienkowski E, Rajski SR, Shen B. 2011. Dedicated ent-kaurene and ent-atiserene synthases for platensimycin and platencin biosynthesis. Proceedings of the National Academy of Sciences 108(33): 13498-13503.

Sun T-P, Kamiya Y. 1994. The Arabidopsis GA1 locus encodes the cyclase ent-kaurene synthetase A of gibberellin biosynthesis. Plant Cell 6: 1509-1518.

Triplett LR, Hamilton JP, Buell CR, Tisserat NA, Verdier V, Zink F, Leach JE. 2011. Genomic analysis of Xanthomonas oryzae isolates from rice grown in the United States reveals substantial divergence from known X. oryzae pathovars. Applied and Environmental Microbiology 77(12): 3930-3937.

Tully RE, van Berkum P, Lovins KW, Keister DL. 1998. Identifcation and sequencing of a cytochrome P450 gene cluster from Bradyrhizobium japonicum. Biochimica et Biophysica Acta 1398: 243-255.

Wang L, Makino S, Subedee A, Bogdanove AJ. 2007. Novel candidate virulence factors in rice pathogen Xanthomonas oryzae pv. oryzicola as revealed by mutational analysis. Applied and Environmental Microbiology 73(24): 8023- 8027.

Wichmann F, Vorholter FJ, Hersemann L, Widmer F, Blom J, Niehaus K, Reinhard S, Conradin C, Kolliker R. 2013. The noncanonical type III secretion system of Xanthomonas translucens pv. graminis is essential for forage grass infection. Molecular Plant Pathology 14(6): 576-588.

Wiemann P, Sieber CM, von Bargen KW, Studt L, Niehaus EM, Espino JJ, Huss K, Michielse CB, Albermann S, Wagner D, et al. 2013. Deciphering the cryptic genome: genome-wide analyses of the rice pathogen Fusarium fujikuroi reveal complex regulation of secondary metabolism and novel metabolites. PLoS Pathogens 9(6): e1003475.

Wild M, Daviere JM, Cheminant S, Regnault T, Baumberger N, Heintz D, Baltz R, Genschik P, Achard P. 2012. The Arabidopsis DELLA RGA-LIKE3 is a direct target of MYC2 and modulates jasmonate signaling responses. Plant Cell 24(8): 3307-3319.

Yabuta T, Sumiki Y. 1938. On the crystal of gibberellin, a substance to promote plant growth. Journal of Agricultural and Food Chemistry Soc. Japan 14: 1526.

Yamaguchi S, Saito T, Abe H, Yamane H, Murofushi N, Kamiya Y. 1996. Molecular cloning and characterization of a cDNA encoding the gibberellin biosynthetic enzyme ent-kaurene synthase B from pumpkin (Cucurbita maxima L.). Plant Journal 10: 101-111.

Yang DL, Li Q, Deng YW, Lou YG, Wang MY, Zhou GX, Zhang YY, He ZH. 2008. Altered disease development in the eui mutants and Eui overexpressors indicates that gibberellins negatively regulate rice basal disease resistance. Molecular Plant 1(3): 528-537.

Yang DL, Yao J, Mei CS, Tong XH, Zeng LJ, Li Q, Xiao LT, Sun TP, Li J, Deng XW, et al. 2012. Plant hormone jasmonate prioritizes defense over growth by interfering with gibberellin signaling cascade. Proceedings of the National Academy of Sciences 109(19): E1192-1200.

Yoshii M, Yamazaki M, Rakwal R, Kishi-Kaboshi M, Miyao A, Hirochika H. 2010. The NAC transcription factor RIM1 of rice is a new regulator of jasmonate signaling. Plant Journal 61(5): 804-815.

Figures

A) rhizobia CYP112 CYP114 Fd SDR CYP117 GGPS CPS KS ()IDI B) OPP OPP OPP IDI GGPS CPS KS

OPP

Xoc CYP115 CYP112 CYP114 Fd SDR CYP117 GGPS CPS KS IDI C)

Figure 1. Putative gibberellin biosynthesis operon from the rhizobia. A) Schematic representation of operon showing core genes defined in the text, plus accessory isopentenyl diphosphate isomerase (IDI) found in some cases. B) Scheme illustrating enzymatic activities demonstrated for certain examples from the rhizobia. C) Schematic representation of the putative gibberellin operon from Xoc.

Figure 2. Xoc exhibits the ability to produce ent-kaurene. GC-MS selected ion (m/z = 272) chromatograms demonstrating production of kaurene from GGPP by XocCPS and XocKS, along with an authentic standard.

Figure 3. Quantitative expression analysis of A) XocCPS; B) XocCYP112; or C) XocKS transcript levels in culture or rice leaves 1, 2, 3, 4 or 5 days after inoculation (DAI).

A) i) ii) B) i) ii) C)

WT cps WT ks

Figure 4. Knocking out CPS or KS reduces virulence of Xoc. A) Box plots of lesion lengths from paired inoculations matching wild-type (WT) with either cps (i) or ks (ii) knock-out strains, as measured 10 days after infiltration. These are representative data from one of three independent experiments, each of which was carried out with three replicates, each comprised of 10 leaves, with significant differences observed in each case in paired student’s t-test (p-values indicated by * < 0.05 or ** < 0.01). B) Representative image from paired inoculation of WT with either cps (i) or ks (ii) strains. C) Reduced number of viable bacteria from infected plants. Time course illustrating results from bacterial colony numbers from infected leaves. Infiltrated region was calculated from three independent experiments with three infiltrations in five leaves for each strain of Xoc (p-values indicated by * < 0.05 or ** < 0.01).

A) B)

Figure 5. Complementation restores virulence of cps and ks strains of Xoc. Box plots of lesion lengths from paired inoculation of wild-type (WT) with either cps+CPS or ks+KS strains in paired student’s t-test, as indicated (no significant difference was observed).

Figure 6. Knocking out CYP112 also reduces Xoc virulence. Box plot of lesion lengths from paired inoculation of WT and cyp112 in paired student’s t-test (** indicates p-value < 0.01).

A) C)

B) D)

Figure 7. Quantitative comparison of A) OsLOX; B) OsAOS2; C) OsPR1a; or D) OsPR1b gene expression in rice following inoculation with either wild-type or cps Xoc either shortly (~30 min.) or 1, 2, 3, 4 or 5 days after inoculation.

Supplementary Results

Figure S1. PCR analysis of the disruption of Xoc CPS, KS or CYP112. Lane 1: 1 and 2 as the primers, and the genomic DNA of knockout cps strain as the template; Lane 2: 1 and 2 as the primers, and the genomic DNA of WT Xoc BLS256 as the template; Lane 3: 3 and 4 as the primers, and the genomic DNA of knockout ks strain as the template; Lane 4: 3 and 4 as the primers, and the genomic DNA of WT Xoc BLS256 as the template. Primer M13R is specific to pZeroBlunt.

Figure S2. RNA expression of CPS, KS, CYP112, CYP114, IDI, and 16S genes in WT, cps, ks and cyp112 strains of Xoc (as indicated).

Figure S3. Phylogenetic tree from alignment of the core diterpenoid biosynthesis operon from the various bacterial species discussed here. In particular, those from the Xanthomonas genus, abbreviated here as Xoc (X. oryzae pv. oryzicola), Xtg (Xanthomonas translucens pv. graminis) and Xtt (X. translucens pv. translucens), along with those from the rhizobia, which are abbreviated as Ml (Mesorhizobium loti), Re (Rhizobium etli), Bj (Bradyrhizobium japonicum) and Sf (Sinorhizobium fredii).

Table S1. Primers used in this study

Primer Name Sequence XocCPS F 5'-CACCATGAACGCGCTGTGCGAACAGATTTTC-3' XocCPS R 5'-TCATGGCGCAGCTCCCGTATGCGTGAC-3' XocKS F 5'-CACCATGATCAAGGACGCGCTGGAGCAGG-3' XocKS R 5'-TCAGACGGGCGCATGCTGGCTCCC-3' XocCPS 403F 5'-GCGTTGTCGAACGGCGCGGCCTTTCCGCGC-3' XocCPS 1103R 5'-TGCGGTTGGTTTCGACGTAGGTCCTGGTGG-3' XocKS 249F 5'-GGCCCAGCTTGACGCGGTTGAGCGAGCGCTG-3' XocKS 649R 5'-CAGCGACCTGGTCCTTGCCGCGCCCGTGC-3' XocCYP112 351F 5'-GATCGCGGCGCAGTTGTTCGAGACGCTGGC-3'

XocCYP112 850R 5'-CGCCCGGCGGATACATGCGGAGGATTTCCT-3' XocIDI F 5'-ATGCGCCCGTCTGAACCCGCGCCC-3' XocIDI 301R 5'-TGGCATCGGCGCGCGGCATACCGC-3' Xoc16SrRNA F 5'-ATGAAGCAGACTGAGCGCCGCATC-3' Xoc16SrRNA R 5'-TCAGAAATCCGGATCCCAGTCGAC-3'

M13 Foward-R 5'-GTAAAACGACGGCCAG-3' XocCPS 200F 5'-TTCCGCTATTCCGGCATGCGCCCA-3' XocKS 50F 5'-AGGGCTTCTCCTACCAGCATGCCAC-3' XocCYP112 50F 5'-TGAACCCTCGCCTGCCATGCTGGCA-3'

CHAPTER 3: FUNCTIONAL CONSERVATION OF THE CAPACITY FOR ENT-

KAURENE BIOSYNTHESIS AND AN ASSOCIATED OPERON IN CERTAIN

RHIZOBIA

A paper published in Journal of Bacteriology 2014, 196 (1), 100-106

David M. Hershey†,‡, Xuan Lu†, Jiachen Zi, and Reuben J. Peters*

Department of Biochemistry, Biophysics & Molecular Biology, Iowa State University,

Ames, IA, USA

†These authors contributed equally

Abstract

Bacterial interactions with plants are accompanied by complex signal exchange processes. Previously, the nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum was found to carry adjacent genes encoding two sequentially acting diterpene cyclases that together transform geranylgeranyl diphosphate to ent-kaurene, the olefin precursor to the gibberellin plant hormones. In order to understand the ecological and genomic context of such diterpene production, bioinformatics searches were carried out to identify other rhizobia that might have the capacity to produce ent-kaurene. Species from the three other major genera of rhizobia contained homologous terpene synthase genes. Cloning and functional characterization of a representative set of these enzymes confirmed the capacity of each genus to produce ent-kaurene. Moreover, comparison of their genomic context revealed that these diterpene synthases are found in a conserved

43 operon, which includes an adjacent isoprenyl diphosphate synthase, shown here to produce the geranylgeranyl diphosphate precursor. In addition, the rest of the operon consists enzymatic genes that presumably lead to more elaborated diterpenoid natural products, although the production of gibberellins was not observed. Nevertheless, it has previously been shown that the operon is selectively expressed during nodulation, and further bioinformatics analysis suggests that the operon has been distributed in a scattered pattern throughout the rhizobia via independent horizontal gene transfer within the symbiotic plasmid or genomic island, indicating that such diterpenoid production likely modulates the interaction of these particular symbionts with their host plants.

Introduction

Bacteria play critical roles in biogeochemical cycles such as the fixation of nitrogen. Although nitrogen makes up approximately 80% of the earth’s atmosphere, its bioavailability remains a major limitation – e.g., to plant growth (Vance, 2001). This is due to the inability of plants to assimilate the diatomic nitrogen that occurs naturally in the atmosphere (Den Herder and Parniske, 2009). Among plants, legumes uniquely host bacteria from the rhizobiales order of α-proteobacteria in nodules formed following invasion of their root cortical cells. Inside these nodules the rhizobia develop into endosymbiont bacteroids, fixing nitrogen in exchange for carbon from their plant hosts

(Jones et al., 2007). This agriculturally important collaboration is thought to be the main biological route for nitrogen fixation. Of particular relevance here, a number of the rhizobia have been shown to produce plant growth hormones, such as the gibberellins, which are thought to further promote growth of the host plant (Bottini et al., 2004).

Both legume and rhizobial species exhibit a surprising amount of specificity with respect to symbiotic partners (Fauvart and Michiels, 2008). Only rarely can a given rhizobia species nodulate more than a few closely related plants. This specificity is due to bacterial and plant factors (Den Herder and Parniske, 2009). The host plant secretes flavonoid inducers that elicit rhizobial production of lipochitooligosaccharide Nod factors, which is recognized by the host plant, with subsequent steps in the nodulation process dependent on recognition of bacterial cell surface chemistry and effector proteins as well. However, the host plant also applies the usual defense mechanisms – e.g., microbe associated molecular pattern triggered immunity and R-gene recognition of bacterial effectors – to restrict nodulation by unwanted strains (Jones et al., 2007; Den

Herder and Parniske, 2009; Yang et al., 2010). This complex signal exchange process exerts extreme evolutionary pressure on the rhizobia (Johnston et al., 1978; Flores et al.,

2005; Gonzalez et al., 2010), which can be inferred, in part, from the presence of large plasmids or genomic “islands” with distinct G+C content relative to the rest of the genome, which contain the large set of genes required for nodulation, including nitrogen fixation (Kaneko et al., 2000; Kaneko et al., 2002; Mavingui et al., 2002; Gonzalez et al.,

2006; Fauvart and Michiels, 2008). This presumably reflects the ability of horizontal transfer to spread these distinct genetic elements, enabling nodulation by the recipient rhizobia (Johnston et al., 1978; Sullivan and Ronson, 1998). Intriguingly, the prevalence of insertion sequences and phage integration is thought to promote rearrangement within these symbiotic modules (Viprey et al., 2000).

Previously we characterized two terpene synthases from B. japonicum USDA110

(Morrone et al., 2009). These proved to be diterpene cyclases capable of successively

45 converting the general diterpenoid precursor geranylgeranyl diphosphate (GGPP) into ent-copalyl diphosphate (ent-CPP) and, hence, to ent-kaurene, a precursor to the gibberellin phytohormones (Figure 1). The relevant genes, blr2149 and blr2150, then encode an ent-CPP synthase (CPS) and ent-kaurene synthase (KS) respectively. Notably, these two genes fall into a more extensive operon that was originally defined by Tully et al. (Tully et al., 1998a), and suggested to be present in all rhizobia (Keister et al., 1999), although it did not appear to affect the ability of B. japonicum to nodulate soybean

(Glycine max) or fix nitrogen in the resulting nodules (Tully and Keister, 1993). Here we have investigated the functional conservation of the CPS and KS from this operon and demonstrated production of the upstream GGPP by the isoprenyl diphosphate synthase encoded by the adjacent gene in the operon. In addition, our results more precisely define the overall gene content of the operon and its scattered distribution in the rhizobia. While examples are found in all four major genera, with conservation of the ability to produce ent-kaurene, the uneven distribution of the operon suggests that such diterpenoid production provides a selective advantage only under certain conditions.

Results

Identification of ks and cps homologs in rhizobia

As previously noted, the BjKS that directly produces ent-kaurene exhibits distinct sequence homology relative to other characterized bacterial diterpene synthases.

Accordingly, the BjKS sequence was used in initial BLAST searches of the NCBI database to identify bacteria from the rhizobiales order that contain homologous diterpene synthases. Notably, homologs were found in species from all four major genera

46 of rhizobia – i.e., Rhizobium, Sinorhizobium, and Mesorhizobium, in addition to

Bradyrhizobium. In each case, immediately upstream of BjKS homolog was a homolog to

BjCPS (Table 1), with overlapping open-reading frame just as found in B. japonicum.

Interestingly, these were not conserved by bacterial phylogeny – e.g., the KS from the various species of Bradyrhizobium share less sequence identity than that between BjKS and the KS from Mesorhizobium loti, which is not only in a distinct genus but also falls into the separate Phyllobacteriaceae family. Accordingly, the KS and CPS appear to have been distributed via horizontal gene transfer. Consistent with such an inheritance mechanism, the KS gene is not found in all rhizobia [e.g., no homolog is present in

Rhizobium leguminosarum, whose genome has been fully sequenced (Young et al.,

2006), and nor, despite the presence of a homologous protein sequence annotated as being encoded by Sinorhizobium meliloti, are homologs present in any of the genome sequences reported for various strains of S. meliloti (Galibert et al., 2001; Schneiker-

Bekel et al., 2011; Weidner et al., 2013)].

Functional characterization of representative CPS and KS

To investigate the ability of the CPS and KS homologs found in our bioinformatics search to cooperatively produce ent-kaurene, we analyzed these from one species from each genus, specifically examples with complete genome sequences reported, Mesorhizobium loti MAFF303099 (Kaneko et al., 2000), Sinorhizobium fredii

NGR234 (Schmeisser et al., 2009), and Rhizobium etli CFN42 (Gonzalez et al., 2006).

We cloned and characterized the CPS and KS from each of these species much as previously described for those from B. japonicum (Morrone et al., 2009). Briefly, each

47 pair of CPS and KS homologs were co-expressed in recombinant E. coli also expressing a

GGPP synthase, which led to the production of kaurene (Figure 2). To demonstrate stereospecificity, each CPS was expressed in recombinant E. coli with GGPS from Abies grandis (AgGGPS) and the KS from Arabidopsis thaliana (AtKS), which is specific for ent-CPP. In addition, each KS was expressed in recombinant E. coli with AgGGPS and the ent-CPP producing CPS from Arabidopsis thaliana (AtCPS). In each case, this led to the production of ent-kaurene (data not shown), demonstrating stereochemistry consistent with that of the gibberellin plant hormones. These results confirmed a common catalytic activity for these distributed enzymatic genes and, importantly, each genus identified here then contains the capacity to produce ent-kaurene from GGPP.

Defining a rhizobial diterpenoid biosynthetic operon

Given that BjCPS and BjKS are neighboring genes in what has been proposed to be a more extensive operon (Tully and Keister, 1993), we examined the genomic context for each of the characterized CPS and KS to determine if these were similarly set in a more broadly conserved operon. Indeed, the CPS is invariably found immediately upstream of the KS, and homologs to all of the other genes also were present in the same order as found in the B. japonicum operon. In particular, homologs to the adjacent cytochromes P450 CYP112 and CYP114, a ferredoxin (Fd), a short-chain alcohol dehydrogenase/reductase (SDR), another cytochrome P450 (CYP117), an isoprenyl diphosphate synthase that presumably makes GGPP (GGPS), as well as the orthologous

CPS and KS. Although it should noted that some of these genes are fused together in certain cases – i.e., the CYP114 and Fdx in R. etli, and the Fdx and SDR in M. loti, these

48 still exhibit clear homology to the separate genes found elsewhere. Thus, these genes then define a core diterpenoid biosynthetic gene cluster/operon that is conserved across all four of the major rhizobial genera, sharing 80-92% nucleotide sequence identity.

Notably, the GGPS gene in R. etli appears to be disrupted. While some homologous sequence is present, large deletions and various frame-shifts result in a clearly compromised open reading frame (accordingly, we suggest that this is a pseudo- gene and refer to it as GGPSp). However, R. etli contains another isoprenyl diphosphate synthase in close proximity to its core operon. Although this is not closely related to the

GGPS found within the operon, and is in the opposite orientation, we hypothesized that this might serve the same function (and refer to it here as GGPS2). There is an intervening gene. However, this gene appears to encode an isopentenyl diphosphate isomerase (IDI), which balances isoprenoid precursor supply and, thus, similarly has a plausible role in (di)terpenoid biosynthesis as well. Further analysis demonstrated that a homologous IDI gene also occurs at the same position (3ʹ to the KS) in M. loti.

Intriguingly, M. loti further has a gene encoding a major facilitator superfamily (MFS) member immediately downstream of its IDI, and we speculate that this might be involved in secretion of the diterpenoid natural product. Accordingly, in R. etli and M. loti accessory genes appear to have been appended to the core diterpenoid biosynthesis operon (Figure 3).

Upon sequence comparisons of the core operon, that from R. etli appears to be the most divergent, sharing < 82% identity, while the others are ≥ 90% identical with each other. Even when excluding the compromised GGPSp, comparison of the other genes from the R. etli operon reveals that these are only 86-89% identical with those from the

49 other rhizobia, less than the identity shared by the other rhizobia. Accordingly, the R. etli operon is clearly the most divergent, consistent with distribution of the entire operon by horizontal gene transfer – e.g., despite their common phylogenetic origin in the

Rhizobiaceae family, R. etli and S. fredii contain the most disparate rhizobial diterpenoid biosynthesis operons.

Demonstrating production of GGPP and capacity for ent-kaurene production

Diterpenoid biosynthesis generally proceeds via initial formation of a hydrocarbon skeletal structure, followed by oxidative elaboration (Peters, 2010). In looking at the organization of the rhizobial diterpenoid biosynthesis operons, it is notable that the genes predicted to be involved in oxidation are in the 5ʹ region, with all those predicted to be involved in the formation of the cyclized olefin ent-kaurene falling in the

3ʹ region. This might suggest that these form nominally independent sub-clusters, although no such sub-clusters appear in the currently available sequence information.

Nevertheless, we took advantage of this gene arrangement to demonstrate both the production of GGPP by the isoprenyl diphosphate synthase and, hence, ability of the operon to lead to production of ent-kaurene. While initial efforts were directed at heterologous expression of the putative GGPP synthase in E. coli for use in co-expression studies such as those described above, that ultimately proved unsuccessful. We then turned to recombinant expression in a more closely related bacterium, specifically the

1021 strain of Sinorhizobium meliloti, whose reported genome sequence does not contain the rhizobial diterpenoid biosynthesis operon (Galibert et al., 2001). Accordingly, we overexpressed the 3ʹ region of the operon from the closely related S. fredii, either a

50 fragment containing GGPS-CPS-KS or CPS-KS only. Consistent with the usual lack of

GGPP production in bacteria, expression of the CPS-KS genes alone in S. meliloti 1021 did not result in the production of ent-kaurene, while expression of the GGPS-CPS-KS genes did (Figure 4). These results then not only confirm that the associated isoprenyl diphosphate synthase produces GGPP, but also the capacity of the operon to confer the ability to produce at least ent-kaurene.

Evidence for a role in symbiosis

In each of the rhizobia examined here, the diterpenoid biosynthesis operon is located in the symbiotic module (plasmid or genomic island), suggesting that the resulting natural product may play a role in the symbiotic relationship of these rhizobia with their host plants. Consistent with such a role, in B. japonicum the genes in the diterpenoid biosynthesis operon are expressed upon bacterial differentiation in the nodules to the nitrogen-fixing bacteroid form, and to a much lesser extent under microanaerobic culture conditions, which mimic those found in the nodules (Pessi et al.,

2007). This expression is dependent on the symbiotic specific sigma factor RpoN and associated transcription factor NifA (Hauser et al., 2007). Similarly, expression of the genes from the diterpenoid biosynthesis operon in R. etli also depends on RpoN and

NifA, and occurs under similar conditions – i.e., in nodules and under microanaerobic conditions (Salazar et al., 2010). In addition, upstream of the diterpenoid biosynthesis operon in the other two genera, there are sequences suggestive of regulation by the

RpoN-NifA regulon as well. In particular, putative binding sites for both NifA and RpoN,

51 specifically the same sites identified upstream of the diterpenoid biosynthesis operon in

R. etli (Salazar et al., 2010).

Given that ent-kaurene is the olefin intermediate in biosynthesis of the gibberellin phytohormones in both plants and fungi (Hedden et al., 2001), coupled to previous reports of bacterial production of gibberellins (Bottini et al., 2004), we have hypothesized that the capacity to produce this diterpene is indicative of the ability to produce gibberellins (Morrone et al., 2009a). However, although it has been reported that B. japonicum produces gibberellin A3 (GA3) under standard liquid culture conditions

(Boiero et al., 2007), we have been unable to detect this, or ent-kaurene, from B. japonicum grown under the previously described conditions, or even under microanaerobic conditions. In addition, we were unable to detect production of ent- kaurene from any of the other rhizobia examined here under aerobic or microanaerobic culture conditions.

While the rhizobial symbiotic modules (plasmid or genomic island) are known to be distributed via horizontal gene transfer, the diterpenoid biosynthetic operon seems to be separately distributed, albeit via selective integration into the symbiotic module. For example, it has been noted that this cytochrome P450 rich gene cluster exhibits a different G+C content relative to the rest of the symbiotic plasmid in S. fredii NGR234

(Freiberg et al., 1997). The phylogenetic relationship of the CPS genes from the analyzed rhizobial diterpenoid biosynthesis operons, for which homologs are evident in other bacteria, is distinct from that for the gene for the nitrogenase subunit NifK (Figure 5).

This suggests that the diterpenoid biosynthetic operon, which is not found in all such rhizobial symbiotic modules in any case, has been independently incorporated into these

52 symbiotic modules. Although it has already been shown that polar disruption of the initial

CYP112 gene in B. japonicum has no discernable effect on its ability to nodulate soybean or fix nitrogen (Tully and Keister, 1993), the resulting diterpenoid natural product may provide some selective advantage in the symbiotic growth phase of the rhizobia where it has been incorporated.

Discussion

The results presented here demonstrate scattered distribution within the rhizobia of a diterpenoid biosynthetic operon, with functional conservation of at least the capacity for production of ent-kaurene (Figures 2 and 4), although the final diterpenoid end product(s) remains unclear at this time. The location of this operon in the symbiotic module of the relevant rhizobia, along with its previously demonstrated transcription in response to bacteroid differentiation in nodules, indicates a role for the resulting diterpenoid natural product in the symbiotic relationship of these rhizobia with their host plants. While the exact physiological function of this diterpenoid remains unclear at this time as well, the scattered distribution of the operon, which appears to be a result of its apparently independent horizontal gene transfer between symbiotic modules, suggests that it provides a selective advantage only under certain conditions. Nevertheless, the striking conservation of this diterpenoid biosynthetic operon hints at its importance.

Intriguingly, the characterized operons are all from rhizobia associated with determinate, rather than indeterminate, nodules. While nodule type is specified by the host plant species (Oke and Long, 1999; Oono et al., 2010), rhizobial specificity for plant host species indirectly controls bacterial nodulation phenotypes. Accordingly, it is tempting to

53 speculate that the diterpenoid product of this operon specifically plays a role in rhizobial interactions within determinate nodules, providing a direction for future investigations.

Materials And Methods

General

Unless otherwise noted, all molecular biology reagents were purchased from

Invitrogen and all other chemicals from Fischer Scientific. B. japonicum USDA110 was obtained from Dr. Michael Sandowsky (Univ. Minnesota), Mesorhizobium loti

MAFF303099, Sinorhizobium fredii NGR234 and Rhizobium etli CFN42 were all obtained from Dr. Philip Poole (John Innes Centre), while Sinorhizobium meliloti 1021 was obtained from Dr. Kathryn Jones (Florida State University). Escherichia coli was grown at 37 or 16 °C on either NZY (for cloning) or TB medium (for expression).

Rhizobia were cultured with YEM medium at 28 °C. When necessary, 1.8% agar was added to the relevant medium to pour plates. Where applicable, antibiotics were used at the following concentrations: chloramphenicol – 30 µg/mL, carbenicillin – 50 µg/mL, spectinomycin – 50 µg/mL, kanamycin – 50 µg/mL. Liquid cultures were grown with vigorous shaking (200 rpm), generally in 250 mL Erlenmeyer flasks with 50 mL media.

Microanaerobic cultures were grown in YEM media with 10 mM KNO3 under an atmosphere of nitrogen gas (N2) and ~0.5% oxygen with moderate shaking (80 rpm) in rubber-stoppered flasks, with the atmosphere exchanged every 12 hours (N2 blown into the flasks for 15 min).

Sequence retrieval and analysis

All sequences were retrieved from the National Center for Biotechnology

Information (NCBI) website. The amino acid sequence of the previously characterized

KS from B. japonicum (BjKS) was used as a query for BLAST searches against the rhizobiales order (taxid: 356) on the NCBI website (Altschul et al., 1990). This also was done using the CPS from B. japonicum (BjCPS) amino acid sequence as the query sequence. Sequence analyses were carried out with the CLC Main Workbench program

(version 6.8.4). Alignments used the following parameters: Gap Open Cost – 10, Gap

Extension Cost – 10, End Gap Cost – As any other. Trees were prepared using the

Neighbor Joining algorithm with a bootstrap analysis of 1000 replicates. PAUP analysis was used to confirm the topology of the resulting trees. The presented phylogenetic analyses were carried out using genes encoding biochemically analogous proteins from as distantly related species as possible as the out-group sequences. For CPS the out-group sequence was that for the ent-CPP producing CPS from Streptomyces sp. KO-3988

(GenBank AB183750)(Kawasaki et al., 2004), which falls within the Actinobacteria phylum. For NifK the out-group sequence was from Azotobacter vinelandii (GenBank gene ID Avin_01400), which falls within the proteobacteria phylum, but is in the distinct gammaproteobacteria class.

Cloning and characterization of CPS and KS

Genomic DNA was isolated from rhizobia using the Generation Capture Kit

(Qiagen) following the manufacturers protocol. Each CPS and KS gene was amplified via

PCR from genomic DNA and cloned into pENTR-SD-dTOPO (Invitrogen). Biochemical

55 characterization of the CPS-KS pair from each species of rhizobia was carried out as described previously for B. japonicum (Morrone et al., 2009a). Briefly, the KS genes were sub-cloned into the plasmid pGG-DEST, which carries a GGPP synthase gene, and the CPS genes sub-cloned into pDEST14. This enabled use of a previously described metabolic engineering system, which included analogous constructs with the CPS and KS from Arabidopsis thaliana (Cyr et al., 2007). Accordingly, the E. coli strain OverExpress

C41 (Lucigen) was transformed with the various combinations of pGG-DEST::CPS and pDEST14::KS plasmids described below, along with pIRS [i.e., to increase the isoprenoid precursor pool, as described (Morrone et al., 2010)]. Liquid cultures (50 mL) of the resulting recombinant E. coli were induced at an optical density of 0.6, the pH adjusted to 7.1 and grown at 16 °C for 72 hours. The cultures were then extracted with an equal volume of hexanes, which was separated out, dried in a rotary evaporator, and the residue resuspended in 100 µL hexanes. This concentrated extract was analyzed by gas chromatography (GC) carried out on a Varian (Palo Alto, CA) 3900 GC with Saturn 2100 ion trap mass spectrometer (MS) in electron ionization (70 eV) mode. Samples (1 µL) were injected in splitless mode at 50 °C and, after holding for 3 min. at 50 °C, the oven temperature was raised at a rate of 14 °C/min. to 300 °C, where it was held for an additional 3 min. MS data from 90 to 600 m/z were collected starting 12 min. after injection until the end of the run. The production of ent-kaurene was verified by comparison of mass spectra and retention time to an authentic standard (enzymatically produced by the characterized CPS and KS from A. thaliana).

Mapping the diterpenoid biosynthesis operon

A 20 kB region surrounding each biochemically characterized KS was downloaded from the NCBI website and further analyzed. The predicted genes that were either homologous to those in the B. japonicum operon or had plausible predicted functions in (di)terpenoid biosynthesis were identified by alignment and open reading frame prediction. In each case, the boundaries of each operon were clear from the predicted functions of the adjacent genes and their position in the genome. Putative RpoN and NifA binding sites were identified by searching for identical matches in the upstream region of each operon to previously defined 16-nucleotide motifs (Salazar et al., 2010).

Characterization of GGPS

Fragments from the 3ʹ region of the operon from S. fredii including genes for

GGPS-CPS-KS or CPS-KS only were amplified via PCR from genomic DNA. These were cloned into pZeroBlunt, and then sub-cloned into a previously described S. meliloti overexpression vector (Jones, 2012), pstb-LAFR5 (obtained from Dr. Kathryn Jones,

Florida State University), using BamHI and EcoRI restriction sites introduced by PCR.

The resulting constructs were transformed in to S. meliloti 1021 by conjugation from E. coli. These recombinant strains were grown for three days, then the total culture extracted with an equal volume of hexanes. This organic extract was separated out and dried under a gentle stream of N2, with the residue resuspended in 100 µL of hexanes for analysis by

GC-MS, as described above.

Analysis of rhizobial diterpenoid production

Liquid cultures grown under aerobic or microanaerobic conditions were harvested

3, 6 or 9 days after inoculation, and the cells separated from the spent media by centrifugation (15 min. at 10,000 × g). For analysis of gibberellin content, the supernatant was acidified to pH 2.5 with acetic acid, and then extracted with an equal volume of ethyl acetate saturated with acetic acid (1% v/v). This organic extract was separated and passed over a 1 mL HP-20 resin column, which was then eluted with 3 mL each of 1% acetic acid in dH2O, 40% and 80% methanol (v/v in dH2O with 1% acetic acid). Each of these fractions was dried in a rotary evaporator and the residue resuspended in 100 µL acetic acid-saturated ethyl acetate for analysis by GC-MS as described above. For analysis of the ent-kaurene intermediate, the total culture was directly extracted with an equal volume of hexanes, which was separated out and passed over a 1 mL silica gel column to remove contaminating polar compounds, with the resulting organic extract dried under a gentle stream of N2, and the residue resuspended in 100 µL of hexanes, for analysis by

GC-MS, again as described above.

Acknowledgements

The authors thank Dr. Philip Poole (John Innes Centre) for providing rhizobia and helpful discussion, and Dr. Kathryn Jones (Florida State University) for providing the pstb-LAFR5 expression vector. We also thank Dr. Matt Hillwig for productive discussions. This work was supported by a grant from the National Science Foundation

(MCB0919735) to R.J.P.

References

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990). Basic local alignment search tool. Journal of Molecular Biology 215, 403-410.

Boiero, L., Perrig, D., Masciarelli, O., Penna, C., Cassán, F., and Luna, V. (2007). Phytohormone production by three strains of Bradyrhizobium japonicum and possible physiological and technological implications. Applied Microbiology and Biotechnology 74, 874-880.

Bottini, R., Cassan, F., and Piccoli, P. (2004). Gibberellin production by bacteria and its involvement in plant growth promotion and yield increase. Applied Microbiology Biotechnology 65, 497-503.

Cyr, A., Wilderman, P.R., Determan, M., and Peters, R.J. (2007). A Modular Approach for Facile Biosynthesis of Labdane-Related Diterpenes. Journal of the American Chemical Society 129, 6684-6685.

Den Herder, G., and Parniske, M. (2009). The unbearable naivety of legumes in symbiosis. Current opinion in plant biology 12, 491-499.

Fauvart, M., and Michiels, J. (2008). Rhizobial secreted proteins as determinants of host specificity in the rhizobium-legume symbiosis. FEMS microbiology letters 285, 1-9.

Flores, M., Morales, L., Avila, A., Gonzalez, V., Bustos, P., Garcia, D., Mora, Y., Guo, X., Collado-Vides, J., Pinero, D., Davila, G., Mora, J., and Palacios, R. (2005). Diversification of DNA sequences in the symbiotic genome of Rhizobium etli. Journal of Bacteriology 187, 7185-7192.

Freiberg, C., Fellay, R., Bairoch, A., Broughton, W.J., Rosenthal, A., and Perret, X. (1997). Molecular basis of symbiosis between Rhizobium and legumes. Nature 387, 394-401.

Galibert, F., Finan, T.M., Long, S.R., Puhler, A., Abola, P., Ampe, F., Barloy- Hubler, F., Barnett, M.J., Becker, A., Boistard, P., Bothe, G., Boutry, M., Bowser, L., Buhrmester, J., Cadieu, E., Capela, D., Chain, P., Cowie, A., Davis, R.W., Dreano, S., Federspiel, N.A., Fisher, R.F., Gloux, S., Godrie, T., Goffeau, A., Golding, B., Gouzy, J., Gurjal, M., Hernandez-Lucas, I., Hong, A., Huizar, L., Hyman, R.W., Jones, T., Kahn, D., Kahn, M.L., Kalman, S., Keating, D.H., Kiss, E., Komp, C., Lelaure, V., Masuy, D., Palm, C., Peck, M.C., Pohl, T.M., Portetelle, D., Purnelle, B., Ramsperger, U., Surzycki, R., Thebault, P., Vandenbol, M., Vorholter, F.J., Weidner, S., Wells, D.H., Wong, K., Yeh, K.C., and Batut, J. (2001). The composite genome of the legume symbiont Sinorhizobium meliloti. Science 293, 668-672.

Gonzalez, V., Santamaria, R.I., Bustos, P., Hernandez-Gonzalez, I., Medrano-Soto, A., Moreno-Hagelsieb, G., Janga, S.C., Ramirez, M.A., Jimenez-Jacinto, V., Collado-Vides, J., and Davila, G. (2006). The partitioned Rhizobium etli genome: genetic and metabolic redundancy in seven interacting replicons. Proceedings of the National Academy of Sciences 103, 3834-3839.

Gonzalez, V., Acosta, J.L., Santamaria, R.I., Bustos, P., Fernandez, J.L., Hernandez Gonzalez, I.L., Diaz, R., Flores, M., Palacios, R., Mora, J., and Davila, G. (2010). Conserved symbiotic plasmid DNA sequences in the multireplicon pangenomic structure of Rhizobium etli. Applied Environmental Microbiology 76, 1604-1614.

Hauser, F., Pessi, G., Friberg, M., Weber, C., Rusca, N., Lindemann, A., Fischer, H.- M., and Hennecke, H. (2007). Dissection of the Bradyrhizobium japonicum NifA+sigma54 regulon, and identification of a ferredoxin gene (fdxN) for symbiotic nitrogen fixation. Molecular Genetics and Genomics 278, 255-271.

Hedden, P., Phillips, A.L., Rojas, M.C., Carrera, E., and Tudzynski, B. (2001). Gibberellin Biosynthesis in Plants and Fungi: A Case of Convergent Evolution? Journal of Plant Growth Regulation 20, 319-331.

Johnston, A.W.B., Beynon, J.L., Buchanan-Wollaston, A.V., Setchell, S.M., Hirsch, P.R., and Beringer, J.E. (1978). High frequency transfer of nodulating ability between strains and species of Rhizobium. Nature 276, 634-636.

Jones, K.M. (2012). Increased production of the exopolysaccharide succinoglycan enhances Sinorhizobium meliloti 1021 symbiosis with the host plant Medicago truncatula. J Bacteriol 194, 4322-4331.

Jones, K.M., Kobayashi, H., Davies, B.W., Taga, M.E., and Walker, G.C. (2007). How rhizobial symbionts invade plants: the Sinorhizobium-Medicago model. Nature reviews. Microbiology 5, 619-633.

Kaneko, T., Nakamura, Y., Sato, S., Minamisawa, K., Uchiumi, T., Sasamoto, S., Watanabe, A., Idesawa, K., Iriguchi, M., Kawashima, K., Kohara, M., Matsumoto, M., Shimpo, S., Tsuruoka, H., Wada, T., Yamada, M., and Tabata, S. (2002). Complete Genomic Sequence of Nitrogen-ﬁxing Symbiotic Bacterium Bradyrhizobium japonicum USDA110. DNA Research 9, 189-197.

Kaneko, T., Nakamura, Y., Sato, S., Asamizu, E., Kato, T., Sasamoto, S., Watanabe, A., Idesawa, K., Ishikawa, A., Kawashima, K., Kimura, T., Kishida, Y., Kiyokawa, C., Kohara, M., Matsumoto, M., Matsuno, A., Mochizuki, Y., Nakayama, S., Nakazaki, N., Shimpo, S., Sugimoto, M., Takeuchi, C., Yamada, M., and Tabata, S. (2000). Complete genome structure of the nitrogen- fixing symbiotic bacterium Mesorhizobium loti. DNA Research 7, 331-338.

Kawasaki, T., Kuzuyama, T., Kuwamori, Y., Matsuura, N., Itoh, N., Furihata, K., Seto, H., and Dairi, T. (2004). Presence of copalyl diphosphate synthase gene in an actinomycete possessing the mevalonate pathway. J. Antibiot. 57, 739-747.

Keister, D.L., Tully, R.E., and Van Berkum, P. (1999). A cytochrome P450 gene cluster in the Rhizobiaceae. The Journal of general and applied microbiology 45, 301-303.

Mavingui, P., Flores, M., Guo, X., Davila, G., Perret, X., Broughton, W.J., and Palacios, R. (2002). Dynamics of genome architecture in Rhizobium sp. strain NGR234. J Bacteriol 184, 171-176.

Morrone, D., Lowry, L., Determan, M.K., Hershey, D.M., Xu, M., and Peters, R.J. (2010). Increasing diterpene yield with a modular metabolic engineering system in E. coli: comparison of MEV and MEP isoprenoid precursor pathway engineering. Appl. Microbiol. Biotechnol. 85, 1893-1906.

Morrone, D., Chambers, J., Lowry, L., Kim, G., Anterola, A., Bender, K., and Peters, R.J. (2009). Gibberellin biosynthesis in bacteria: Separate ent-copalyl diphosphate and ent-kaurene synthases in Bradyrhizobium japonicum. FEBS Lett. 583, 475-480.

Oke, V., and Long, S.R. (1999). Bacteroid formation in the Rhizobium-legume symbiosis. Current opinion in microbiology 2, 641-646.

Oono, R., Schmitt, I., Sprent, J.I., and Denison, R.F. (2010). Multiple evolutionary origins of legume traits leading to extreme rhizobial differentiation. New Phytol 187, 508-520.

Pessi, G., Ahrens, C.A., Rehrauer, H., Lindemann, A., Hauser, F., Fischer, H.-M., and Hennecke, H. (2007). Genome-wide transcript analysis of Bradyrhizobium japonicum bacteroids in soybean root nodules. Mol. Plant Microbe Interact. 20, 1353-1363.

Peters, R.J. (2010). Two rings in them all: The labdane-related diterpenoids. Nat. Prod. Rep. 27, 1521-1530.

Salazar, E., Diaz-Mejia, J.J., Moreno-Hagelsieb, G., Martinez-Batallar, G., Mora, Y., Mora, J., and Encarnacion, S. (2010). Characterization of the NifA-RpoN regulon in Rhizobium etli in free life and in symbiosis with Phaseolus vulgaris. Appl Environ Microbiol 76, 4510-4520.

Schmeisser, C., Liesegang, H., Krysciak, D., Bakkou, N., Le Quere, A., Wollherr, A., Heinemeyer, I., Morgenstern, B., Pommerening-Roser, A., Flores, M., Palacios, R., Brenner, S., Gottschalk, G., Schmitz, R.A., Broughton, W.J., Perret, X., Strittmatter, A.W., and Streit, W.R. (2009). Rhizobium sp. strain NGR234 possesses a remarkable number of secretion systems. Appl Environ Microbiol 75, 4035-4045.

Schneiker-Bekel, S., Wibberg, D., Bekel, T., Blom, J., Linke, B., Neuweger, H., Stiens, M., Vorholter, F.J., Weidner, S., Goesmann, A., Puhler, A., and Schluter, A. (2011). The complete genome sequence of the dominant Sinorhizobium meliloti field isolate SM11 extends the S. meliloti pan-genome. Journal of biotechnology 155, 20-33.

Sullivan, J.T., and Ronson, C.W. (1998). Evolution of rhizobia by acquisition of a 500- kb symbiosis island that integrates into a phe-tRNA gene. Proc Natl Acad Sci U S A 95, 5145-5149.

Tully, R.E., and Keister, D.L. (1993). Cloning and Mutagenesis of a Cytochrome P-450 Locus from Bradyrhizobium japonicum That Is Expressed Anaerobically and Symbiotically. Appl. Environ. Microbiol. 59, 4136-4142.

Tully, R.E., van Berkum, P., Lovins, K.W., and Keister, D.L. (1998). Identifcation and sequencing of a cytochrome P450 gene cluster from Bradyrhizobium japonicum. Biochim. Biophys. Acta 1398, 243-255.

Vance, C.P. (2001). Symbiotic nitrogen fixation and phosphorus acquisition. Plant nutrition in a world of declining renewable resources. Plant Physiol 127, 390-397.

Viprey, V., Rosenthal, A., Broughton, W.J., and Perret, X. (2000). Genetic snapshots of the Rhizobium species NGR234 genome. Genome biology 1, RESEARCH0014.

Weidner, S., Baumgarth, B., Gottfert, M., Jaenicke, S., Puhler, A., Schneiker-Bekel, S., Serrania, J., Szczepanowski, R., and Becker, A. (2013). Genome Sequence of Sinorhizobium meliloti Rm41. Genome announcements 1.

Yang, S., Tang, F., Gao, M., Krishnan, H.B., and Zhu, H. (2010). R gene-controlled host specificity in the legume-rhizobia symbiosis. Proc Natl Acad Sci U S A 107, 18735-18740.

Young, J.P., Crossman, L.C., Johnston, A.W., Thomson, N.R., Ghazoui, Z.F., Hull, K.H., Wexler, M., Curson, A.R., Todd, J.D., Poole, P.S., Mauchline, T.H., East, A.K., Quail, M.A., Churcher, C., Arrowsmith, C., Cherevach, I., Chillingworth, T., Clarke, K., Cronin, A., Davis, P., Fraser, A., Hance, Z., Hauser, H., Jagels, K., Moule, S., Mungall, K., Norbertczak, H., Rabbinowitsch, E., Sanders, M., Simmonds, M., Whitehead, S., and Parkhill, J. (2006). The genome of Rhizobium leguminosarum has recognizable core and accessory components. Genome biology 7, R34.

Table

Table 1. Rhizobial KS and CPS homologs % Id to % Id to Genus species KS BjKS CPS BjCPS Bradyrhizobium japonicum NP_768790 – NP_768789 – elkanii WP_018270013 91 WP_016845990 92 WSM1253 WP_007600190 89 WP_007600189 91 WSM471 WP_007605894 88 WP_007605892 91 Mesorhizobium loti NP_106894 93 NP_106893 93 alhagi WP_008838313 94 WP_008838314 95 amorphae WP_006204703 93 WP_006204702 93 ciceri YP_004144785 92 YP_004144784 92 STM 4661 WP_006329103 92 WP_006329109 93 WSM4349 WP_018457688 92 WP_018457687 93 Sinorhizobium fredii NP_443948 92 NP_443949 95 meliloti WP_018098888 91 WP_018098887 94 medicae WP_018009727 90 WP_018009726 92 Rhizobium etli NP_659792 87 NP_659791 86 tropici YP_007335933 87 YP_007335932 90 CCGE 510 WP_007636919 88 WP_007636921 87 grahamii WP_016558477 71 WP_016558476 72 mesoamericanum WP_007539161 69 WP_007539159 72

Figures

OPP OPP H

CPS H KS H GGPP ent-CPP ent-kaurene

Figure 1. Production of ent-kaurene by B. japonicum. Shown are the steps catalyzed by the characterized ent-copalyl diphosphate synthase (CPS) and ent-kaurene synthase (KS).

Figure 2. GC-MS selected ion (m/z = 272) chromatograms demonstrating production of ent-kaurene from GGPP by co-expressing CPS and KS from (A) M. loti, (B) S. fredii, and (C) R. etli, along with (D) an authentic standard (from co-expression of the CPS and KS from Arabidopsis thaliana) in E. coli (along with a GGPP synthase).

B. japonicum CYP112 CYP114 Fd SDR CYP117 GGPS CPS KS

S. fredii CYP112 CYP114 Fd SDR CYP117 GGPS CPS KS

M. loti CYP112 CYP114 Fd-SDR CYP117 GGPS CPS KS IDI MFS

R. etli CYP112 CYP114-Fd SDR CYP117 GGPSp CPS KS IDI GGPS2

Figure 3. Schematic of diterpenoid biosynthesis operon from designated rhizobia (gene designations as described in the text).

Figure 4. GC-MS selected ion (m/z = 272) chromatograms demonstrating production of ent-kaurene from S. meliloti 1021 expressing the GGPS-CPS-KS, but not CPS-KS alone, from S. fredii NGR234.

Figure 5. Molecular phylogenetic analysis of (A) genes for the characterized rhizobial CPS and (B) genes for the nitrogenase sub-unit NifK from the same rhizobia. SsCPS and AzNifK are the out-group sequences, as described in the text.

CHAPTER 4: PHYSIOLOGICAL FUNCTION AND METABOLITES LEVEL OF

DITERPENOIDS IN RICE PLANT DEFENSE

Xuan Lu1, Meimei Xu1, Bo Liu2, Julio Rodríguez-Romero3, Ane Sesma3, Dangping Luo2,

Bing Yang2 and Reuben J. Peters1,*

1Department of Biochemistry, Biophysics, and Molecular Biology, Iowa State University,

Ames, IA 50011, USA

2Department of Genetics, Development and Cell Biology, Iowa State University, Ames,

IA 50011, USA

3Centre for Plant Biotechnology and Genomics (CBGP), Universidad Politecnica de

Madrid, Campus de Montegancedo, 28223 Pozuelo de Alarcon, Madrid, Spain

Abstract

Among their responses to microbial infection, plants deploy an arsenal of antibiotic natural products. While these have been identified on the basis of their antibiotic activity in vitro, this leaves open the question of their relevance to defense in planta. The vast majority of such natural products from the important crop plant rice

(Oryza sativa) are diterpenoids whose biosynthesis proceeds via either ent- or syn-copalyl diphosphate (CPP) intermediates, and which were isolated on the basis of their antibiotic activity against the fungal blast pathogen Magnaporthe oryzae. Rice plants in which the syn-CPP synthase (Os-CPS4) is knocked out do not exhibit increased susceptibility to M. oryzae, instead exhibiting decreased allelopathy. Here, rice plants in which the ent-CPP

66 synthase Os-CPS2 is knocked out are found to exhibit increased susceptibility to M. oryzae, but also to the bacterial leaf blight pathogen Xanthomonas oryzae pv. oryzae.

Nevertheless, based on the observation that rice plants direct significant metabolite levels towards syn-CPP derived diterpenoids upon elicitation, further investigation revealed that

Os-CPS4 does play a role in non-host disease resistance. Altogether, these results establish the relevance of diterpenoids to rice plant defense, and also provide insight into their metabolite levels, which seems to have evolved to balance production according to a complex range of physiological roles.

Introduction

Plants deploy an arsenal of antimicrobial natural products during pathogen infection (Bednarek and Osbourn, 2009). This is one means of effective means of defense. However, the underlying studies mostly rely on demonstrated antibiotic activity in vitro, with such antibiotics whose production is strongly induced by infection termed phytoalexins, while those constitutively present are termed phytoanticipans (VanEtten et al., 1994). The relevance of such natural products to plant defense is most rigorously supported by genetic evidence using gene knockout lines in which the compound(s) of interest can no longer be produced, but the remainder of the plant defense response has been retained. Such evidence has been provided in at least a few cases (Ahuja et al.,

2012). For example, camalexin and certain sesquiterpenes in Arabidopsis, as well as triterpenoid saponins in oats, all clearly play a role in plant defense against infection by certain pathogens (Papadopoulou et al., 1999; Thomma et al., 1999; Huang et al., 2012).

Nevertheless, in other cases such genetic analysis does not support the relevance of at

67 least some phytoalexins to the plant defense response. Of particular relevance here, despite being the first identified rice phytoalexins against the fungal blast pathogen

Magnaporthe oryzae (Cartwright et al., 1981), the momilactones do not play an important role in rice plant defense (Xu et al., 2012).

Almost all of the identified antibiotic natural products from the important crop plant rice (Oryza sativa) are diterpenoids (Schmelz et al., 2014). In particular, labdane- related diterpenoids and its biosynthesis can be traced back to that of the gibberellin phytohormones (Peters, 2010). Most of these compounds, specifically momilactones A &

B, oryzalexins A-F, oryzalexin S and phytocassanes A-F, were originally isolated and identified on the basis of their activity against M. oryzae (Peters, 2006) (Fig. 1). Infection by M. oryzae further strongly induces accumulation of these labdane-related diterpenoids

(Hasegawa et al., 2010), which then seem to be phytoalexins. The remaining diterpenoid antibiotics, including oryzalide A and B, oryzalic acid A and B and oryzadione I to III

(Toyomasu, 2008), all of which are termed oryzalide-related here, were isolated and identified on the basis of their activity against the bacterial leaf blight pathogen

Xanthomonas oryzae pv. oryzae (Xoo). The production of these labdane-related diterpenoids is only weakly induced by Xoo infection (Watanabe et al., 1996), and they then seem to be phytoanticipans.

As with the vast majority of phytoalexins, these rice labdane-related diterpenoids have been assumed to be relevant to plant defense on the basis of their antibiotic activity in vitro. However, it should be noted that the momilactones were actually first isolated on the basis of their plant growth suppressing activity (Kato et al., 1973), and are constitutively secreted from the roots into the soil, where they have long been suggested

68 to act as allelochemicals, most specifically momilactone B (Kato-Noguchi and Peters,

2013). This offers an alternative rationale for the production of at least the momilactones.

Nevertheless, momilactones, as well as phytocassanes, accumulate in leaves upon infection by M. oryzae (Hasegawa et al., 2010). In addition, both the momilactones and phytocassanes have recently been shown to accumulate specifically at the site of Xoo infection in rice leaves (Klein et al., 2015).

The defining characteristic of labdane-related diterpenoids is their biosynthetic origin from bicyclization of the general diterpenoid precursor (E,E,E)-geranylgeranyl diphosphate (GGPP) catalyzed by class II diterpene cyclases (Peters, 2010). This most often results in the production of the eponymous labdadienyl/copalyl diphosphate (CPP), with the relevant enzymes then termed CPP synthases (CPSs; Fig. 1). All vascular plants seem to produce the labdane-related diterpenoid gibberellins as phytohormones, and this depends on the production of ent-CPP (Hedden and Thomas, 2012). Rice produces two different stereoisomers of CPP, both the requisite ent-CPP, and the diastereomer syn-

CPP. The production of these is catalyzed by three distinct CPSs in rice. Two of these produce ent-CPP, both Os-CPS1 and Os-CPS2, while the third, Os-CPS4, produces syn-

CPP (Otomo et al., 2004; Prisic et al., 2004; Xu et al., 2004). Of the two ent-CPP producing CPSs, only Os-CPS1 seems to be involved in gibberellin metabolism

(Sakamoto et al., 2004). On the other hand, transcript levels of Os-CPS2 and Os-CPS4 are inducible, and both have been suggested to play a role in the production of more specialized labdane-related diterpenoids (Peters, 2006). The diterpenes whose hydrocarbon structures define the various families of derived diterpenoids are produced from ent- or syn-CPP by class I diterpene synthases (Fig. 1), which are related to the ent-

69 kaurene synthase (KS) required for gibberellin phytohormone biosynthesis and have been termed kaurene synthase-like (KSL).

Given the presence of two ent-CPP producing CPSs, previous work has focused on analysis of the syn-CPP derived labdane-related diterpenoids in rice, whose biosynthesis depends only on Os-CPS4. However, it must be noted that, of the identified antibiotic labdane-related diterpenoids from rice, only momilactones A & B and oryzalexin S are derived from syn-CPP, while the rest are derived from ent-CPP (Fig. 1).

Analysis of a T-DNA insertion mutant of Os-CPS4 has indicated that these knockout Os- cps4 plants are not more susceptible to infection with M. oryzae than the parental/wild- type (WT) line. Additional analysis, including the use of a more specific Os-ksl4 gene knockout line, indicates that the momilactones act as allelochemicals (Xu et al., 2012).

However, another study, using rice plants in which Os-CPS4 expression is knocked- down, but not completely abolished, suggests that OsCPS4-dependent labdane-related diterpenoids do play some role in defense against M. oryzae (Toyomasu et al., 2014).

This discrepancy may reflect the different genetic backgrounds of the rice lines and fungal strains used in each of these studies. In particular, it has been noted that susceptible cultivars produce less momilactones and phytocassanes than resistant cultivars, at least at early time points (Hasegawa et al., 2010; Klein et al., 2015), and Os- cps4 is derived from a susceptible line (ssp. japonica cv. Zhonghua 11). Accordingly, such reduced and/or delayed production (i.e., even in the parental/WT line) may underlie the observed lack of effect with Os-cps4 plants.

Here evidence is provided indicating that Os-CPS4 dependent diterpenoids (i.e., those derived from syn-CPP) are not relevant to rice plant defense against the bacterial

70 pathogen Xoo. Indeed, Os-cps4 plants are more resistant to infection by Xoo. On the other hand, Os-cps2 plants are more susceptible to the bacterial leaf blight pathogen Xoo, as well as the fungal blast pathogen M. oryzae, indicating that the diterpenoids dependent on

Os-CPS2 (i.e., derived from ent-CPP) are relevant to plant defense. Interestingly, diterpene olefin analysis also indicates increased metabolic flux towards the Os-CPS4 dependent branch of labdane-related diterpenoid metabolism in Os-cps2 plants. On the other hand, there is more metabolic flux towards the ent-CPP derived branch of labdane- related diterpenoid metabolism in Os-cps4 plants, which presumably explains the increased resistance of Os-cps4 rice to infection by Xoo. Nevertheless, Os-cps4 plants are more susceptible to Magnaporthe poae, a fungal pathogen that normally does not infect rice, indicating a role for Os-CPS4 in non-host disease resistance. Accordingly, our results provide insight into the physiological role of diterpenoids and their metabolic flux in the microbial defense response of the important crop plant rice.

Results

Characterization of Os-cps2

The production of ent-CPP for more specialized diterpenoid metabolism in rice has been hypothesized to largely depend on Os-CPS2. Based on the availability of an Os-

CPS2 insertion mutant, a reverse genetics approach was undertaken to investigate the biological function of the ent-CPP derived diterpenoids. A homozygous Os-cps2 line was obtained, as identified by PCR verification of the T-DNA insertion into the sixth exon of the gene (Fig. S1). As expected, Os-CPS2 transcripts could no longer be detected in these plants, even following induction with methyl jasmonate (Me-JA) (Fig. 2A). Me-JA has

71 been shown to induce accumulation of transcripts for both Os-CPS2 and Os-CPS4 (Prisic et al., 2004; Xu et al., 2004), along with the derived labdane-related diterpenes (Morrone et al., 2011), and plays a role in induction of diterpenoid phytoalexins (Riemann et al.,

2013; Shimizu et al., 2013), as well as rice plant defense more generally (Yang et al.,

2013). These Os-cps2 plants exhibited only minor reductions in growth (≤ 10%), largely consistent with the previously suggested predominant role of Os-CPS1 in gibberellin metabolism. In addition, following induction with Me-JA, there was a clear reduction in the levels of several ent-CPP derived diterpenes in these plants relative to its parental/WT line (ssp. japonica cv. Nipponbare), most notably ent-sandaracopimaradiene (Fig. 2B).

Susceptibility of Os-cps2 to M. oryzae and X. oryzae pv. oryzae

Given that some ent-CPP dependent diterpenoids have been suggested to serve as phytoalexins against M. oryzae (Peters, 2006), while others have been suggested to be phytoanticipans against Xoo (Schmelz et al., 2014), the susceptibility of Os-cps2 rice relative to its parental/WT line to both microbial pathogens was investigated. Infecting leaves with a rice isolate of M. oryzae (Guy 11) causes disease symptoms on both Os- cps2 and its parental/WT rice (Fig 3). However, there was a significant difference in the percentage of infected/necrotic area of leaves from Os-cps2 plants (9.5 ± 0.8 %), relative to that for its parental/WT line (2.9 ± 0.7 %), indicating that this mutant is more susceptible to M. oryzae and, hence, ent-CPP derived diterpenoids seem to serve an effective role in rice plant defense against this fungal pathogen (p < 0.001 in an unpaired student’s t-test). Moreover, when infected with Xoo via a leaf clip method Os-cps2 rice exhibited significantly longer lesion lengths than its parental/WT line (p < 0.01 in an

72 unpaired student’s t-test) (Fig. 4), indicating that this mutant also is more susceptible to

Xoo. Thus, ent-CPP derived diterpenoids seem to serve an effective role in rice plant defense against this bacterial pathogen as well. This is consistent with the known anti- fungal activity of the ent-cassadiene-derived phytocassanes, and anti-bacterial activity of the oryzalide-related diterpenoids, which are potentially derived from ent-kaurene, and certainly its endocyclic double bond isomer ent-isokaurene (Fig. S2).

Characterization of Os-CPS2 over-expressing plants

In order to provide further evidence for the relevance of ent-CPP-derived diterpenoids to rice defense against microbial pathogens, plants that over-express Os-

CPS2 were generated. Three lines whose enhanced expression of Os-CPS2 was verified by RT-PCR were then selected for further characterization (Fig. S3). Critically, these Os-

CPS2OE plants exhibited normal growth and development, indicating that their gibberellin phytohormone levels were not affected. Nevertheless, there was a clear increase in the levels of ent-CPP derived diterpenes, most notably ent-kaurene, following treatment with Me-JA (Fig. 5A). Indeed, even without Me-JA treatment, increased levels of ent-kaurene and ent-isokaurene were observed (Fig. 5B). Given that these are precursors to the oryzalide-related anti-bacterial phytoanticipans (Fig. S2), plants from all three Os-CPS2OE lines were infected with Xoo, along with parental/WT plants (ssp. japonica cv. Kitaake). The Os-CPS2OE plants exhibited significantly shorter lesion lengths relative to its parental/WT line (p < 0.01 in an unpaired t-test), indicating increased resistance to bacterial leaf blight (Fig. 5C). This result supports a role for ent-

CPP derived diterpenoids as anti-bacterial agents in rice plant defense, at least against

Xoo, consistent with the previously known anti-Xoo activity of the oryzalide-related diterpenoids.

Resistance of Os-cps4 to X. oryzae pv. oryzae infection

The results with Os-CPS2OE suggest that increased metabolite levels in ent-CPP derived diterpenoids increases resistance to Xoo. Examination of the Os-cps2 diterpene content suggested that these plants produce increased amounts of syn-CPP derived diterpenes (Fig. 2B). Hypothesizing that the converse might occur with Os-cps4 (Fig. 1), the diterpene content of this mutant was analyzed and compared to that of its parental/WT line (ssp. japonica cv. Zhonghua 11). Indeed, there is an increase in ent-CPP derived diterpenes upon treatment with Me-JA, particularly ent-kaurene and ent- isokaurene, in Os-cps4 versus its parental/WT plants (Fig. 6A). It has already been reported that Os-cps4 plants are no more susceptible to infection with M. oryzae, as well as the rice fungal pathogen Fusarium fujikuroi, than its parental/WT line (Xu et al.,

2012). Given the resistance to Xoo observed with Os-CPS2OE, analogous infections were carried out with Os-cps4 and its parental/WT line. Intriguingly, Os-cps4 plants exhibited significantly shorter lesion length relative to those from its parental/WT line (p < 0.01 in an unpaired student’s t-test), indicating that this mutant has increased resistance to the bacterial leaf blight disease (Fig. 6B). Again, this supports not only the known anti-Xoo activity of the oryzalide-related diterpenoids, but also, particularly given the dramatic increase observed here, a role for ent-kaurene as a precursor – e.g., to orzyalide B (Fig.

S2).

Characterizing metabolite levels

Beyond the various labdane-related diterpenes shown in Fig. 1, the targeted metabolite analysis employed here also found the more general diterpene casbene.

Casbene was directly formed from GGPP, and its was recently identified as an intermediate in rice diterpenoid phytoalexin (+)-10-oxodepressin biosynthesis (Inoue et al., 2013). Notably, both Os-cps2 and Os-cps4 plants accumulate increased levels of casbene relative to their parental/WT lines (Figs. 2B & 6A). Given the contrasting susceptibilities of these two mutants to infection, this suggests that the derived (+)-10- oxodepressin also is not a major component of rice plant defense against M. oryzae or

Xoo. Nevertheless, it also indicates that there is a distinct pool of GGPP that is tapped for more specialized diterpenoid metabolism. This seems to be distinct from that utilized for gibberellin biosynthesis given the lack of effect of either knocking out or over-expressing

Os-CPS2 on plant growth and development, or that utilized in the production of photosynthetic pigments, the latter of which presumably reflects the predominant metabolic fate for GGPP overall. These distinct pools may reflect the separation of these various metabolic processes into distinct tissues, as Os-CPS1, required for gibberellin metabolism, is expressed in the vascular bundle, while the inducible Os-CPS2 is largely expressed in epidermal cells, separate from the mesophyll where photosynthesis is localized (Toyomasu et al., 2015).

Also noteworthy is the presence in uninduced leaves of the endo/exo double-bond isomers of syn-stemarene (peak 9) and syn-stemodene (peak 10) produced by Os-KSL8

(Fig. 5B) (Xu et al., 2007). Besides that, there are small amounts of syn-pimaradiene, rather than the ent-isokaurene and/or ent-kaurene that might have been expected from the

75 phytoanticipans nature of the derived oryzalide-related diterpenoids (Fig. 5B). The constitutive accumulation of syn-stemarene/stemodene is somewhat surprising as Os-

KSL8 mRNA levels are clearly inducible (Nemoto et al., 2004). While the only known derived diterpenoid, oryzalexin S (Fig. 1), has been defined as a phytoalexin rather than phytoanticipan (Tamogami et al., 1993), this is only a single metabolite, derived from the endo-syn-stemarene major product, and alternative metabolic fates for this seem possible, along with the unknown fates for the other Os-KSL8 products. Thus, it seems likely that there is some other role for the syn-stemarene/stemodene derived diterpenoids. Besides that, consistent with the previous report (Morrone et al., 2011), we can also detect the presence of syn-labdatriene in CPS4WT (Fig. S5),

Quantifying redirected metabolites

The results reported above indicate not only that ent-CPP derived diterpenoids are effective components of rice plant defense against both the fungal pathogen M. oryzae and the bacterial pathogen Xoo, but also that increasing flux to ent-CPP derived diterpenoids can enhance resistance. Hypothesizing that closer examination of the altered flux exhibited by these various lines might offer further insight into the specific roles of the resulting diterpenoids, the content of the defining diterpene precursors in each line relative to its parental/WT plants was quantified. As expected, no syn-CPP derived diterpenes are observed in Os-cps4 plants, and there seems to be a general increase in ent-

CPP derived diterpenes, although this is not significant for ent-cassadiene (Fig. 7A). The increase in levels of ent-isokaurene and ent-kaurene is consistent with the anti-bacterial activity assigned to the derived oryzalide-related diterpenoids. Despite their assignment

76 as phytoanticipans it has been noted that there is a small increase (< 4-fold) in the level of these compounds in response to Xoo infection (Watanabe et al., 1996), consistent with the increased levels of ent-(iso)kaurene observed here following induction. However, it is also possible that the observed resistance may be due, at least in part, to increased levels of other diterpenoids – e.g., oryzalexins A–F, which are derived from ent- sandaracopimaradiene (Fig. 1), the accumulation of which is also significantly increased in Os-cps4 plants.

Consistent with the continuing presence of Os-CPS1, which also produces ent-

CPP, Os-CPS1 derived diterpenes can still be found in Os-cps2 plants, albeit with some reduction in their amounts, along with some increase in the amounts of syn-CPP derived diterpenes, although this is again not significant in all cases (Fig. 7B). Perhaps most intriguing is the lack of any detectable ent-sandaracopimaradiene, the precursor to oryzalexins A–F, as well as the insignificant changes in levels of ent-isokaurene and ent- kaurene, the precursors to the only known anti-bacterial diterpenoids, the oryzalide- related phytoanticipans. The former hints at a role for oryzalexins A–F as phytoalexins against Xoo, while the latter suggests that production of the orzyalide-related diterpenoids does not strictly depend on Os-CPS2. In any case, the decreased resistance of Os-cps2 plants to infection, despite the increase in syn-CPP derived diterpenes, further emphasizes that the resulting phytoalexins (e.g., the momilactones and oryzalexin S) are not major components of rice plant defense against either M. oryzae or Xoo.

As noted above, overexpression of Os-CPS2 leads to striking increases in the level of ent-kaurene and ent-isokaurene in both uninduced and induced plants (Fig. 7C).

In addition, there are significant increases in the levels of all detectable ent-CPP derived

77 diterpenes upon induction (Fig. 7C). The normal development of the Os-CPS2OE plants indicates that there is no change in gibberellin phytohormone metabolism. This is consistent with previous results in Arabidopsis thaliana, where over-expression of the endogenous CPS led to similar increases in ent-kaurene, but no increase in bioactive gibberellins (Fleet et al., 2003). The alternative production of oryzalide-related diterpenoids from ent-(iso)kaurene provides another metabolic fate for these diterpenes, and presumably underlies the increased resistance of Os-CPS2OE plants to Xoo, but does not rule out a role for the other ent-CPP derived diterpenoids in rice plant defense against this bacterial pathogen.

To provide further insight into the observed diterpene accumulation patterns, the expression of all rice KSLs, as well as Os-CPS2 and Os-CPS4 in response to Xoo infection was examined (Fig. S4). The observed accumulation of ent-kaurene and ent- isokaurene in uninduced rice plants indicates that Os-KS and Os-KSL6, respectively, should be constitutively expressed. Indeed, transcripts of both genes are easily detected in uninduced rice leaves. However, despite the lack of accumulation of its ent-pimaradiene product, Os-KSL5 transcripts are also clearly present in uninduced rice leaves as well.

The increased susceptibility of Os-cps2 plants indicates a role for Os-CPS2 in defense, including against Xoo, and it was found that Os-CPS2 transcripts accumulate following

Xoo infection, along with those for Os-KSL7 and Os-KSL10. By contrast, transcripts for

Os-KS1 and Os-KSL6 are not induced, consistent with assignment of the resulting ent-

(iso)kaurene derived oryzalide-related diterpenoids as phytoanticipans. The lack of induction of Os-KSL5 transcripts by Xoo infection suggests that the diterpenoids derived from ent-pimaradiene also may serve as phytoanticipans. Perhaps somewhat surprisingly,

78 despite the decreased susceptibility of Os-cps4 plants indicating that syn-CPP derived diterpenoids do not play a role in defense, not only transcripts for Os-CPS4, but also the subsequently acting Os-KSL4 and the closely related Os-KSL8 & 11, as well as those for

Os-KSL10 that can act on both ent- and syn-CPP, accumulate following Xoo infection.

A role for Os-cps4 in non-host disease resistance

Strikingly, the results presented above indicate that syn-CPP derived diterpenoids do not have a role in defense against either M. oryzae or Xoo. Indeed, the only role indicated for these is as allelochemicals (Xu et al., 2012), which raises the question of why Os-CPS4 transcripts accumulate in response to either elicitation with the fungal cell wall component chitin (Okada et al., 2007), or the defense signaling molecule Me-JA (Xu et al., 2004). Given that allelopathic activity has been convincingly assigned to the Os-

KSL4 dependent momilactones (Xu et al., 2012), it is also unclear why rice would constitutively accumulate the syn-stemarene/stemodene products of Os-KSL8. It has been shown that rice is resistant to a number of other fungal pathogens from the Magnaporthe genus other than rice isolates of M. oryzae (Faivre-Rampant et al., 2008). Hypothesizing that Os-CPS4 might be a factor in such non-host disease resistance, a range of M. oryzae pathovars (isolated from alternative host plants), and Magnaporthe species (M. grisea, M. rhizophila and M. poae) were tested. Given that M. rhizophila and M. poae are root- directed pathogens, and that M. oryzae can infect rice through the roots as well (Sesma and Osbourn, 2004), these infections were directed at the roots of Os-cps4 plants and its parental/WT line. As previously reported (Xu et al., 2012), no significant differences were observed with the Guy 11 strain (rice isolate) of M. oryzae (Fig. 8A). However,

79 with two other pathovars/strains of M. oryzae (G22, isolated from Eragrostis curvula, and JP15, isolated from Elusine coracana), as well as M. grisea (strain BR29, isolated from Digitaria sanguinalis) and M. poae (strain 73-1, isolated from wheat, Triticum aestivum), Os-cps4 exhibited varying increases in susceptibility relative to its parental/WT line (Fig. 8A), perhaps most strongly with M. poae (Fig. 8B). Given the constitutive accumulation of Os-KSL8 produced diterpenes, it is tempting to speculate that these serve as precursors to anti-fungal phytoanticipans. In any case, these results indicate a role for syn-CPP derived diterpenoids in the non-host disease resistance of rice, at least against fungi from the Magnaporthe genus, providing a role for Os-CPS4 in plant defense against microbial pathogens.

Discussion

The sheer number of rice diterpenoid phytoalexins isolated on the basis of antibiotic activity against M. oryzae raised questions regarding their relevance to plant defense. This was heightened by the finding that the Os-CPS4 dependent diterpenoids do not appear to play a significant role against M. oryzae, with only a role in allelopathy assignable to the more specifically Os-KSL4 dependent momilactones (Xu et al., 2012). It was then unclear what role other syn-CPP derived diterpenoids might play. Moreover, the role of the many ent-CPP derived diterpenoids also had not yet been proven.

Here, genetic manipulation of Os-CPS2 and Os-CPS4 was employed to demonstrate that, although Os-CPS4 further does not play a role in defense against the bacterial pathogen Xoo, Os-CPS2 is relevant to rice plant defense against both Xoo and the fungal pathogen M. oryzae. Nevertheless, the observed metabolic flux through Os-

CPS4 to syn-CPP derived diterpenes indicated that these must play some role in plant defense, and additional investigation indicates that the derived diterpenoids are important in non-host disease resistance against Magnaporthe species of fungi that are not adapted to rice. Indeed, M. oryzae has already been shown to metabolize momilactones

(Hasegawa et al., 2010), which presumably is at least partially responsible for its ability to infect rice. Thus, rice diterpenoid natural products seem to play a number of roles, including acting as antibiotics against adapted bacterial and fungal pathogens, as well as in non-host disease resistance, along with acting as allelochemicals. Accordingly, the complexity of rice diterpenoid metabolism seems to be matched by the variety of physiological roles played by these natural products.

More generally, the results reported here indicate that investigation of metabolic flux, enabled by biochemical analyses defining the relevant enzymatic genes, can inform reverse genetic approaches. In particular, these results suggest that evolution is unlikely to allow significant metabolic flux without an important physiological function. It is now of interest to determine if further dissection of the roles played by rice diterpenoids, and the clusters of biosynthetic genes that (partially) underlie the production of these natural products (Yamane, 2013; Zi et al., 2014), supports this broad hypothesis.

Materials and Methods

Chemicals

Unless otherwise noted, all molecular biology reagents were purchased from

Invitrogen (Carlsbad, CA, USA) and all other chemicals from Fischer Scientific

(Waltham, MA, USA).

Plant materials

Rice plants (Orzya sativa L. ssp. japonica cv. Nipponbare, Kitaake and Zhonghua

11) were cultivated to the 6th leaf stage in standard growth chambers cycling between 12 h light at 28 °C and 12 h dark at 24 °C.

The Os-cps2 T-DNA insertion mutant RGT1043_5.1 (insert in the sixth exon of

Os-CPS2; TIGR ID, LOC_ Os02g36210; Fig. S1A) was obtained from the UC-Davis

Transposable Element Insertion Database. This mutant is in the ssp. japonica cv.

Nipponbare background. Homozygous Os-cps2 plants were selected by RT-PCR of Os-

CPS2 (i.e., the absence of any cDNA) from the T1 generation. Homozygous Os-cps2 plants were verified in the T2 generation and propagated to provide seeds for the experiments described here.

The Os-cps4 T-DNA insertion mutant RMD_03Z11UZ33, and selection of a homozygous line, has been previously described (Xu et al., 2012). This mutant is in the cv Zhonghua 11 background.

Transcriptional expression analysis

RT-PCR was used to determine the level of gene expression following induction with methyl-jasmonate (Me-JA). For induction, leaves were detached from 1 month old plants, treated with 2 mL 0.1% Tween 20 + 0.2% Me-JA (vol/vol) per leaf and then incubated for 24 h, while control leaves were only treated with the 0.1% Tween 20 carrier solution.

Total RNA was extracted from these leaves 24 h after induction with Me-JA, using the RNeasy Plant Mini Kit (Qiagen). Template cDNA was generated from 500 ng

82 of DNAse-treated RNA using the iScript Reverse Transcription kit (BioRad, Hercules,

CA, USA), following the manufacturer’s instructions. This cDNA preparation was used as the template for PCR with gene-specific primers CPS2RT F and CPS2RT R.

Transcription Factor TFIIAγ5 (LOC_Os05g01710) was used as internal control for rice genes.

Over-expression vector construction and rice transformation

The open reading frame for Os-CPS2 was amplified from cDNA by PCR using the primer OE CPS2F and OE CPS2R, introducing unique restriction sites at the 5ʹ and 3ʹ ends, EcoRI and SpeI, respectively. The PCR fragment was cloned into the EcoRI ⁄ SpeI restriction sites of the binary vector pBY02, such that expression is under the control of a cauliflower mosaic virus (CaMV) 35S promoter. The resulting vector construct was transformed into Agrobacterium tumefaciens EHA105 by electroporation. Immature embryo derived callus cells of rice (cv. Kitaake) were transformed by adapting a previously described protocol (Hiei et al., 1994). Transformed calli and regenerated plantlets were selected on media containing hygromycin B, with the resulting plants grown to seed, and Os-CPS2 overexpression confirmed by PCR. Selection was carried out for three more generations to obtain non-segregating homozygous Os-CPS2OE plants. Three independent lines that exhibited strong over-expression of Os-CPS2 were selected for further analysis (Fig. S5).

Extraction and quantification of diterpenes

Rice plants were cultivated to the 6th leaf stage. For induction, detached leaves (3 g) were induced with 10 mL 0.2% Me-JA (in 0.1% Tween 20), wrapped with aluminum foil and incubated in the dark at room temperature (24 °C) for 72 h. The leaves were then frozen and ground to powder in liquid nitrogen, and the residue suspended in 200 mL hexanes, with the addition of 53 µg of the diterpene cembrene A (Sigma-Aldrich) as an internal standard (i.e., cembrene A is not found in rice) per g fresh leaf weight. This slurry was stirred for 12 hours at room temperature. The hexane was collected by filtration through Whatman filter paper and the residue was extracted twice more with

200 mL hexanes. The pooled 600 mL hexane extract was then dried by rotary evaporation. The residue was re-suspended in 12 mL hexanes. The organic extract was fractionated by passage over an open column filled with 20 mL silica gel (J.T. Baker, PA,

USA), which was then washed with a further 60 mL of hexanes, with 6 x 12 mL fractions collected in glass tubes throughout. Each fraction was dried under nitrogen gas, and the residue was re-suspended in 100 µL of hexanes for quantification.

The concentrated fractions were analyzed by gas chromatography (GC) carried out on an HP-5ms column (Agilent, 0.25 µm Film, 0.25 mm ID, Length 30 m) using a

Varian (Palo Alto, CA, USA) 3900 GC with detection via a Saturn 2100 ion trap mass spectrometer (MS) in electron ionization (70 eV) mode. Samples (1 µL) were injected in splitless mode at 50 °C and, after holding for 3 min at 50 °C, the oven temperature was raised at a rate of 18 °C/min to 194 °C. Then the oven temperature was raised at a rate of

0.1 °C/min from 194 °C to 195 °C. After that, the temperature was raised at a rate 20

°C/min to 300°C, where it was held for an additional 2 min. MS data from 90 to 300 m/z

84 was collected starting 10 min after injection until the end of the run. Diterpenes were identified by comparison of their retention times and mass spectra with those of authentic standards. For each sample, all fractions containing diterpenes were pooled together

(generally fractions 2-4) for relative quantification, with normalization accomplished by comparison of the total ion peak areas for the rice diterpenes to that of the internal standard.

Due to the partial overlap of certain diterpenes in the method described above, a longer temperature ramp was employed to separate these, particularly syn-labdatriene

(Fig. S5), as well as verify that the recently identified ent-beyerene product of Os-KSL2

(Tezuka et al., 2015) was not detected in our samples. This was accomplished by injecting 1 µL samples in splitless mode at 60 °C and, after holding for 3 min at 50 °C, the oven temperature was raised at a rate of 3 °C/min to 246 °C. After that, the temperature was raised at a rate 20 °C/min to 300°C, where it was held for an additional

2 min. MS data from 90 to 300 m/z was collected starting 5 min after injection until the end of the run. Again, diterpenes were normalized by comparison of their peak area to that of the internal standard.

Disease resistance assay

Magnaporthe infection assays were carried out as previously described (Sesma and Osbourn, 2004). Xoo infections were carried using the leaf-clipping method

(Kauffman et al., 1973), using the fully expanded leaves of 8-week-old rice plants and inoculating with Xoo strain PXO99A. Xoo was grown for 2-3 days on TSA (1% wt/vol

Tryptone, 1% sucrose, 0.1% glutamic acid, 1.5% agar, pH 6.6-6.8) plates at 28 °C,

scraped off and resuspended in sterile dH2O. This suspension was then adjusted to an optical density of 0.5 at 600 nm before use for infections. After clipping, lesion lengths consisting of grayish chlorotic coloration below the inoculated leaf tips along the main vein was measured 14 days after inoculation.

References

Ahuja, I., Kissen, R., and Bones, A.M. (2012). Phytoalexins in defense against pathogens. Trends Plant Sci 17, 73-90.

Bednarek, P., and Osbourn, A. (2009). Plant-microbe interactions: chemical diversity in plant defense. Science 324, 746-748.

Cartwright, D.W., Langcake, P., Pryce, R.J., Leworthy, D.P., and Ride, J.P. (1981). Isolation and characterization of two phytoalexins from rice as momilactones A and B. Phytochemistry 20, 535-537.

Faivre-Rampant, O., Thomas, J., Allegre, M., Morel, J.B., Tharreau, D., Notteghem, J.L., Lebrun, M.H., Schaffrath, U., and Piffanelli, P. (2008). Characterization of the model system rice--Magnaporthe for the study of nonhost resistance in cereals. New Phytol 180, 899-910.

Fleet, C.M., Yamaguchi, S., Hanada, A., Kawaide, H., David, C.J., Kamiya, Y., and Sun, T.-P. (2003). Overexpression of AtCPS and AtKS in Arabidopsis confers increased ent-kaurene production but no increase in bioactive gibberellins. Plant Physiol. 132, 830-839.

Hedden, P., and Thomas, S.G. (2012). Gibberellin biosynthesis and its regulation. Biochem J 444, 11-25.

Hiei, Y., Ohta, S., Komari, T., and Kumashiro, T. (1994). Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. The Plant journal : for cell and molecular biology 6, 271-282.

Huang, M., Sanchez-Moreiras, A.M., Abel, C., Sohrabi, R., Lee, S., Gershenzon, J., and Tholl, D. (2012). The major volatile organic compound emitted from

Arabidopsis thaliana flowers, the sesquiterpene (E)-beta-caryophyllene, is a defense against a bacterial pathogen. The New phytologist 193, 997-1008.

Inoue, Y., Sakai, M., Yao, Q., Tanimoto, Y., Toshima, H., and Hasegawa, M. (2013). Identification of a novel casbane-type diterpene phytoalexin, ent-10-oxodepressin, from rice leaves. Biosci Biotechnol Biochem 77, 760-765.

Kato, T., Kabuto, C., Sasaki, N., Tsunagawa, M., Aizawa, H., Fujita, K., Kato, Y., Kitahara, Y., and Takahashi, N. (1973). Momilactones, growth inhibitors from rice, Oryza sativa L. Tetrahedron Lett. 14, 3861-3864.

Kato-Noguchi, H., and Peters, R.J. (2013). The role of momilactones in rice allelopathy. J. Chem. Ecol. 39, 175-185.

Kauffman, H.E., Reddy, A.P.K., Hsieh, S.P.Y., and Merca, S.D. (1973). Improved Technique for Evaluating Resistance of Rice Varieties to Xanthomonas-Oryzae. Plant Dis Rep 57, 537-541.

Morrone, D., Hillwig, M.L., Mead, M.E., Lowry, L., Fulton, D.B., and Peters, R.J. (2011). Evident and latent plasticity across the rice diterpene synthase family with potential implications for the evolution of diterpenoid metabolism in the cereals. The Biochemical journal 435, 589-595.

Nemoto, T., Cho, E.-M., Okada, A., Okada, K., Otomo, K., Kanno, Y., Toyomasu, T., Mitsuhashi, W., Sassa, T., Minami, E., Shibuya, N., Nishiyama, M., Nojiri, H., and Yamane, H. (2004). Stemar-13-ene synthase, a diterpene cyclase involved in the biosynthesis of the phytoalexin oryzalexin S in rice. FEBS Lett. 571, 182-186.

Okada, A., Shimizu, T., Okada, K., Kuzuyama, T., Koga, J., Shibuya, N., Nojiri, H., and Yamane, H. (2007). Elicitor induced activation of the methylerythritol phosphate pathway towards phytoalexin biosynthesis in rice. Plant Mol Biol 65, 177-187.

Papadopoulou, K., Melton, R.E., Leggett, M., Daniels, M.J., and Osbourn, A.E. (1999). Compromised disease resistance in saponin-deficient plants. Proceedings of the National Academy of Sciences of the United States of America 96, 12923- 12928.

Peters, R.J. (2006). Uncovering the complex metabolic network underlying diterpenoid phytoalexin biosynthesis in rice and other cereal crop plants. Phytochemistry 67, 2307-2317.

Peters, R.J. (2010). Two rings in them all: The labdane-related diterpenoids. Nat. Prod. Rep. 27, 1521-1530.

Prisic, S., Xu, M., Wilderman, P.R., and Peters, R.J. (2004). Rice contains two disparate ent-copalyl diphosphate synthases with distinct metabolic functions. Plant Physiol. 136, 4228-4236.

Riemann, M., Haga, K., Shimizu, T., Okada, K., Ando, S., Mochizuki, S., Nishizawa, Y., Yamanouchi, U., Nick, P., Yano, M., Minami, E., Takano, M., Yamane, H., and Iino, M. (2013). Identification of rice Allene Oxide Cyclase mutants and the function of jasmonate for defence against Magnaporthe oryzae. Plant J 74, 226-238.

Sakamoto, T., Miura, K., Itoh, H., Tatsumi, T., Ueguchi-Tanaka, M., Ishiyama, K., Kobayashi, M., Agrawal, G.K., Takeda, S., Abe, K., Miyao, A., Hirochika, H., Kitano, H., Ashikari, M., and Matsuoka, M. (2004). An overview of gibberellin metabolism enzyme genes and their related mutants in rice. Plant Physiol. 134, 1642-1653.

Schmelz, E.A., Huffaker, A., Sims, J.W., Christensen, S.A., Lu, X., Okada, K., and Peters, R.J. (2014). Biosynthesis, elicitation and roles of monocot terpenoid phytoalexins. Plant J 79, 659-678.

Sesma, A., and Osbourn, A.E. (2004). The rice leaf blast pathogen undergoes developmental processes typical of root-infecting fungi. Nature 431, 582-586.

Shimizu, T., Miyamoto, K., Miyamoto, K., Minami, E., Nishizawa, Y., Iino, M., Nojiri, H., Yamane, H., and Okada, K. (2013). OsJAR1 contributes mainly to biosynthesis of the stress-induced jasmonoyl-isoleucine involved in defense responses in rice. Biosci Biotechnol Biochem 77, 1556-1564.

Tamogami, S., Mitani, M., Kodama, O., and Akatsuka, T. (1993). Oryzalexin S structure: A new stemarane-type rice plant phytoalexin and its biogenesis. Tetrahedron 49, 2025-2032.

Tezuka, D., Ito, A., Mitsuhashi, W., Toyomasu, T., and Imai, R. (2015). The rice ent- KAURENE SYNTHASE LIKE 2 encodes a functional ent-beyerene synthase. Biochem Biophys Res Commun 460, 766-771.

Thomma, B.P., Nelissen, I., Eggermont, K., and Broekaert, W.F. (1999). Deficiency in phytoalexin production causes enhanced susceptibility of Arabidopsis thaliana to the fungus Alternaria brassicicola. The Plant journal : for cell and molecular biology 19, 163-171.

Toyomasu, T. (2008). Recent Advances Regarding Diterpene Cyclase Genes in Higher Plants and Fungi. Biosci. Biotechnol. Biochem. 72, 1168-1175.

Toyomasu, T., Usui, M., Sugawara, C., Kanno, Y., Sakai, A., Takahashi, H., Nakazono, M., Kuroda, M., Miyamoto, K., Morimoto, Y., Mitsuhashi, W., Okada, K., Yamaguchi, S., and Yamane, H. (2015). Transcripts of two ent- copalyl diphosphate synthase genes differentially localize in rice plants according to their distinct biological roles. Journal of experimental botany 66, 369-376.

Toyomasu, T., Usui, M., Sugawara, C., Otomo, K., Hirose, Y., Miyao, A., Hirochika, H., Okada, K., Shimizu, T., Koga, J., Hasegawa, M., Chuba, M., Kawana, Y., Kuroda, M., Minami, E., Mitsuhashi, W., and Yamane, H. (2014). Reverse- genetic approach to verify physiological roles of rice phytoalexins: characterization of a knockdown mutant of OsCPS4 phytoalexin biosynthetic gene in rice. Physiologia plantarum 150, 55-62.

VanEtten, H.D., Mansfield, J.W., Bailey, J.A., and Farmer, E.E. (1994). Two Classes of Plant Antibiotics: Phytoalexins versus "Phytoanticipins". The Plant cell 6, 1191-1192.

Xu, M., Wilderman, P.R., Morrone, D., Xu, J., Roy, A., Margis-Pinheiro, M., Upadhyaya, N., Coates, R.M., and Peters, R.J. (2007). Functional characterization of the rice kaurene synthase-like gene family. Phytochemistry 68, 312-326.

Yamane, H. (2013). Biosynthesis of phytoalexins and regulatory mechanisms of it in rice. Bioscience, biotechnology, and biochemistry 77, 1141-1148.

Yang, D.L., Yang, Y., and He, Z. (2013). Roles of plant hormones and their interplay in rice immunity. Mol Plant 6, 675-685.

Zi, J., Mafu, S., and Peters, R.J. (2014). To Gibberellins and Beyond! Surveying the Evolution of (Di)Terpenoid Metabolism. Annu Rev Plant Biol 65, 259-286.

Figures

OPP OsCPS4 OsCPS1 OPP OPP GGPP OsCPS2

Momilactones A & B ??? ??? ??? Oryzalides A-C Phytocassanes A-F GA Oryzalexin S Oryzalexins A-F

Figure 1. Rice labdane-related diterpenoids biosynthetic network. CPS, copalyl diphosphate synthase. KSL, kaurene synthase like. Thicker arrow indicates enzymatic reactions specifically involved in GA metabolism; dashed arrows indicate multiple enzymatic reactions.

Figure 2. Transcriptional and chemical characterization of Os-cps2 mutant A) Transcriptional identification of Os-cps2. After Me-JA treatment, transcriptional expression of OsCPS2 in the its wild-type and Os-cps2. Actin expression was used as a control. B) Chemical characterization of Os-cps2 mutant with Me-JA treatment. Gas chromatograph verification of decrease in Os-CPS2 dependent diterpenes in rice leaves. Numbers indicate the corresponding diterpenes: 1. Internal Standard (cembrene) 2. syn- pimara-7,15-diene 3. casbene 4. ent-pimara-8(14), 15-diene 5. ent-sandaraco-pimaradiene 6. ent-isokaurene 7. syn-stemar-13-ene 8. ent-kaurene 9. syn-stemod-12-ene 10. putative syn-exo-stemar-13-ene 11. ent-cassa-12, 15-diene 12. syn-stemod-13(17)-ene

Figure 3. Os-cps2 has increased susceptibility to M. oryzae fungal blast disease (* percentage of diseased area) (***, P < 0.001). Left panel: phenotypes of susceptibility of Os-cps2 to M. oryzae infection. Right panel: percentage of diseased area of Os-cps2 compared with WT.

Figure 4. Os-cps2 has increased susceptibility to Xoo bacterial blight disease. Left panel: phenotypes of susceptibility of cps2 to Xoo infection. Right panel: percentage of diseased area of Os-cps2 compared with WT. (**, P < 0.01) (n=15).

Figure 5. Chemical analysis and phenotype of Os-CPS2OE with its wild-type. A) is with Me-JA treatment and B) is without Me-JA treatment. Gas chromatograph verification of increase in OsCPS2 dependent diterpenes in Me-JA induced Os-CPS2OE and un-induced Os-CPS2OE rice leaves respectively. 1. Internal standard (cembrene) 2. syn-pimara-7,15-diene 3. casbene 4. ent-pimara-8(14), 15-diene 5. ent-sandaracopimaradiene 6. ent-isokaurene 7. syn-stemar-13-ene 8. ent-kaurene 9. syn-stemod-12-ene 10. putative syn-exo-stemar-13-ene 11. ent-cassa-12, 15-diene 12. syn-stemod-13(17)- ene. C) Effect of Os-CPS2OE on Xoo blight disease resistance. Comparison of Lesion length of wild-type and Os-CPS2OE infected with Xoo. P-values: ** < 0.01, n=15.

Figure 6. Chemical analysis and phenotype of Os-cps4 with its wild-type. A) Gas chromatograph verification of increase in Os-CPS2 dependent diterpenes in induced cps4ko rice leaves 1. Internal Standard (cembrene) 2. syn-pimara-7,15-diene 3. casbene 4. ent-pimara-8(14), 15-diene 5. ent-sandaraco-pimaradiene 6. ent-isokaurene 7. syn-stemar-13-ene 8. ent-kaurene 9. syn-stemod-12-ene 10. putative syn-exo-stemar-13- ene 11. ent-cassa-12, 15-diene 12. syn-stemod-13(17)-ene B) Effect of Os-cps4 on Xoo blight disease resistance. Comparison of Lesion length of wild-type and Os-cps4 infected with Xoo. P-values: ** < 0.01, n=15.

Figure 7A. Production of diterpenes in wild-type and Os-cps4 mutant after induction by methyl jasmonate. N. D. non-detectable, Internal Standard 53µg/g FW leaf * p<0.05, ** p<0.01, *** p<0.001

Figure 7B. Production of diterpenes in wild-type and Os-cps2 mutant after induction by methyl jasmonate. N. D. non-detectable, Internal Standard 53µg/g FW leaf * p<0.05, ** p<0.01, *** p<0.001

Figure 7C. Production of diterpenes in wild-type and Os-CPS2OE after induction by methyl jasmonate. N. D. non-detectable, Internal Standard 53µg/g FW leaf * p<0.05, ** p<0.01, *** p<0.001

Figure 8A. Test Os-cps2 and wild-type with different Magnaporthe species and strains. Numbers indicate the degree of infection area. % indicates the proportion to the total disease degree.

Figure 8B. Effect of Os-cps4 on Magneporthe poae non-host disease susceptibility. M. o. indicates Mageneporthe oryzae. M. p. indicates Magneporthe poae.

Supplemental Results

Figure S1. Genome structure of Os-cps2 mutant, tDNA sequence is inserted the sixth exon of OsCPS2 in Os-cps2. Red arrowheads indicate the locations of primers used for genomic PCR.

Figure S2. Representative of putative Oryzalide-Related Diterpenoid Biosynthetic Pathway, dashed arrows indicate multiple enzymatic reactions.

Figure S3. Transcriptional expression of Os-CPS2 in its wild-type and over-expression (OE) lines. Actin expression was used as a control. KitWT indicates the Kitaake cultivar wild-type rice.

Figure S4. Transcriptional expression of rice diterpene synthases (OsCPSs and OsKSLs) at 0, 24 and 48 hours after Xoo infection.

Figure S5. Detection of syn-labdatriene in CPS4WT in GC-MS-selected ion (* indicates the peak corresponding to syn-labdatriene in CPS4WT. Selected Ion Count indicate m/z = 272.

CHAPTER 5: INVESTIGATING THE BIOLOGICAL FUNCTION OF

PHYTOCASSANES IN RICE DEFENSE

Abstract

Among the various responses to microbial infection, rice has developed an arsenal of antibiotic secondary metabolites, which mostly are diterpenoids. While these compounds have been identified on the basis of their antibiotic activity in vitro, this leaves open the question of their relevance to defense in planta. Our previous results show that Os-CPS2 (copalyl diphosphate synthase) pathway is relevant to rice defense against rice fungal blast and bacterial blight disease. However there are several specific pathway downstream Os-CPS2 pathway, it is not clear which specific pathway is the most relevant pathway to rice defense. Phytocassanes, as one of the potent phytoalexins in rice, were isolated from Magnaporthe oryzae (M. oryzae) infected lesions and exhibited antibiotic activity. Here, with the knock-out and over-expression constructs of the key gene Os-KSL7 in phytocassane biosynthetic pathway, results indicate that the biological function of phytocassanes is relevant to defense against both fungal pathogen

M. oryzae and bacterial blight pathogen Xanthomonas oryzae pv. oryzae (Xoo). Our results indicate that previously identified phytoalexins might have the broader biological functions in their ecological contexts.

Introduction

Rice is susceptible to microbial diseases. As one of the responses to these biotic stresses, rice produces phytoalexins as phytochemical arsenals to defend against pathogenic attack (Schmelz et al., 2014). The majority of these phytoalexins are

100 diterpenoids, and their antibiotic activities were identified on their activities against rice blast fungal pathogen M. oryzae. Phytocassanes, as one group of potent rice phytoalexins, were also isolated from rice leaves infected by M. oryzae (Koga, 1995). They exhibited antibiotic activity against M. oryzae spore germination. Different from other rice phytoalexins, besides M. oryzae, phytocassanes were also isolated from rice stems infected by pathogenic fungi Rhizoctonia solani (R. solani), and exhibited the biological activity inhibiting the hyphal growth of R. solani (Koga, 1995). Recently it is shown that phytocassanes can be strongly induced by bacterial pathogen Xoo, and accumulate specifically at the site of Xoo infection in rice leaves (Klein et al., 2015). Interestingly, besides rice leaves, phytocassane A-F are constitutively produced from rice roots as well

(Toyomasu et al., 2008).

Biosynthesis of phytocassanes is derived from the common diterpenoid precursor geranylgeranyl diphosphate (GGPP), catalyzed by ent-copalyl diphosphate (ent-CPP) synthases (Os-CPS2) (Prisic et al., 2004). Followed by this, kaurene synthase-like gene

Os-KSL7 (kaurene synthase-like) catalyzes the further cyclization to diterpene cassane skeleton, which is the precursor of phytocassane A-F (Fig. 1). The downstream oxidative steps are elaborated by cytochrome P450s for decorating phytocassanes A-F biosynthesis.

Momilactone was the first identified rice phytoalexin against the fungal blast pathogen M. oryzae. However genetic evidence show that the momilactones do not play a critical role in rice plant defense, and momilactones acts as allelochemicals instead (Xu et al., 2012). Moreover, some diterpene olefins isolated from maize seedling were inducible to fungal infection but didn’t display anti-fungal activity (Mellon and West 1979). That controversy indicates the needs for genetic evidence to support relevance of phytoalexins.

101

With the reverse genetic evidence, Os-ksl7 knock-down plants are more susceptible to the fungal blast pathogen M. oryzae, as well as the bacterial leaf blight pathogen Xoo, indicating the relevance of phytocassanes to plant defense.

Results

Characterization of Os-ksl7 knockdown mutant

From ent-CPP, production of ent-cassadiene diterpenoid biosynthesis depends on

Os-KSL7. Based on the availability of Os-KSL7 insertion mutant, we ordered a mutant that T-DNA was inserted in the last intron. After two generations of screening, we screened out the homozygous Os-ksl7 line. Os-KSL7 transcripts decrease significantly after treatment with methyl-jasmonate (Me-JA) (Fig. 2A). This shows Os-ksl7 is actually knockdown mutant. Furthermore, from diterpene profiling we detected the production ent-cassadiene as the indicator for final bioactive phytocassanes. There was a significant reduction in the levels of ent-cassadiene in Os-ksl7 plants relative to its parental/WT line

(ssp. japonica cv. Nipponbare) (Fig. 2B).

Susceptibility of Os-ksl7 to M. oryzae and X. oryzae pv. oryzae

Given that some phytocassanes have been suggested to serve as phytoalexins against M. oryzae, the susceptibility of Os-ksl7 rice relative to its parental/WT line against M. oryzae was investigated. Infecting leaves with rice specific isolate of M. oryzae (O-254), this pathogen causes disease symptoms on both Os-ksl7 and its parental/WT rice (Fig 3). It turns out that infected/necrotic areas of leaves in Os-ksl7 plants are much more severe relative to that for its parental/WT line, indicating the

102 susceptibility of Os-ksl7 to M. oryzae. Thus, consistent with the previous literature, ent- cassadiene derived phytocassanes seem to serve an effective role in rice plant defense against M. oryzae.

It has been shown that phytocassanes were inducible at Xoo infection site (Klein et al., 2015). In order to test whether phytocassanes exhibited anti-Xoo activity, we infected Os-ksl7 and its parental wild-type with Xoo infection via a leaf clip method. Os- ksl7 rice exhibited significantly longer lesion lengths than its parental/WT line (p < 0.01 in an unpaired t-test), indicating that this mutant is more susceptible to Xoo (Fig. 4).

Thus, phytocassanes seem to serve an effective role in rice plant defense against this bacterial pathogen Xoo as well.

The resistance of Os-KSL7 over-expressing to X. oryzae pv. oryzae

In order to provide further evidence for the relevance of phytocassanes to rice defense against Xoo, Os-KSL7 over-expressing (OE) were generated. Enhanced expression of Os-KSL7 was verified by RT-PCR, were then selected for further characterization. There was a clear increase in the levels of several ent-cassadiene following Me-JA treatments (Fig. 5A).

Os-KSL7OE plants were infected with Xoo, along with parental/WT plants (ssp. japonica cv. Kitaake). Os-KSL7OE plants exhibited significantly shorter lesion lengths relative to its parental/WT line (p < 0.01 in an unpaired student’s t-test), indicating increased resistance to bacterial leaf blight (Fig. 5B). This result further supports the role for phytocassanes as anti-bacterial agents in rice plant defense, at least against Xoo, consistent with the inducible activity of the phytocassanes at Xoo infection site. These

103 results further have broadened the bioactivity of phytocassanes, which open up the possibility that the relevance of phytoalexins to defense against more microbial pathogens.

Discussion

Previously, the identification of phytoalexins was based on their in vitro antibiotic activity. That raises the questions about their relevance to plant defense. Although phytocassanes and momilactones accumulate in leaves upon infection by M. oryzae

(Hasegawa et al., 2010), there is no genetic evidence to support its biological function with relevance of defense in planta. Biological role of phytocassanes is defined in this study.

Oryzalide A and B, oryzalic acid A and B and oryzadione I to III, oryzalide- related diterpenoids, were isolated and identified on the basis of their activity against Xoo activity (Toyomasu, 2008). However, there is no report about the relevance of phytocassanes to anti-Xoo activity. Here we show the direct relevance of phytocassanes to defense against Xoo. Phytocassanes are constitutively produced from rice roots as well

(Toyomasu et al., 2008), which is consistent with the report that M. oryzae can infiltrate through rice roots (Sesma and Osbourn, 2004). It is possible that the production of phytocassanes from rice root also exert an inhibitory effect on pathogenic fungi around the rice roots in the paddy soil natural conditions. Considering phytocassanes were isolated from rice stems infected by R. solani, it is possible to speculate that phytocassanes may function as phytoanticipans, so they are associated with rice defense

104 throughout all vegetative tissues in rice. Overall, the broader biological functions of phytocassanes will be investigated in the paddy soil condition in the future.

Materials And Methods

Chemicals

Unless otherwise noted, all molecular biology reagents were purchased from

Invitrogen (Carlsbad, CA, USA) and all other chemicals from Fisher Scientific

(Waltham, MA, USA).

Plant Materials

Rice plants (Orzya sativa L. ssp. japonica cv. Nipponbare and Kitaake) were cultivated to the 6th leaf stage in standard growth chambers cycling between 12 h light at

28 °C and 12 h dark at 24 °C. The Os-ksl7 T-DNA insertion mutant PFG_3A-07700.L

(insert in the last intron of Os-KSL7; TIGR ID, LOC_ Os02g36140) was obtained from the PFG Element Insertion Mutant Database from South Korea. This mutant is in the ssp. japonica cv. Nipponbare background. Homozygous Os-ksl7 plants were selected by RT-

PCR of Os-KSL7 (i.e., the absence of any cDNA) from the T1 generation. Homozygous

Os-ksl7 plants were verified in the T2 generation and propagated to provide seeds for the experiments described here.

Transcriptional Expression Analysis

The characterization of ksl7 mutant was identified by RT-PCR with Me-JA treatment. The detailed procedure was described in Chapter 4. Total RNA was extracted

105 from these leaves 24 h after induction with Me-JA, using the procedure for RT-PCR was described previously (Lu et al., 2015). This cDNA preparation was used as the template for PCR with gene-specific primers KSL7RT F and KSL7RT R. Transcription Factor

TFIIAγ5 (LOC_Os05g01710) was used as internal control for rice genes.

Over-Expression Vector Construction and Rice Transformation

The open reading frame for Os-KSL7 was amplified from cDNA by PCR, and the procedure to construct the over-expression lines was followed by methods section described in Chapter 4.

Extraction and Identification of Diterpenes

Rice plants were cultivated to the 6th leaf stage. For induction, detached leaves (3 g) were induced with 10 mL 0.2% Me-JA (in 0.1% Tween 20), wrapped in aluminum foil and incubated in the dark at room temperature (24 °C) for 72 h. GC-MS analysis of diterpenes is the same method as described in Chapter 4.

Disease Resistance Assay

M. oryzae (O-254) were streaked out and grown on potato dextrose agar plate at

26 °C for around 7 days. After that, the plate was poured about 10 ml sterile water, and scraped conidia by spoon and discarded the water. The fungal plate was incubated under the blue fluorescence light to induce conidia formation for three days. When the fungus was ready for infection, 20 ml of sterilized water was poured into the plate and collected fungal conidia with spoon. The fungal conidia were filtered through cheesecloth into a

106

50-ml conical plastic tube, and determined the concentration of conidia in the filtrate by a microscope. 0.05% Tween 20 was added in the conidia. After that, rice plants were sprayed with conidia until the leaf was completely wet. After 4-5 days of infection, the number of blast lesion in susceptible rice was counted. Xoo infections were carried out using the leaf-clipping method (Kauffman et al., 1973), and the detailed procedure was described in Chapter 4.

Acknowledgments

The work was supported by the grant from US Department of Agriculture

(USDA-AFRI-NIFA 2014-67013-21720) to R.J.P. The rice expression vector pBY02 was provided by Dr. Bing Yang’s lab at Iowa State University.

References

Kauffman, H.E., Reddy, A.P.K., Hsieh, S.P.Y., and Merca, S.D. (1973). Improved Technique for Evaluating Resistance of Rice Varieties to Xanthomonas-Oryzae. Plant Dis Rep 57, 537-541.

Koga, J., Shimura, M., Oshima, K., Ogawa, N., Yamauchi, T. and Ogasawara, N. . (1995). Phytocassanes A, B, C and D, Novel Diterpene Phytoalexins from Rice, Or-pa sativa L. tetrahedron 51, 7907-7918.

Lu, X., Hershey, D.M., Wang, L., Bogdanove, A.J., and Peters, R.J. (2015). An ent- kaurene-derived diterpenoid virulence factor from Xanthomonas oryzae pv. oryzicola. The New phytologist 206, 295-302.

107

Prisic, S., Xu, M., Wilderman, P.R., and Peters, R.J. (2004). Rice contains two disparate ent-copalyl diphosphate synthases with distinct metabolic functions. Plant physiology 136, 4228-4236.

Schmelz, E.A., Huffaker, A., Sims, J.W., Christensen, S.A., Lu, X., Okada, K., and Peters, R.J. (2014). Biosynthesis, elicitation and roles of monocot terpenoid phytoalexins. Plant J 79, 659-678.

Sesma, A., and Osbourn, A.E. (2004). The rice leaf blast pathogen undergoes developmental processes typical of root-infecting fungi. Nature 431, 582-586.

Toyomasu, T. (2008). Recent Advances Regarding Diterpene Cyclase Genes in Higher Plants and Fungi. Biosci. Biotechnol. Biochem. 72, 1168-1175.

Toyomasu, T., Kagahara, T., Okada, K., Koga, J., Hasegawa, M., Mitsuhashi, W., Sassa, T., and Yamane, H. (2008). Diterpene phytoalexins are biosynthesized in and exuded from the roots of rice seedlings. Bioscience, biotechnology, and biochemistry 72, 562-567.

Figures

Figure 1. Biosynthetic pathway to phytocassane A-F from diterpenoid precursor GGPP

108

Figure 2. Transcriptional and chemical analysis of Os-ksl7 knock-down mutant. A) Transcriptional identification of Os-ksl7. After Me-JA treatment, transcriptional expression of Os-KSL7 in its wild-type and Os-ksl7. Actin expression was used as a control. B) Gas chromatograph analysis of ksl7 mutant (* indicates the peak corresponding to ent-cassadiene)

109

Figure 3. Susceptibility of Os-ksl7knock-down on M. oryzae fungal blast disease

Figure 4. Susceptibility of Os-ksl7 on Xoo bacterial blight disease. The red arrow indicates the ending point of lesions. Left panel: phenotypes of susceptibility of Os-ksl7 to Xoo infection. Right panel: lesion length of Os-ksl7 compared with its WT. (***, P < 0.01, n=15)

110

Figure 5. Phenotype and chemical characterization of Os-KSL7OE with its wild-type. A) Lesion lengths of Os-KSL7OE against Xoo bacterial blight disease. B) Gas chromatograph verification of increase of ent-cassadiene in Os-KSL7OE plants

111

CHAPTER 6: EXPLORATION OF ORYZALIDE-RELATED DITERPENOIDS

IN RICE DEFENSE WITH GENOME EDITING TECHNOLOGY

Abstract

Among the various effective means of defenses against microbial infection, rice has deployed an arsenal of phytoalexins, which mostly are diterpenoids. While these have been identified on the basis of their antibiotic activity in vitro, this leaves open the question of their relevance to defense in planta. Our previous results show that the Os-

CPS2 (copalyl diphosphate synthase) pathway is relevant to rice defense against rice blast fungal disease and blight bacterial disease. However, it is not clear which specific pathway downstream of Os-CPS2 is relevant to rice defense. Here with the successful application of CRISPR/Cas9 technology targeted to Os-KSL5 and Os-KSL6 (Δcrksl5 and

Δcrksl6), we show that the products of Os-KSL5 and Os-KSL6 pathways are relevant to rice defense. Moreover through the generation of Os-KSLs over-expression plants, our preliminary results indicate the additive effect might explain why rice produces an array of diterpenoid phytoalexins.

Introduction

Rice is exposed to many biotic stresses including infection by fungi, bacteria, and viruses. As one of the strategy against microbial attack, rice produces phytoalexins as response to pathogenic infection (Schmelz et al., 2014). Most of these phytoalexins are diterpenoids, and their antibiotic activities were originally characterized against rice blast fungal pathogen Magneporthe oryzae (M. oryzae).

112

Different from most of diterpenoid phytoalexins, oryzalide related diterpenoids were extracted and isolated especially from rice resistant cultivar (Watanabe et al., 1996), which caused the incompatible reaction with Xanthomonas oryzae pv. oryzae (Xoo).

These oryzalide related diterpenoids include oryzalide A and B, oryzalic acid A and B and oryzadione I to III (Toyomasu, 2008). Although oryzalide-related diterpenoids have been isolated mostly from resistant rice cultivars. It is not known in susceptible rice cultivars the relevance of oryzalide-related diterpenoids to defense.

Oryzalide related diterpenoids biosynthesis is evolved from gibberellin biosynthesis. Derived from the common diterpenoid precursor GGPP, ent-copalyl diphosphate (ent-CPP) synthase Os-CPS1 and Os-CPS2 catalyze GGPP to ent-CPP

(Prisic et al., 2004). After that, Os-KSL6 converts the cyclization from ent-CPP to diterpene ent-isokaurene, which could be the precursor of oryzalide related diterpenoids.

Interestingly, a putative redundant gene Os-KSL5 is just located next to Os-KSL6. In rice subspecies indica, Os-KSL5 and Os-KSL6 are actually the redundant gene while in rice subspecies japonica, they are different gene. We biochemically characterized that the product of Os-KSL5 is ent-pimaradiene instead of ent-isokaurene, but the metabolic fate and its biological role are still unknown (Xu et al., 2007).

Analysis of a T-DNA insertion mutant of Os-CPS4 has indicated that these knockout Os-cps4 plants are not more susceptible to infection with M. oryzae than the parental/wild-type (WT) line. Additional analysis, including the use of a more specific

Os-ksl4 gene knockout line, indicates that the momilactones act as allelochemicals (Xu et al., 2012).

113

As one of the cutting-edge genome editing technologies, the clustered regulatory interspersed short palindromic repeat (CRISPR)-associated protein (Cas9) system has been successfully applied in various plant species. Compared with TALEN (transcription activator-like nuclease) and zinc finger nuclease technology, CRISPR/Cas9 system is much easier to construct the target sequence because only 20 base pair guide RNA can simply recognize the target DNA sequences. The only requirement for the target sequence is that followed by NGG PAM (protospacer adjacent motif). With the Cas9 nuclease, it will lead to double-strand break several nucleotides upstream the PAM.

Interestingly, CRISPR/Cas9 technology can be applied to cause the large chromosome deletion in rice diterpenoid phytoalexin study (Zhou et al., 2014).

Here we take advantage of CRISPR/Cas9 technology to investigate the relevance of Os-KSL5 and Os-KSL6 pathways to plant defense. Also, by the generation of the Os-

KSLOE, the results indicate that the additive effect may explain the reason rice produces so many diterpenoid phytoalexins.

Results

Construction and Identification of Δcrksl5 and Δcrksl6 Plants

To elucidate physiological function in oryzalide-related diterpenoids in planta, we target on the first exon of redundant gene Os-KSL5 and Os-KSL6 respectively (Fig. 1).

Without T-DNA mutants available for Os-KSL5 and Os-KSL6, we seek for CRISPR/Cas9 genome editing method. Since Os-KSL5 and Os-KSL6 share 93% of cDNA identity, to prevent the off-target, we design only one single guide with 20bp for each gene. At T0 generation, we genotyped the transgenic plants by sequencing the target fragments (Fig.

114

2). In Δcrksl5 and Δcrksl6, total mutagenesis rate for Δcrksl5 and Δcrksl6 are up to 92% and

85% respectively. Types of mutagenesis include heterozygous, homozygous and diallelic mutation. As for Δcrksl5, the most common type of mutagenesis is one or two base pairs addition or deletion. For Δcrksl6, besides one or two base pair addition or deletion, we detected 14bp deletion homozygous plants. We sequenced the whole Os-KSL5 and

Os-KSL6 genes, in case off-target frequency may happen. There is no trace of any nucleotide changes at other sites. After the self-fertilization T0 generation, in T1 generation sequencing results showed that mutagenesis is inherited stably in a Mendelian fashion. OsCas9 has been segregated it out with 28% due to the loss of T-DNA in T1 generation. The homozygous mutants from T2 plants were selected and used as the following experimental material.

Susceptibility of Δcrksl5 and Δcrksl6 to X. oryzae pv. oryzae

We infect infected Δcrksl5, Δcrksl6 and their parental WT (Orzya sativa ssp. japonica cv. Kitaake) with Xoo via a leaf clip method. Both Δcrksl5 and Δcrksl6 rice exhibited significantly longer lesion length than WT line (p < 0.01 in an unpaired t-test), indicating that both mutants are more susceptible to Xoo infection (Fig. 3). Thus, oryzalide related diterpenoids seem to serve more effective role to defend against Xoo.

This result is consistent with the known anti-bacterial activity of the oryzalide-related diterpenoids, which are potentially derived from ent-isokaurene. Moreover, it indicates the biological function of oryzalide related diterpenoids is not only in resistant cultivar, but in susceptible cultivar as well. Although the final product of Os-KSL5 is not clear, this result indicates that the final products in Os-KSL5 pathway are relevant to defense.

115

Os-KSL over-expressing (OE) plants indicating the additive effect

In order to investigate the relevance of further downstream of ent-CPP derived diterpenoids to rice defense, we planned to infect all ent-CPP derived Os-ksl mutant and compare their lesion lengths infected by Xoo. However, Os-cps2, Os-ksl7, and Os-ksl10

T-DNA mutants are in the different genetic background. Thus, we generated Os-KSL5,

Os-KSL6, Os-KSL7 and Os-KSL10 over-express plants all in the same genetic background (cv. Kitakee) to investigate their biological activity along with Os-CPS2OE plants. Os-KSL5OE, Os-KSL6OE, Os-KSL7OE and Os-KSL10OE plants were identified by presence of cDNA in the genomic DNA and checking transcriptional expression (data not shown).

Although Os-KSL6OE and Os-KSL7OE showed shorter lesion lengths than parental WT line (p < 0.05 in an unpaired student’s t-test), Os-CPS2OE plants exhibited the shortest lesion length relative to its parental/WT line (p < 0.01 in an unpaired t-test).

That indicates diterpenoids produced from Os-CPS2OE exert the most efficient inhibition against Xoo (Fig. 4). The major product from Os-CPS2OE is ent-kaurene, but the level of ent-cassadiene and ent-isokaurene also get increased slightly in Os-CPS2OE plants.

Indeed, given that ent-kaurene and ent-isokaurene are precursors to the oryzalide-related anti-bacterial phytoanticipans. These results might suggest it is the additive effect of all

Os-CPS2 derived diterpenoids including final bioactive products of ent-isokaurene, ent- cassadiene and ent-sandaracopimaradiene that all lead to the most significant inhibitory effects against Xoo.

116

Discussion

This study is the successful application of CRISPR/Cas9 technology in plant secondary metabolism. Unlike in the field of plant development and reproduction, there are very few publications on the application of genome editing in plant secondary metabolism. This study is a timely successful application to narrow this gap. Although the natural products have been isolated, their biological functions in planta are still in the air. With this technology, we can explore the some unexplored functions of secondary metabolites in their native ecological contexts.

Results from Δcrksl6 show that oryzalide related LRDs are relevant to defense against Xoo. That is consistent with the previous report that oryzalide related LRDs were isolated and identified in resistant cultivar by their activities against Xoo. We broadened the biological function of oryzalide-related in susceptible cultivar genetic background. As for Δcrksl5, although it is still clear the products of Os-KSL5, the results indicate the products of Os-KSL5 are also relevant to defense against Xoo. Now with Δcrksl5, we can dig into the intermediate and final product of Os-KSL5 by chemical analysis with wild- type. The additive effect of phytoalexin production was reported in pea phytoalexin

(Kazuhiro Toyoda, 1995). To further confirm this model, crossing the individual

CRISPR/Cas9 mutant plants in the same genetic background could provide more supporting evidence.

117

Materials and Methods

Chemicals

Unless otherwise noted, all molecular biology reagents were purchased from

Invitrogen (Carlsbad, CA, USA) and all other chemicals from Fisher Scientific

(Waltham, MA, USA).

Plant materials

Rice plants (Orzya sativa L. ssp. japonica cv. Kitaake) were cultivated to the 6th leaf stage in standard growth chambers cycling between 12 h light at 28 °C and 12 h dark at 24 °C.

CRISPR/Cas9 construction and genotyping of transgenic plants

Commercial pENTR4 vector was modified by Dr. Bing Yang’s lab. For sgRNA constructs, the cutting sites of BtgZI or BsaI type II restriction enzymes were ligated downstream of the U6 promoter. 4-bp overhangs were added to seed sequences for recognition of BtgZI or BsaI, after annealing forward and reverse primers, the seed sequences were phosphorylated to clone into the pENTR4 vector. U6p-F1 was used to sequence inserted seed fragments. After that, the seed sequence cassettes in pENTR- sgRNA vectors were mobilized to pBY02-CCDB vector using the Gateway LR clonase

(Life Technologies). The pBY02 vectors with seed sequences were confirmed by enzyme digestion by BamHI or HindIII. The resulting vector construct was transformed into

Agrobacterium tumefaciens EHA105 by electroporation. Immature embryo-derived callus cells of rice (cv. Kitaake) were transformed using a previously described (Hiei et

118 al., 1994). T0 transgenic plants were detected with PCR for each plant, and PCR products from the targeted regions in each line were subjected to Sanger sequencing. After self- fertilization in T1 transgenic rice plants, we sequenced the targeted regions again in T1.

Over-expression vector construction and rice transformation

The open reading frame for Os-KSL5, Os-KSL6, Os-KSL7 and Os-KSL10 were amplified from cDNA by PCR and the procedure to construct the over-expression lines was followed by methods section described in Chapter 4.

Disease resistance assay

Xoo infections were carried using the leaf-clipping method (Kauffman et al.,

1973), and was described in Chapter 4.

Acknowledgments

The authors thank Dr. Huanbin Zhou and Honghao Bi for helpful discussion on

CRISPR-Cas9 construction. The modified cloning vector pENTR4 and expression vector pBY02-CCDB vectors were kindly provided by Dr. Bing Yang’s group. The work was supported by the grant from US Department of Agriculture (USDA-AFRI-NIFA 2014-

67013-21720) to R.J.P.

119

References

Kauffman, H.E., Reddy, A.P.K., Hsieh, S.P.Y., and Merca, S.D. (1973). Improved Technique for Evaluating Resistance of Rice Varieties to Xanthomonas-Oryzae. Plant Dis Rep 57, 537-541.

Kazuhiro Toyoda, K.M., Yuki Ichinose, Tetsuji Yamada and Tomonori Shiraishi. (1995). Plant Lectins Induce the Production of a Phytoalexin in Pisum sativum. Plant and Cell Physiology 36, 799.

Prisic, S., Xu, M., Wilderman, P.R., and Peters, R.J. (2004). Rice contains two disparate ent-copalyl diphosphate synthases with distinct metabolic functions. Plant physiology 136, 4228-4236.

Schmelz, E.A., Huffaker, A., Sims, J.W., Christensen, S.A., Lu, X., Okada, K., and Peters, R.J. (2014). Biosynthesis, elicitation and roles of monocot terpenoid phytoalexins. Plant J 79, 659-678.

Toyomasu, T. (2008). Recent Advances Regarding Diterpene Cyclase Genes in Higher Plants and Fungi. Biosci. Biotechnol. Biochem. 72, 1168-1175.

Xu, M., Wilderman, P.R., and Peters, R.J. (2007). Following evolution's lead to a single residue switch for diterpene synthase product outcome. Proceedings of the National Academy of Sciences of the United States of America 104, 7397-7401.

Zhou, H., Liu, B., Weeks, D.P., Spalding, M.H., and Yang, B. (2014). Large chromosomal deletions and heritable small genetic changes induced by CRISPR/Cas9 in rice. Nucleic acids research 42, 10903-10914.

120

Figures

Figure 1. Sequences of DNA fragment induced by sgRNAs in the first exon of Os-KSL5 and Os-KSL6 respectively. The PAM sequence is colored in red.

121

Figure 2. Sequencing confirmation of the site-specific mutations (diallelic) in the T0 transgenic lines. The PAM sequence is colored in orange. Dashes denote nucleotide deletions and blue-colored letters denote nucleotide insertions. The arrow indicates the transgenic lines used as the experimental material.

122

Figure 3. Effects of Δcrksl5 and Δcrksl6 on Xoo bacterial blight disease. The red arrow indicates the ending point of lesions. Left panel: phenotypes of susceptibility of Δcrksl5 and Δcrksl6 to Xoo infection. Right panel: lesion length of Δcrksl5 and Δcrksl6 compared with WT. (*, P < 0.05; **, P < 0.01)

123

Figure 4. Effects of Os-KSL5OE, Os-KSL6OE, Os-KSL7OE and Os-KSL10OE against Xoo bacterial blight disease. Left panel: phenotypes of Os-KSLOE to Xoo infection compared with Kitaake WT. Right panel: lesion length of overexpression KSLs to WT. (*, P < 0.05; **, P < 0.01)

1 24

CHAPTER 7: CONCLUSIONS AND FUTURE DIRECTIONS

Conclusions

Diterpenoids provide a rich source of natural products to mediate plant-microbe interaction. In this thesis, I focus on the roles of diterpenoids rice-Xanthomonas oryzae interaction. In Chapter 2 from the side of X. oryzae pv. oryzicola (Xoc), with genome mining in Xoc, we found a gene operon that is highly relevant to GA biosynthesis. We biochemically characterized that this pathogen had the capability to produce the precursor of gibberellin ent-kaurene. By insertion mutagenesis, it turns out that putative gibberellin acts as the virulence factor for Xoc. The molecular mechanism is via GA antagonistic to jasmonate so as to down-regulate rice defense. With more genome mining, we found that

GA biosynthetic operon is not only found in pathogenic bacteria but in plant nodulating bacteria as well. Chapter 3 is a side project about this story. We found the GA operon in the genome of four genera of rhizobia. Similarly, we biochemically characterized the diterpene synthases in rhizobia. Moreover, the preliminary evidence suggests that the production of putative gibberellin might be relevant to the nodulation process.

From the side of rice, it produces phytoalexin phytochemicals as responses to defend against microbial pathogens. While diterpenoid phytoalexins have been identified on the basis of their antibiotic activity in vitro, this leaves open the question of their relevance to defense in planta. With reverse genetics method in Chapter 4, we found that the relevance of diterpenoids to rice plant defense depends on Os-CPS2 pathway.

Moreover, metabolic flux keeps balanced in Os-CPS2 and Os-CPS4 pathway.

Downstream the Os-CPS2 pathway, in Chapter 5 more specific phytocassane pathway is

125 relevant to defense against pathogenic fungi M. oryzae, but pathogenic bacteria Xoo as well. Besides phytocassanes in Chapter 6, with the advent of CRISPR/Cas9 technology, we explored the biological function of more specific pathways downstream ent-CPP. We successfully generate mutants of redundant gene Os-KSL5 and Os-KSL6 with

CRISPR/Cas9. It seems that Os-KSL5 and Os-KSL6 pathway are relevant to defense against Xoo. Moreover, over-expressing Os-KSL5, Os-KSL6, Os-KSL7 and Os-KSL10 downstream Os-CPS2 indicates the rice phytoalexins might reach the additive effect on inhibitory against pathogens in their biological contexts.

Future Directions

As for the putative GA biosynthetic operon in Xoc, we have biochemically characterized that it had the capacity to produce the precursor of gibberellin ent-kaurene.

The next obvious step is to biochemically characterize all P450s in the operon. Motivated by the results that nodule of Bradyrhizobium japonicum has the capacity to produce GA9

(Mendez et al., 2014), our preliminary findings showed the GA biosynthetic operon in the

Xanthomonas translucens pv. translucens, four P450s in the operon have the capacity to catalyze ent-kaurene to GA4. That supports our hypothesis in Chapter 2 that the final bioactive product of ent-kaurene is GA4. After solving this major question, based on the previous model that GA and jasmonic acid competing for bind with DELLA protein (Bari and Jones, 2009), the next question we are interested is the molecular mechanism why

GA4 functions as the virulence as for Xoc. We want to know how GA4 and jasmonic acid compete to bind with DELLA protein in rice.

126

In Chapter 4, we detected and quantified the level of intermediate diterpenes as the indicators for the final bioactive diterpenoids in Os-cps2, Os-cps4 and Os-CPS2OE plants. Our next step is to detect and quantify these final bioactive diterpenoids. In addition, our results showed that Os-cps4 plants are relevant to non-host fungal pathogen

Magneporthe poae. This opens the possibility that Os-cps2 and Os-cps4 might be susceptible to more host and non-host pathogens. If possible, we can even set up the field trail test to mimic the natural ecological conditions.

Besides the biotic stress, crop phytoalexins are also relevant to abiotic stresses.

Rice momilactones are involved in allelopathy to suppress the growth of neighboring lettuce and barnyard grass (Xu et al., 2012). In maize, the production of phytoalexin kauralexins and zealexins are relevant to drought and salt stresses (Vaughan et al., 2015).

Thus, it is reasonable to speculate that rice diterpenoid phytoalexins might also be relevant to abiotic stress. We will set up the drought and salt stress experiment with Os- cps2 and Os-cps4 plants to test their responses to different stresses.

In Chapter 5, the last unfinished piece for phytocassanes study is to conduct the chemical analysis for ksl7 knock-down so that we can get to the conclusion whether metabolic fluxes towards other specific pathways downstream ent-CPP. In order to provide more evidence of phytocassane defense against Xoo, we’ll mimic the physiological concentration of phytocassanes, and conduct the in vitro assay to test the inhibitory effect on phytocassanes against Xoo. In addition, phytocassane A-F are constitutively produced from rice roots, but they do not seem to be involved in allelopathy (Toyomasu et al., 2008). Thus, the production of phytocassanes from roots

127 might be relevant to root pathogens or abiotic stresses. It is worth testing the physiological functions of phytocassanes in roots.

In Chapter 6 with the application of CRISPR/Cas9 technology, we successfully generate Δcrksl5 and Δcrksl6 mutants. To finish the complete scientific story, we will conduct the chemical analysis for Δcrksl5 and Δcrksl6 so as to get the clue of the metabolic fluxes. Moreover, with Δcrksl5 mutant, we will investigate metabolic fate of Os-KSL5 pathway, such as the intermediate and final product of ent-pimaradiene in Os-KSL5 pathway by comparing the metabolic profiling with the parental wild type.

Furthermore, with the successful application of CRISPR/Cas9 technology, we are interested in targeting on the downstream cytochrome P450s. The most straightforward question to answer is which specific diterpenoid phytoalexin is most relevant to rice defense. For example, there are six diterpenoids in the group of phytocassane A-F, with

CRISPR/Cas9 we can elucidate the most bioactive phytocassane in planta. In addition, the previous in vitro tests show that several P450s (CYP76M5-M8) shared the same enzymatic activity, and single P450 (CYP701A8) react with several substrates (Wang et al., 2012b; Wang et al., 2012a). With CRISPR/ Cas9, we can explore whether they also function redundantly in planta and investigate the order actual the catalytic steps by comparing the metabolic profiling.

Besides targeting on individual gene, CRIPSR/ Cas9 has been applied in large chromosomal deletions (Zhou et al., 2014), which targets on two diterpenoid biosynthsis gene clusters located on chromosome 2 and 4 respectively. The gene cluster on chromosome 4 spans 170 kb while the other one on chromosome 2 is around 245kb (Fig.

1). In the cooperation project with Dr. Bing Yang’s group, now we have acquired the

128 homozygous gene cluster mutants and double gene cluster deletion. The next step we will carry out the chemical analysis to detect diterpenes and diterpenoids metabolic profiling.

In Chapter 6, our preliminary results indicate the production of many rice phytoalexins might give rise to the additive effect. In order to provide more evidence for this model, next step is to generate double gene mutant by crossing the single gene mutant. With these mutants, we will compare lesion lengths and counting colony-forming units after Xoo infection.

Overall, in rice-X. oryzae interactions diterpenoids play indispensible role. For

Xoc, important diterpenoid GA can be utilized as the virulence factor. From the perspective of rice, it can take advantage of diterpenoid phytoalexins to defense against microbial pathogens such as X. oryzae. With traditional reverse genetic and new

CRISPR/Cas9 technology, we established the relevance of specific diterpenoids to rice defense against both microbial pathogens and showed the redirection of metabolic flux as well. Based on these, we provide not only some insight into the flux balance in specialized metabolism but potential agricultural application to control rice disease in the future as well.

References:

Bari, R., and Jones, J.D. (2009). Role of plant hormones in plant defence responses. Plant Mol Biol 69, 473-488.

Mendez, C., Baginsky, C., Hedden, P., Gong, F., Caru, M., and Rojas, M.C. (2014). Gibberellin oxidase activities in Bradyrhizobium japonicum bacteroids. Phytochemistry 98, 101-109.

Toyomasu, T., Kagahara, T., Okada, K., Koga, J., Hasegawa, M., Mitsuhashi, W., Sassa, T., and Yamane, H. (2008). Diterpene phytoalexins are biosynthesized in

129

and exuded from the roots of rice seedlings. Bioscience, biotechnology, and biochemistry 72, 562-567.

Vaughan, M.M., Christensen, S., Schmelz, E.A., Huffaker, A., McAuslane, H.J., Alborn, H.T., Romero, M., Allen, L.H., and Teal, P.E. (2015). Accumulation of terpenoid phytoalexins in maize roots is associated with drought tolerance. Plant Cell Environ 38, 2195-2207.

Wang, Q., Hillwig, M.L., Wu, Y., and Peters, R.J. (2012a). CYP701A8: a rice ent- kaurene oxidase paralog diverted to more specialized diterpenoid metabolism. Plant physiology 158, 1418-1425.

Wang, Q., Hillwig, M.L., Okada, K., Yamazaki, K., Wu, Y., Swaminathan, S., Yamane, H., and Peters, R.J. (2012b). Characterization of CYP76M5-8 indicates metabolic plasticity within a plant biosynthetic gene cluster. The Journal of biological chemistry 287, 6159-6168.

Figure

Figure 1. Scheme of two Labdane related diterpenoid gene clusters in rice.

130

ACKNOWLEDGEMENTS

First and foremost, I would like to express my gratitude to my major professor,

Dr. Reuben Peters, for his invaluable guidance, encouragement and support throughout the course of my research. He is an excellent model to inspire me develop critical thinking skills and overcome the difficulties and obstacle during these years.

I also express my sincere appreciation rest of my POS committee, Dr. Gwyn

Beattie, Dr. Gustavo Macintosh, Dr. Alan Myers, and Dr. Bing Yang for their helpful advice and valuable suggestions.

In addition, I wish to thank our lab manager Meimei Xu for her willingness with experimental procedures. I would like to thank all colleagues and friends in Reuben’s lab graduate students, postdoc and undergraduate students for their sincere help and friendship. I’m fortunate to work in such an enjoyable working environment these years.

And also, I would thank my friends and collaborators in Dr. Bing Yang’s lab and Dr.

Adam Bogdanove’s lab for their scientific discussion and selfless help.

Finally, I would like to show my deepest gratitude to my wife Yunting Pu and my parents for encouragement, understanding and supporting me throughout my dissertation work. I want to especially thank my wife for cheer me up when I am down in my research. Without them, I cannot achieve what I have accomplished in my research.