Transcriptome Profiling of Two Arabidopsis Farnesyl

Diphosphate Synthase Mutants for Understanding

Terpenoids Metabolism

YU, Pui Man

A Thesis Submitted in Partial Fulfillment

of the Requirements for the Degree of

Master of Philosophy

in

Biology

August 2009 /Vl:统系餘書圖\\

(I.”— m ….'TV肩 Thesis/Assessment Committee

Professor L. Jiang (Chairman)

Professor Diane D. J. Guo (Thesis Supervisor)

Professor Y. S. Wong (Committee Member)

Professor Y. M. Zhu (External Examiner)

ii statement

All experimental work reported in this thesis was performed by the author, unless stated otherwise.

二？丨 ffW•一 -TO Pui Man

iii Acknowledgements

First of all, I sincerely thank my supervisor, Prof. Diane Dianjing Guo, for giving me a precious chance to gain multidisciplinary research experience in the field of plant metabolic engineering. Her professional advices and constant concerns have

given me great encouragement during these two years of M.Phil study.

I am glad to receive the valuable suggestions and guidance from my thesis

committee members, Prof. Liwen Jiang and Prof. Yum Shing Wong during my thesis

‘research. It is my honor to have Prof. Yanming Zhu (Northeast Agricultural

University, China) as my external examiner who has provided valuable comments to

the draft of this thesis.

My research works would not be possible without the kindest supports from all

laboratory mates in G94.1 would like to give special thanks to Ms. Yan Qi and Mr.

Wei Wang for the guidance in the field of molecular biology, Dr. Wei Wu and Mr.

Baohui Cheng for advice and assistance in the chemical analyses part of the project. I

would also like to express my appreciation to Mr. Qing Zhang for instructing me in

the microarray data analysis. Thanks were extended to Mr. Yiu Man Chan, Mr.

Yejun Wang and Mr. Peng Tien for the discussions and advices on the research.

Lastly, my thanks would go to my beloved family for their loving considerations

and great support throughout my study.

iv Abstract

Terpenoid is the largest class of natural product essential for both primary and secondary metabolism of the plants. Many terpenoids also serve as important components of natural pharmaceuticals and commercial products, such as the anti-malarial agent artemisinin, the anti-cancer agent paclitaxel (Taxol), eleutherobin, dyes, flavors, and perfumes etc. Elucidating terpenoid biosynthesis pathway and its regulatory mechanism is therefore of utmost importance to metabolic engineering aiming at enhanced terpenoids production.

Two distinctive terpenoids biosynthetic pathways exist in cytosolic and plastidial respectively in higher plants, namely, mevalonate (MVA) and methylerythritol 4-phosphate (MEP) pathways. Although located in separate compartments, it has been suggested that crosstalk exists between the MVA and

MEP routes. Farnesyl diphosphate synthase (Farnesyl pyrophosphate synthase, FPS) is one of the key enzymes situates at the multi-branching point of terpenoids pathway, leading to the production of sesquiterpenes, triterpenes, polyterpenes, and sterols.

Two FPS genes (fpsl and fps2) have been identified and cloned in Arabidopsis.

Although their temporal and spatial expression patterns have been well characterized, the exact functional roles of these two distinctive enzymes are still largely unknown.

The goal of this project is to investigate the roles of these two FPSs in controlling the metabolic flux of terpenoid biosynthesis. Using Affymetrix GeneChip, gas chromatography-mass spectrometry (GC-MS), and high-performance liquid chromatography (HPLC) approaches, I systematically investigated the transcriptome profiles and targeted metabolite profiles in Arabidopsis fps mutants. Compared to wild type, knockdown of the fps gene resulted in only limited effect on the transcriptome, with slightly higher impact on genes participate in the MEP pathway than those in the MVA pathway. The differentially expressed genes were grouped according to Gene Ontology (GO) terms under the category of molecular function.

Except for the un-annotated genes, the top two categories were “enzyme activity” and “transport”. Genes involved in stress and defense response, such as glutaredoxin

(GRX) and thioredoxin (TRX), were altered in both mutants. This suggests that reduction in fps gene expression may trigger cellular stress-like response in plant

system. More genes in the fps2 mutant showed differential expression compared to

the fpsl mutant, indicating the distinctive roles of these two enzymes in terpenoids

biosynthesis.

GC-MS analysis showed no significant changes in triterpene and sterols content.

HPLC analysis revealed that carotenoids and chlorophyll content were dramatically

changed in the mutants. In addition to the fact that perturbation of the terpenoid

cytosolic pathway enzyme FPS caused only limited gene expression change at

transcript level but significant metabolite change suggests that post transcriptional

regulation may play an important role in control the metabolic flux in MVA pathway.

Furthermore, perturbation of cytosolic MVA pathway gene FPS at the downstream

of IPP caused significant changes in plastidial MEP pathway genes and downstream

metabolites levels, indicating crosstalk between the two compartments may exist at

multiple points apart from EPP. 摘要

萜類化合物是最大的一類天然產物，是植物初生代謝和次生代謝中必不可

少的物質。許多萜類化合物也是重要的天然藥物和商品，比如抗瘧藥物青 蒿 素 、

抗癌藥物紫杉醇( Taxol) 和軟珊瑚提取物Eleutherobin、染料 、香料 、香水 等 等 。

因此萜類化合物合成途徑以及它的調控機制的硏究，對增加萜類生產的代謝工

程是極爲重要的。

高等植物中一般存在兩條萜類化合物合成途徑：甲羥戊酸（MVA) 途徑和

2-C-甲基-D-赤藻糖醇-4-磷 酸 (MEP)途 徑 ，分別存在於細胞質和質體中。雖然

定位的細胞器不同，但是據硏究IPP 是萜類化合物的共用前體，這兩條合成途

徑 應 該 在 IPP 位置有交叉互換。然而目前，能夠產生許多潛在藥物的MVA途

徑還沒有被深入地硏究。法呢基焦磷酸合酶(FPS) 是位於多個合成途徑分支點

的關鍵酶，通過干擾該酶，可以影響倍半萜、三萜 、多萜和甾醇的合成。在模

式植物擬南芥中發現了兩種FPS (FPS1和 FPS2) ，而且它們已經被克隆得到了

完整的編碼基因。雖然已經鑑定了這兩個基因的時間與空間表達特徵，但是它

們在萜類化合物合成途徑中的確切功能基本上還是未知的。

本硏究的目的就是要通過兩個FPS基因敲除突變體來改變萜類化合物生物

合成途徑，從而硏究兩個 FPS 酶在萜類化合物的代謝途徑中的角色。通過

Affymetric基 因 芯 片 、GC-MS、HPLC等技術手段，我們系統地硏究了擬南芥兩

種 FPS突變體的轉錄本和針對性的代謝圖的變化。

實驗結果表明FPS的干擾對於轉錄本的影響不明顯，MEP途徑中基因的變

化稍大於參與MVA途徑的基因。差異表達的基因都使用GO 分類的方法根據分

子功能進行了分類，除了未被註釋的基因外，屬於酶活性和轉運這兩個組別的

vii 差異基因最多。還有一些與脅迫和防禦反應有關的基因，比如編碼谷氧還蛋白

(GRX)和硫氧還蛋白(TRX)的 基 因 ，也在兩種FPS突變體中有所變化。這表 明 ，

減少了 FPS的基因表達可能會觸發植物系統中的類細胞脅迫反應。FPS2突變體

中的差異表達基因比較多，同時 FPS1和 FPS2突變體的差異表達基因不同，也

證明兩個突變體的基因調控模式不同，以上結果表明FPS1和 FPS2這兩個酶在

萜類化合物合成中的功能不同。

氣相色譜分析顯示突變體中的三萜和甾醇含量無顯著變化，高效液相色譜

法分析卻表明類胡蘿蔔素和葉綠素含量均顯著改變。在有限的轉錄水平的變化

上 ，代謝產物卻有重大的變化，這也進一步證明了轉錄後調控在MVA 途徑代

謝通量的控制中扮演著重要的角色。另外 ，據報道 FPS酶 存 在 於 MVA途 徑 ，

也就是位於細胞質中，但是 FPS酶活性的降低也會導致細胞質體中MEP途徑的

下游基因表達量減少，下游代謝產物起變化，這就説明除了 IP P 位 點 ，其他位

置的化合物也存在著交叉互換。 Table of Contents

Acknowledgements iv

Abstract v

Table of Contents ix

List of Figures xii

List of Tables xiv

List of Abbreviations xv

Chapter 1. General Introduction 1

Chapter 2. Literature Review 5

2.1 The importance of terpenoids 5

2.2 The difficulties in synthesizing terpenoids 8

2.3 Structure and classification of terpenoids 9

2.4 MVA and MEP pathways of terpenoid biosynthesis in higher plants 12

2.4.1 The MVA pathway 16

2.4.2 The MEP pathway 18

2.5 The crosstalk between MVA and MEP routes 20

2.6 The famesyl diphosphate is a key enzyme in terpenoid biosynthetic pathway 20

2.7 The glutaredoxin system 22

Chapter 3. Materials and Methods 25

3.1 Plant materials and growth condition 25

3.2 DNA extraction and screening offps mutants 25

3.3 Validation of the fps mutant by semi-quantitative RT-PCR 26

3.4 Semi-quantitative RT-PCR analysis of the fps mutants 28

3.5 Genechip analysis oifps mutants 29 3.6 Enzyme assays 29

3.7 Triterpene and sterol analysis of加 mutants 30

3.8 Preparation of carotenoid standards for carotenoid analysis 31

3.9 Carotenoids analysis oifps mutants by HPLC 31

3.10 Subcellular localization ofFPSl and FPS2 by transient expression 33

Chapter 4. Results 36

4.1 Screening offpsl and fps2 homozygous mutants 36

4.2 Validation offps mutants by RT-PCR and enzyme activity assay 36

4.3 Genechip analysis of two fps mutants 40

4.3.1 Quality control and normalization of microarray sample 40

4.3.2 Normalization and identification of differentially expressed genes 42

4.3.3 GO annotation of differentially expressed genes in fps mutants 43

4.3.4 Genes participate in stress and defense response were differentially

expressed in both fpsl and fps2 mutants 48

4.3.5 Genes in the plastidial pathway were down-regulated 51

4.4Effects of FPS mutations on pathway enzymes 53

4.5 Effects oifps mutants on terpenoids and sterol metabolism 55

4.6 Comparison on carotenoids and chlorophyll contents 55

4.7 Subcellular localization ofFPSl and FPS2 61

Chapter 5. Discussion 62

Chapter 6. Conclusion 67

Reference 68

Appendices 79

Appendix A. Primers designed for homozygous mutant screening for fps mutants

79

Appendix B. Primer pairs designed for fps ORF, common sequence and fpsl

Y specific region 80

Appendix C. Primer pairs designed for studying expression level of the downstream genes of FPS 81

Appendix D. Annotations of differentially expressed genes in fpsl mutant 82

Appendix E. Annotations of differentially expressed genes in fps2 mutant 84

Appendix F. Log fold changes of terpenoid pathway genes involved in FPS mutants 94

xi List of Figures

Figure 1 Schematic diagram showing an overview of the major reactions in different compartmentation involved in the terpenoid biosynthetic pathway in higher plants.. 2 Figure 2 Illustration showing the priming effect of plant volatiles emission 7 Figure 3 Schematic diagram showing the formation of different classes of terpenes in the MVA and MEP pathways 10 Figure 4 Chemical structure of some examples of pharmaceutical terpenes 11 Figure 5 The interconnection between primary and secondary metabolisms 14 Figure 6 The prenyltransferase reaction 15 Figure 7 Enzymatic reactions involved in the MVA route 17 Figure 8 The seven enzymatic reactions involved in the MEP roiite 19 Figure 9 The diagram showing exon and intron regions of the fpsl and fps2 22 Figure 10 Enzymatic reactions of the thioredoxin and glutaredoxin 24 •Figure 11 Illustration showing the key regulatory role of the GRX480 in the crosstalk between the SA pathway and JA pathway 24 Figure 12 The primer positions and predicted size ofPCR products 27 Figure 13 A schematic diagram of the MVA pathway showing the major enzymes involved in the pathway and the genes chosen in the preliminary semiquantative RT-PCR study. 28 Figure 14 Fluorescent fusion constructs of the FPSs 35 Figure 15 Gel photos of the PCR analysis for mutant screening 37 Figure 16 Morphology of wild type Arabidopsis plants and fps mutants 38 Figure 17 The alignment offpsl and fps2 gene sequences 39 Figure 18 Gel photos showing the result of PCR verification offpsl mutant 39 Figure 19 Farnesyl diphosphate synthase (FPS) activity in the fps mutants and Col-0 wild type 40 Figure 20 Gel photos showing total RNA extracted from leaves of wild type Columbia-0 and fpsl, fps2 41 Figure 21 RNA degradation plot for the six genechips 41 Figure 22 Boxplot of the six genechips showing the normalization of the data distribution of each dataset 42 Figure 23 Pie charts showing the distribution of the differential expressed genes

xii grouped according to their GO molecular functions 46 Figure 24 Pie charts showing the distribution of the differential expressed genes grouped according to their GO biological processes 47 Figure 25 Schematic diagram of the terpenoid biosynthesis pathway in Arabidopsis thaliana) 51 Figure 26 Sesquiterpene cyclase activity in the fps mutants and Col-0 wild type•.…53 Figure 27 Squalene synthase (SQS) activity in the fps mutants and Col-0 wild type. 54 Figure 28 GC-MS profile of A) Col-0 wild type, B) fpsl mutants and C) fps2 mutants. 57 Figure 29 Relative corrected peak areas of the fps mutants and Col-0 wild type 58 Figure 30 HPLC profile of A) Col-0 wild type, B) fpsl mutant and C) fpsl mutant Arabidopsis at 450nm 59 Figure 31 Cartenoids content in the fps mutants and Col-0 wild type plant 60 Figure 32 Fluorescent microscopic images of A) FPSl-GFP; and C) FPS2-GFP visualized using a confocal laser-scanning microscope 61 Figure 33 Summary of the changes in gene expression level, enzyme activities and metabolites of the two fps mutants in the present study. 62

xiii List of Tables

Table 1 fps mutants used in this study 25 Table 2 Primer pairs designed for FPSl-GFP construct and FPS2-RFP construct.... 35 Table 3 Genes that were regulated in both fpsl and fps2 mutants 49 Table 4 Differentially regulated GRX in fps mutants 50 Table 5 Differentially expressed TRX in fps mutants 50 Table 6 Summary of Log2 fold change in expression level of terpenoid pathway genes 52

xiv List of Abbreviations

Acetoacetyl-CoA Thiolase AAT Acetonitrile ACN base pair bp Border Primer BP Bovine Serum Albumin BSA Cycloartenol Synthase CAS complementary DNA cDNA Coenzyme A Co-A Columbia-0 Col-0 Cetrimonium Bromide CTAB Dichloromethane DCM Differential Expressed DE Digoxigenin DIG “Dimethylally Diphosphate DMAPP Deoxyribonucleic Acid DNA 1 -Deoxy-D-Xylulose 5-Phosphate DOXP Disintegrations Per Minute DPM DOXP Reductoisomerase DXR 1 -Deoxy-D-Xylulose 5-Phosphate Synthase DXS Famesyl Pyrophosphate FPP Famesyl Diphosphate Synthase FPS Gas Chromatography-Mass Spectrometry GC-MS Green Fluorescent Protein GFP Geranylgeranyl Pyrophosphate GGPP Green Leafy Volatiles GLV Geranyl Diphosphate GPP Glutaredoxin GRX Monothiol Glutaredoxin Family Genes GRXS Glutathione GSH Homozygous Mutant HM Hydroxymethylglutaryl-CoA HMG-CoA HMG-CoA Reductase HMGR HMG-CoA Synthase HMGS High-Performance Liquid Chromatography HPLC Heterozygous Mutant HZ xvi Isopentenyl Pyrophosphate IPP Jasmonic Acid JA Liquid Chromatography LC Left Primer LP Lupeol Synthase LUP Methylerythritol 4-Phosphate MEP 2-(N-Morpholino)Ethanesulfonic Acid MES Murashige and Skoog MS tert-Butyl Methyl Ether MtBE Mevalonate MVA Mevalonic Pathway MVA pathway 5-Diphosphomevalonate Decarboxylase MVD MVA Kinase MVK 5-Phosphomevalonate MVP 5-Diphosphomevalonate MVPP National Institute of Standards and .Technology-Mass Spectral NIST-MS Nonexpressor of Pathogenesis-Related proteins 1 NPRl Optical Density OD Open Reading Frame ORF 3，-Phosphoadenosine -5 ‘ -Phosphosulphate PAPS Polymerase Chain Reaction PCR Photodiode Array Detector PDA 5 -Phosphomevalonate Kinase PMK Polytetrafluoroethylene PTFE Red Fluorescent Protein RFP Robust Multichip Analysis RMA Ribonucleic Acid RNA Right Primer RP Relative Standard Deviation RSD Reverse Transcription Polymerase Chain Reaction RT-PCR Salicylic Acid SA Squalene Monoxygnease SQP Squalene Synthase SQS The Arabidopsis Information Resource TAIR Thin Layer Chromatography TLC Thioredoxin-Dependent Peroxidase IPX Thioredoxin TRX

xvi � World Health Organization WHO Wild Type ？\mt ^ ( WT ,:.”〜.，〜 iapfer 诱r�:.r.mn K :“<.� 、.Vts、. > ‘•

j • ... * �..i•丨•

. •• • . • . •

T.. ；" ri'im mv ！K：广、、丄•杰 “.：.<、•:、-';..'...;,“‘ . “.,::� ...... j ... -• • ‘ 二.. ,»••. •

• .. , .一.. ‘ .v. - ‘" 缚 ‘ fen!::.射 ；： :，： ：• r^r.in^ j: W't-VV • ...• . ：.*^、• 询 产、，产....�:":.4't .k •.,�•-：...、广；“�� !V^- . ,.、..％.'.. - . .... ‘

—. .-.、，-‘叉.. •

‘‘•‘殺 ri^y/srniKJi-i^^to；；.! u,’�� >:v ...... 一.」.;：；沈t. - •....V. 」• ” •• • • ‘ • - . . • . ；•.： -，arid -'rl'V!;.. ...丄.^..j-v： ：众f..., • - - . • • .>• • • .... • - • � ..•, ‘‘

• -f ‘ •

- �, “ ••‘

r-n;、. ：...... • - • -

fV'/iH ^ ； . ‘.人-:一 ：. -r._- 1 . .It"‘ ‘‘. ： � ••• .

.,..-‘ - • • i-ir.u-'A • ： , .：.•：.- , ‘ .... mSn OT^ii H . :.:,�'‘.:-. : .,.‘.::� •: , -;.,,.,..�.... _ .�.,. ‘ V > ；—— ‘ r s » .

- • r S� * • • .

•：；. 灼‘如於、vV - ‘�

--、、..•…- ： ....'-々 • ’ ... ：‘-. uf. ../、. ...‘.-: .. ；..... ‘ ？. • • • • ；.;•. • . .;.•�/'�V)-- . •. ....:广.…丄 •..�•.-.:.. - V - ‘ • ...,.• • • . ： • ‘ ‘ • , ‘ .'^： :.' . .. . . , . , • •• “ 、》是?; Knf'玄《”今—.m躲:.？化。：...：： ..； C ���

.» "‘ » • “ - • •

^ ：• . - ‘ • ‘ . 、- Chapter 1. General Introduction

Plant secondary metabolites are natural products often associated with the defense behavior of plants, but not directly essential for their survival and growth.

Terpenoid is one of the largest family of secondary metabolite found in many plant species (Hemmerlin et al., 2004). It consists of a diverse group of organic compounds important to plants in various aspects. For example, they serve as cell membrane components, hormones and photoprotection pigments. Some terpenes also play important roles in ecology as they participate in indirect plant-insect interaction and plant-plant interaction. Terpenoid also form the basis of pharmaceuticals and commercial products (Withers & Keasling，2007)，such as, the anti-malarial agent artemisinin, the anti-cancer agent paclitaxel (Taxol), eleutherobin, dyes, flavors, and perfumes etc.

In higher plants, terpenoids can be synthesized in both cytosol and plastid. The cytosolic mevalonic (MVA) route and the plastidial methylerythritol 4-phosphate

(MEP) route produce the common precursor isopentenyl pyrophosphate (IPP), which is subsequently used to generate other terpenoids (Figure 1).

Much research efforts have been made in order to elucidate the terpenoids

biosynthetic pathway, and many pathway enzymes have been identified and

functionally characterized (Lichtenthaler, 1999; Rodriguez-Concepcion & Boronat，

2002; Hunter, 2007). However, the regulatory aspect of terpenoid biosynthesis is

still poorly understood.

1 I t Mitoyt^mu.. ^^

vK^Acetyl CoA ->HMG CoA ->MVA ->IPP ->FPP ->sesquiten^ne^ " 这资

Figure 1 Schematic diagram showing an overview of the major reactions in different compartmentation involved in the terpenoid biosynthetic pathway in higher plants (modified from Bouwmeester, 2006). MEP route is located in the plastid while MVA route occurs in the cytosol. The cytosolic pathway starts from acetyl Co-A to FPP that is the precursor off sesquiterpenes whereas the plastidial pathway starts with pyruvate and G3P, producing monoterpene as the end product with the following metabolites involved: Acetyl Co-A, Acetyl Coenzyme A; HMG Co-A, 3-hydroxyl-3-inethylglutaryl-Coenzyme A; MVA, (3R)-3,5-dihydroxy-3-methylpentanoic acid; IPP, isopentenyl diphosphate; DMAPP, dimethylallydiphosphate; FPP, farnesyl diphosphate; G3P, gIyceraldehyde-3-phosphate; MEP, 2C-inethyl-D-erythritol-4-phosphate; GPP, geranyl diphosphate.

20 Investigation of the terpenoid pathways can be achieved through pathway perturbation. Using drugs that block the MVA and MEP routes respectively, previous study suggested a possible crosstalk between cytosolic and plastidial routes in

Arabidopsis (Laule et al., 2003).

Famesyl diphosphate synthase (FPS) is one of the key enzymes responsible for the production of FPP, which is further converted into various types of sesquiterpenes, triterpenes, and polyterpenes. Two FPSs genes {fpsl and fpsl) have been identified and cloned in Arabidopsis. Although their temporal and spatial expression pattern has been characterized (Cunillera et al., 2000)，the exact functional roles of these two enzymes remain largely unknown (Masferrer et al.,

2002; Manzano et al., 2004，Manzano et al., 2006). A comprehensive survey of transcriptome and targeted metabolic profile in fps mutants will facilitate our understanding about the roles of the two FPSs in terpenoid biosynthesis and provide new insights into the regulatory mechanisms of terpenoids biosynthesis at systematic

level.

In the present research, Affymetrix Genchip technology was used to monitor the

transcriptome profile and GC-MS/HPLC methods were used to detect the targeted

metabolites in two fps mutants. The high-throughput data were compared and

contrasted with that from wild type Arabidopsis. The results showed no direct

correlation between gene expression and metabolite concentration, suggesting that

post-transcriptional regulation may play important roles in controlling the metabolic

flux in terpenoids biosynthesis. More genes were found differentially modulated in fps2 mutant than that in fpsl mutant, suggesting the distinctive roles of these two

3 enzymes. The fact that knockdown of FPS, inhibition of FPP (downstream of IPP), results in changes in metabolites produced from plastidial MEP pathway indicates that crosstalk between the two compartments may exist at other points besides IPP.

.. —. 1 - ’

4 Chapter 2. Literature Review

2.1 The importance of terpenoids

Primary metabolites are organic compounds participate in nutrition and essential metabolic processes. On the contrary, secondary metabolites are not necessary for the survival and growth of an living organism. They are usually associated with defense and/or interspecific competition behavior, and hence are

.important for the long term survival and fitness of the organism. For example, they prevent plants from pathogen or herbivorous attack and enhance insect pollinations.

Terpenoids is the largest class of plant secondary metabolites with over 25,000 individual compounds described (Douglas & Rodney, 1995; Buchanan et al.，2000;

Hemmerlin et al., 2004). They play important roles in various aspects in plant metabolisms. Terpenoids not only serve as membrane components (sterols) and hormones (gibberellins, abscisic acid, cytokinins), they are also involved in electron

transport system (quinones), photoprotection (carotenoids) and subcellular targeting

and regulation etc. Some terpenoids (monoterpenes, sesquiterpenes, and diterpenes)

also act as messengers for the plant to interact with their surrounding environment,

such as plant defense compounds and pollinators attractants (Chappell, 1995a;

Dixon, 2001). Terpenes can be detected by antennae of insect and recognized in a

metabolite specific manner (Schultz, 2002; Schmelz et al., 2007). Plants produce

terpenoids to attract specific predatory or parasitic insects (De Moraes et al, 2001;

5 Schnee et al.，2006; Weech et al., 2008 ), which in turn discourages other herbivores from laying eggs and protects the plants against grazing (Kessler & Baldwin, 2001).

In plant-plant interaction, damaged plants emit green leafy volatiles (GLV; e.g.

6-carbon aldehydes, alcohols and esters) to induce the jasmonic acid production and trigger the terpenoid emission like sesquiterpenes (Schnee et al., 2006; Steeghs et al.,

2004). The mixture of GLV and terpenes give neighboring plant a priming effect

(Baldwin et al., 2006) (Figure 2), which is similar to vaccination in human. After exposure to the volatile mixture, specific chemicals that mimic endogenous defense-related signaling compounds are produced by the neighboring plants

(Baldwin et al., 2006). The exposed plants subsequently receive less damage, while the receiver plants is benefited by reducing investment in defenses until the onset of actual herbivory. Therefore, terpenoids act as main communication messengers among plants and their environment.

Terpenoids also have great potentials in pharmaceuticals and commercial application. They form major components of natural pharmaceuticals and commercial products (Withers & Keasling，2007), such as anti-malarial agent artemisinin, anti-cancer agent paclitaxel (Taxol), eleutherobin, dyes, flavors, and perfumes. Artemisinin is a sesquiterpene lactone that relieves the situation of the wide spread quinine resistant protozoan Plasmodium falciparum causing the most

dangerous malaria. Artemisinin-based therapies has become the most promising

treatment for malaria recently (WHO, 2003; WHO, 2006). Artemisinin and its

derivatives have also been suggested to have potential use for the treatment of other

diseases such as schistosomiasis and cancer (Xiao, 2005; Efferth, 2006).

6 (

inmlll ii f ^SeSSdatile • 働 Artemisia tridentata Nicotiana attenuata Figure 2 Illustration showing the priming effect of plantn volatiles emissio n (modified from Baldwin et al., 2006). When the neighbouring plant receives plant volatiles generated from herbivorous insect attack, it will exhibit a more vigorous defense response towards the same insect attack.

7 Taxol is a complex diterpenoid effective against lung, ovarian, and breast cancer. Taxol is isolated from the bark of the Pacific yew {Taxus brevifolia) tree. It acts as a mitotic inhibitor that inhibits mitosis and cancer growth (Jordan & Wilson,

2004). Eleutherobin is a another diterpene with similar function as Taxol (Long et al.,

1998; Kingston, 2009). Different from Taxol, eleutherobin is extracted from the marine soft coral.

Terpenoid biosynthetic pathway can be activated when the plants are under oxidative stress (Baxter et al., 2007; Reinhold et al., 2008). Terpene emission in plants following the daily photosynthetic patterns is found to be able to relieve the oxidative stressful condition caused by high temperature and intense irradiation

(Loreto et al., 1996; Penuelas & Llusi‘ 1999’ 2000; Mayrhofer et al., 2005).

Common volatile terpenes like isoprene and monoterpenes in plants are found to be released under stress. Even when the stomata are closed, terpenes built up in the intracellular region can still diffuse out of the plant (Bertin & Staudt，1996). As a result, the cell membrane is stabilized under the stress condition.

2.2 The difficulties in synthesizing terpenoids

Many plants only produce limited amount of important terpenoids. For

example, the yield of artemisinin from the Artemisia annua plant is typically about

0.05-0.2%, with a maximum of less than 1%. Similarly, the Taxol isolated from the

plant is not enough for clinical trial because they are obtained from slow growing

8 and endangered plants Taxus sp. (Lixin & Arnold, 2005). Enhancing the desirable compounds from plant extracts is therefore the ultimate goal for metabolic engineering. Due to the structural complexity of many terpenoids, total chemical synthesis from commercial precursors (Lixin & Arnold, 2005) is not feasible.

Although semi-synthesis is available (Lixin & Arnold, 2005; Ro et al.，2006)，the chemical synthesis of terpenoids remains difficult and costly for scale-up mass production (Chang et al, 2007). Metabolic engineering provides a promising avenue for producing pharmaceutically important terpenoids in a cost effective way.

However, limited knowledge on the regulatory mechanism underlying terpenoid biosynthesis has hindered the metabolic engineering effort to efficiently overproduce desired terpenoids compounds.

2.3 Structure and classification of terpenoids

Terpenes can be classified on the basis of their carbon numbers (Figure 3).

Terpenoids are modified terpenes with oxygen group added or methyl groups moved or removed, but the term "terpenes" refers to both terpenes and terpenoids. They have the carbon skeleton consisting of different combination of isoprene unit and are classified according to the number of units. Compounds with two isoprene units, with ten carbon atoms is named as monoterpenoid while compound with three isoprene units (15 carbons) is called sesquiterpenoid. Similarly, diterpenoid, triterpenoids and tetraterpenoids have four (20 carbons), six (30 carbons) and eight

(40 carbons) isoprene units, respectively. Compounds with more than eight units are

grouped into polyterpenoids. Many pharmaceuticals fall into different categories in

terpenoid family (Figure 4).

9 Isopentenyl Diinethylallyl •H«inilerp«nes \__// diph明pha丨e — i删咖diphosphate (Isoprcocs) J~

• Head- tail condensation c 丨。

dipSphale ^ Monoterpene« plastidial 广一 ^Ji^CH.O®® (IPP)

2X 5rhSpha.P 隱 cytosolic PA^CH/柳卵 丨

I Ceranylgeranyl • Diterpenes plaslidial diphosphale 、‘令 2 f&Bi

Squalen^ » Tril^rpenes cytosolic �- •2

Phytoene Tetralerpenes plastidial

1 丄 chdM Jn

Polyprenyl diphosphate PotytcrpoKs cytosolic

Figure 3 Schematic diagram showing the formation of different classes of terpenes in the MVA and MEP pathways (modified from Buchanan et al.，2000). The classification of terpenoid skeletons is based on the number of carbon atom which is shown on the left hand side of the figure: C5 denotes five carbon atoms while CIO denotes 10 and so on. The arrows indicate the direction of biosynthesis, whereas the label plastidial or cytosolic on the right hand side indicate the compartments that the terpenes are belonging to.

10 / o

H i 八 /^OMe = o A) Artemisinin B) Sarcodictyin A

r^ 0 o

W OH \=ol 人、OAc

^： C) Paclitaxel (Taxol) D) Eleutherobin

Figure 4 Chemical structure of some examples of pharmaceutical terpenes. A) Artemisinin, a sesquiterpene lactone; B) Scarcodictyidn A; C) Taxol; and D) Eleutherobin, diterpene alkaloids

11 2.4 MVA and MEP pathways of terpenoid biosynthesis in higher plants

Terpenoid biosynthetic pathway generates both secondary and primary metabolites (Figure 5). The five carbon isopentenyl diphosphate (IPP) is recognized as isomer or an isoprene unit (Figure 3). The head to tail condensation of IPP and dimethylally diphosphate (DMAPP) yields geranyl diphosphate (GPP) for formation of monoterpene and some monoterpene alkaloids. With one more IPP molecule added to the GPP, famesyl diphosphate (FPP) generate sesquiterpene with 15 carbons. Two FPP molecule form one squalene molecule, which then produces triterpenes (Chappell, 1995b). On the other hand, FPP can react with IPP to form

•geranylgeranyl diphosphate (GGPP), which produces diterpenes. Two GGPP combine to form the precursor of tetraterpene. When large number of isoprene units are condensed, the long polymer polyterpenes are generated (Figure 3).

In higher plants, two distinctive routes lead to terpenoid biosynthesis. The traditionally known mevalonate (MVA) pathway is localized in the cytosol whereas the subsequently found methylerythritol 4-phosphate (MEP) pathway is confined to the plastid (Rodriguez-Concepcion & Boronat，2002, Lunn, 2007; Figure 1).

Terpenoid biosynthesis in higher plant is fundamentally consisted of four main steps:

1. Synthesis of the common precursor isopentenyl diphosphate (IPP),

2. Subsequent additions of IPP to form a series of prenyl diphosphate

homologs, the famesyl diphosphate (FPP), geranylgeranyl diphosphate

(GGPP) and polydiphosphate by the prenyltransferase reaction (Figure

12 6)

3. Formation of terpenoid skeletons by conversion of the prenyl

diphosphate with specific terpenoid synthase to its appropriate

conformation

4. Enzymatic modifications (redox reactions) of the skeletons

13 C〇2 HSI^SSH

^ I

I ‘ ^^^

Figure 5 The interconnection between primary and secondary metabolisms (modified from Taiz & Zeiger，2002). Terpenes are one of the major end products of secondary metabolism and all of their precursors are generated from primary metabolism. On the other hand，the terpenes produced are involved in both primary and secondary metabolisms. MEP: Methylerythritol 4-phosphate

14 j .. •�八。pp + IPP 丄 hT 卿 R 八Z�)麵

A Hylic diphosphate oster (ii carbons) AHylic diphosphate eslei (ii + 5 carbons!

Figure 6 The prenyltransferase reaction (modified from Buchanan et al., 2000). The diphosphate is first ionized by a divalent metal cation resulting in an anion. The anion reacts with IPP molecule to generate isoprenolog that would become a five carbons longer diphosphate after final deprotonation.

• •.

15 2.4.1 The MVA pathway

The cytosolic MVA pathway (Figure 7) starts from acetyl coenzyme A

(acetyl-CoA), which is then converted to hydroxymethylglutaryl-CoA (HMG-CoA) by acetoacetyl-CoA thiolase and HMG-CoA synthase. The HMG-CoA reductase

(HMGR) produces the intermediate mevalonate (MVA) from the HMG-CoA. There are three more enzymatic reactions before the completion of the IPP synthesis: MVA kinase, 5-phosphomevalonate (MVP) kinase and 5-diphosphomevalonate (MVPP) decarboxylase. The IPP produced from MVA pathway is transformed to famesyl pyrophosphate (FPP), which is used for the production of sesquiterpene, sterols and ubiquinone. Famesyl diphosphate synthase (FPS) is the key enzyme to generate FPP, which then generates a great variety of terpenes in the cytosol.

Mevalonate is the precursor for triterpenes and sterols. In the past, it was believed that the biochemical reactions and enzymes involved in terpenoids biosynthesis all occur in cytosol (Brechbiihler-Bader, 1968; Flores-Sanchez et al,

2002). IPP was originally thought to be generated by MVA pathway only (Porter &

Spurgeon, 1981). However, not all IPP was produced from mevalonate in green alga

and eubacteria (Rohmer et al., 1993，1996; Schwender et al, 1996). Some terpenoids,

such as isoprene, carotenoids (tetraterpene), diterpene, monoterpene and chlorophyll

can be produced by a non-mevalonate pathway in the plastid as well (Eisenreich et

al., 1996，1997; Lichtenthaler et al, 1997; Dubey et al., 2003). Other terpenoids like

polyterpene, sesquiterpene, triterpene and phytosterol are considered to be generated

in the cytosol through IPP produced by the MVA pathway (Rodriguez-Concepcion

& Boronat, 2002). Recent findings proved that sesquiterepene in higher plants can

be generated from IPP precursors in MEP pathway as well (van Klink et al, 2003;

16 Sallaud et al., 2009).

: ^

Acetyl Coenryme A

f AGetoaeetyl-GoA-thiolase- AAT)

Acetoacetyl Coenzyme A

^^^^^^^nM^e^ynthase-(HMGS) JQ� f“- �

3"hydroxy-3"nwlhylgkit«iyl"Co>n^nfA

^―^―-5RM^eS3duetase-(HMGR)

！

-^MVA-kinase-(MVK-)

Mevalonate-S-phosphate

警 S-phosphofnevalBnate-kiiaase (PMK)

"x^j^�丄。

Mevatonata-S-pyropho«phate

S^^phof^alonate decarboxylase (MVD)

Dimj^gj^gpwte l8open|w^|K>8phate

isopentenyl pyrophodsphate isomerase (IPI)

Figure 7 Enzymatic reactions involved in the MVA route from the primary metabolite Acetyl-CoA to the common end product IPP for both MVA and MEP pathway.

7 2.4.2 The MEP pathway

The non-mevalonate route occurs in plastid. The first enzyme in the route, namely, 1 ^deoxy-D-xylulose 5-phosphate synthase (DXS), was identified in 1998

(Lange et al., 1998). Other enzymes involved in MEP pathway have also been characterized afterwards (Lichtenthaler, 1999; Rodriguez-Concepcion & Boronat，

2002). The plastidial MEP pathway (Figure 8) involves condensation of pyruvate and glyceraldehyde-3-phosphate for the first intermediate, 1 -deoxy-D-xylulose

5-phosphate (DOXP), which is then converted to 2-C-methyl-D-erythritol-4- phosphate (MEP) for' the synthesis of many important compounds including carotenoid, diterpene, monoterpene, and abscisic acid. The plastoquinone, isoprene, and the side chains of chlorophylls are also produced by the MEP route and subsequent downstream reactions (Dubey et al., 2003).

MEP pathway is absent in humans but present in some severe human pathogens

including malaria (Govindarajan et al., 2007)，tuberculosis, and a series of sexually

transmitted diseases (Hunter, 2007). Enzymes in the MEP pathway down to the

point of IPP generation specifically consume their substrates and generate a specific

product or function that cannot be compensated by another enzyme (Lichtenthaler,

1999; Rodriguez-Concepcion & Boronat，2002; Hunter, 2007). Accordingly,

enzymes in the MEP pathway provide attractive targets for drug treatment.

18 "V + ^^本 pyruvate G»ycefaldehyde-3-phosphate (G3P)

丄 DXS ^ ^^P x丨爾 Vt"个 1 -deoxy-D-xykilose-S-phopshate (DXP)

ISPC %？磁 i 0尋 2C-methyl-D-erythritoM-phosphate (R^P)

、，D o o o ^

""""^‘pC�如> 4

O O O o d i i 1 II • �p^\�T个z�十个 4^diphMphocytidyl>2CHnelhyl~I>«iythrilol<2

2C«iythrito(<2’4-cydodiphosphate (ME-cPP) IspG

im in 4-hydroxy-3-fnethyl-but-2-enyl pyrophosphate (HMBPP) IspH IspH ^ \ typel typell

“•P卜_ A纏-•七‘ isopentenyl diphosphate (IPP) dimethylallydiphosphate (DMAPP) Figure 8 The seven enzymatic reactions involved in the MEP route (Modified from Hunter， 2007). The words in blue indicate the enzymes involved in each step with their secondary structure shown by ribbon diagrams next to it DXS, 1 -deoxy-D-xylulose-5-phosphate synthase; IspC, 1-Deoxy-D-Xylulose 5-Phosphate Reductoisomerase; IspD, 2C-methyl-D-erythritol cytidylyltransferase; IspE, 4-diphosphocytidyl-2-C-methylerythritol kinase; IspF. 2C-methyl-D-erythritol 2,4-cyclodiphosphate synthase; IspQ, hydroxy-methylbutenyl 4-diphosphate synthase; IspH, 4-hydroxy-3-methylbut-2-enyl diphosphate reductase; IPI, isopentenyl diphosphate isomerase (There are type IIPI converts DMAPP to IPP and the type II IPI converts IPP to DMAPP).

19 2.5 The crosstalk between MVA and MEP routes

Although located in separate compartments, it was suggested that crosstalk exist between MVA and MEP routes (Hemmerlin et al., 2003; Laule et al, 2003).

The interactions between the cytosolic and plastidial routes, however, were poorly documented (see Lichtenthaler, 1999; Dubey et al., 2003; Lange & Ghassemian，

2003; Hemmerlin et al., 2006 )

Using drug lovastatin (HMGR inhibitor) and fosmidomycin (DXR inhibitor) to

block the MVA and MEP routes respectively, previous studies (Laule et al., 2003)

showed that the gene expression patterns of pathway enzymes, e.g. DOXP

reductoisomerase (DXR) and HMGR (HMGl) did not show significant change

compared to the control. Contrarily，the terpenoids levels changed rapidly upon the

inhibitor treatment, suggesting that inhibitor mediated metabolic flux redirection

between the MVA and MEP pathways are likely not regulated transcriptionally.

Instead, it may be controlled by increasing level of translation, posttranslational

enzyme modification, posttranscriptional regulatory processes or combined effect of

the above. These evidences suggested further divergence of metabolic fluxes and

cross flow between the compartments (Laule et al., 2003).

2.6 The famesyl diphosphate is a key enzyme in terpenoid biosynthetic pathway

Famesyl diphosphate synthase (FPS) is the key enzyme lies at the multiple

branching point of the terpenoid biosynthesis pathway. It catalyzes the synthesis of

famesyl pyrophosphate (FPP), the critical volatile precursor, by the head and tail

20 condensation of one dimethylallyldiphosphate (DMAPP) with two isopentenyl diphosphate (IPP). FPP can be used to produce different groups of terpenes, including sesquiterpene, triterpenes, phytosterol, dolichol and polyterpene.

Four FPS isoforms were found in Arabidopsis thaliana. They are: FPS IS,

FPSIL, FPS2.1 and FPS2.2. FPSl is the first known FPS and the

famesyl-diphosphate synthase 2 (FPS2) in Arabidopsis thaliana was discovered in

1996 by Southern blot analysis (Cunillera et al., 1996). Both fpsl (AT5g47770) and fps2 (AT4gl7190) genes have been cloned and characterized (Cunillera et al., 1996). fpsl and fps2 share a high level of sequence similarity in both the intron and exon

regions. fps2 gene consists of 11 exons and 10 introns whereas the fpsl gene has 12

exons and 11 introns (Figure 9). The exon 4 in the fps2 gene corresponds to the

exons 4 and 5 in the fpsl gene (Figure 9). FPSIL was the first FPS reported to

possess the mitochondrial transit peptide NH2-tenninal extension of 41 amino-acid

residues. The mitochondria targeting FPS was found common in eukaryotes (Martin

et al., 2007). To distinguish the two FPSl isoforms, the one without NH2 terminal is

subsequently referred as FPSIS. FPS2.2 is an isoform of FPS2 produced by

alternative splicing and the name FPS2.1 and FPS2.2 indicate they are originated

from the same FPS.

The FPSIS and FPSIL localize in the cytosol and the mitochondria respectively.

Northern Blotting Analysis (Cunillera et al., 1997) showed that they have different

expression pattern. FPSIS mRNA has the highest expression in roots and

inflorescences. The FPSIL and FPS2 mRNA express preferentially in inflorescence

but with a lower expression level compared to FPSIS (Cunillera et al, 2000).

Transgenic plant overexpressing FPSIS exhibits senescence-like response due to the

21 decline of HMGR (Masferrer et al., 2002; Manzano et al., 2004). The phenotype of

FPSIL overexpressor could be partially rescued by HMGR restoration and oxidative stress was found in the mitochondria compartment (Manzano et al.，2006).

Researchers speculate a tight control of the cytosolic pool of FPS in Arabidopsis

(Aharoni et al., 2003).

AT5G47770.1 FPS1 --mm • mmmr-'-mm- mmmr AT4G17190.1 FPS2.1

粉---.^minn^. - - - ^^ — .�flifr一 ——一 ^hbh^^^maumm^ AT4G17190.2 FPS2.2

‘ •….• • * — 一 〜* .— Key: MHM- :exon :intron

Figure 9 The diagram showing exon and intron regions of the fpsl and fi)s2. The structure between of fpsl and Jps2 are similar and the exon 4 and 5 in fpsl are corresponding to the exon 4 of fps2.1 (commonly known as fpsl). The FPS2.2 is the alternative splicing isoform of fi)s2 (FPS2.1)

2.7 The glutaredoxin system

Glutaredoxin (GRX) system participates in cellular stress and defense response ubiquitously. It was previously considered as a backup system for thioredoxin (TRX) system because both glutaredoxins and thioredoxins have similar gene numbers in

Arabidopsis (Rouhier et al., 2004) and similar cellular function (the thiol-disulfide redox reaction) in the formation of deoxyribonucleotide reductase and

3 ‘-phosphoadenosine -5 ‘-phosphosulphate (PAPS) reductase (Porat et al., 2007)

22 (Figure 10). They both also involve in the defense response against oxidative stress

(Femandes & Holmgren, 2004).

Thioredoxin and glutaredoxin belong to the thiol-disulfide oxidoreductases protein family with the typical active site CxxC (two cysteine residues with intramolecular disulfide bond) or CxxS (monothiol bond). Thioredoxins play important roles in redox signaling that senses the redox state and transmits this information to other signaling molecules in chloroplasts，mitochondria and cytosol

(Fujino et al, 2006). Other roles of glutaredoxin in the cellular function are still mysterious (Rouhier et al., 2004).

Although both glutaredoxin and thioredoxin function in the same redox

reaction, they are slightly different in their recycling mechanism. Thioredoxin is

directly reduced by the thioredoxin reductase after oxidation (Figure 10). In contrast,

the reduction of the glutaredoxin is catalyzed by the glutathione (GSH) but no

glutaredoxin reductase (Figure 10; Holmgren, 1989). When the thioredoxin system is

inactive, the glutaredoxin system can compensate for their biological function.

The glutaredoxin GRX480 is the key regulator in crosstalk of the salicylic acid

(SA) and jasmonic acid (JA) pathway (Ndamukong et al., 2007). As the downstream

regulator in the SA response, GRX480 was regulated by the nonexpressor of

pathogenesis-related proteins 1 (NPRl) (Figure 11). The increased GRX480 has

inhibitory effect on the jasmonate-dependent defensin. TGA2, TGA5 and TGA6 are

required for GRX480 function as regulator itself and they together form the

inhibitory complex to repress the plant defensin PDF 1.2 gene transcription.

23 NADPH+rf Thioredoxin reductase V J

TRX-S2 TRX-(SH)2 z

1)Soi ' >PAPS PAPS reductase ^ PAP+SOs »Cystein > Protein 2) Met-SO Mei~so reductase ^ Met »Protein 3) rNDP Ribonucleotkte reductase ^ dNDP > dNTP > DMA

GRX-(SH)2 GRX-S2 ^^^

GSSG 2GSH ^ -r< Glutathione reductase I \ NADPH+rf NADP*

Figure 10 Enzymatic reactions of the thioredoxin and glutaredoxin. Thioredoxin is directly reduced by thioredoxin reductase, while glutaredoxin is reduced by glutathione. Reductase is available for glutathione instead of glutaredoxin. TRX-S2： oxidized thioredoxin; TRX- (SH)2： reduced thioredoxin; GRX-Si： oxidized glutaredoxin; GRX-(SH)2： reduced glutaredoxin (Modified from Holmgren, 1989)

SA ERFI Salicylicacid-^^y _, pofi.i/

Figure 11 Illustration showing the key regulatory role of the GRX480 in the crosstalk between the SA pathway and JA pathway (Modified from Ndamukong et al., 2007).

24 Chapter 3. Materials and Methods

3.1 Plant materials and growth condition

Table 1 fps mutants used in this study

Locus Gene name T-DNA Mutant name Location At5G47770 fpsl SALK_073576 jpsl Exon 2 At4G17190 fps2 SALK_045864 fps2 promoter

Seeds of A. thaliana fpsl and fps2 mutants (F3) were obtained from Arabidopsis

• Biological Resource Center (Ohio State University) (Table 1). Seeds were stored for

more than two days at 4 and sterilized with 30% bleach for two minutes. The

seeds were then rinsed for five times in sterilized water before germination on solid

Murashige and Skoog (MS) medium containing basal salt mixture (MS)

(Sigma-Aldrich M5524) supplemented with 1% sucrose, 0.5g/L

2-(N-morpholino)ethanesulfonic acid (MES), pH 5.7，0.9% Bacto agar with 100

mg/L kanamycin (Mentewab & Stewart，2005). Seeds were put under 4°C for two

days to break the dormancy before germination under 16 /8 hours light/dark

condition at 22 °C with 70% humidity.

3.2 DNA extraction and screening of fps mutants

DNA was extracted from leaves with Kang's cetrimonium bromide (CTAB)

DNA extraction method (Kang et al, 1998). The identities of the mutants were

25 further confirmed by polymerase chain reaction (PCR) using primers designed by the T-DNA Primer Design Tool of the Salk Institute Genomic Analysis laboratory

(see Appendix A.). Ten to twenty plants of each mutant were used for screening.

PCR was performed with three primers, border primer LBbl (BP) for priming on the t-DNA insert, left primer (LP) and right primer (RP) for priming on the genomic DNA of SALK—073576 and SALK_045864 (Figure 12). Two sets of PCR were conducted on each sample using LP-RP pair and the BP-RP pair.

Electrophoresis in agarose gel with 0.5|il/ml ethidium bromide, was used to identify

the PCR products.

The selected homozygous mutant lines were grown until maturation and their

seeds were harvested. The homozygosity of the progenies was further validated by

repeating the PCR screening. The homozygous lines (F5) were used in the

subsequent experiments.

3.3 Validation of the fps mutant by semi-quantitative RT-PCR

Because fpsl and fps2 share high sequence similarity with only about lOObp

fpsl specific region, probes in microarray genechip may not be able to distinguish

their mRNAs. In order to confirm the knockdown of FPS gene expression in the fps

mutants, further validation by semi-quantitative RT-PCR with specific primers is

necessary (See appendix B.). Arabidopsis fps mutants and wild type control

(Columbia Ecotype) were planted for 35 days, with most of them reached flowering

stage (Boyes et al., 2001). Whole vegetative parts of at least eight individual mutants

26 and control plants (Columbia-0 ecotype) were harvested separately and frozen in liquid nitrogen. RNA was extracted using the RNeasy Plant Mini Kit (Qiagen) and reverse transcribed to complementary DNA (cDNA) using the superscript II reverse transcriptase (Stratagene). Ribonucleic acid (RNA) was extracted from three types of tissues in wild type and the mutants using RNeasy Plant Mini Kit (Qiagen), and reverse transcribed to complementary DNA (cDNA) using the superscript II reverse transcriptase (Stratagene). The cDNA was used as template for different PGR reactions. The ubiquitin UBC21 was used as internal control for the normalization of reverse transcription polymerase chain reaction (RT-PCR) (Czechowski et al., 2005).

The number of cycles in the PCR program was adjusted to ensure the amount of end product in the PCR reaction was still in the exponential phase to avoid the saturation in the plateau phase. The PCR products were agarose gel electrophoresized and the

intensity of bands were quantified using Quantity One Software (Bio Rad).

Genome JBUI^^^^^ WT HZ HM ooobpHQ^muijuji^miim

Figure 12 The primer positions and predicted size of PCR products LP: left primer (forward); BP: border primer (forward); RP: right primer (reverse); WT: wild type; HZ: heterozygous mutant; HM: homozygous mutant; LP-RP reaction PCR product is about lOOObp, BP-RP reaction PCR product is about 400 to 700bp. WT and HM would have the lOOObp band and the 400 to 700bp band only respectively, while the HZ individual would have both of the two bands.

27 3.4 Semi-quantitative RT-PCR analysis of the fps mutants

The homozygous fps mutants were used in the investigation. In order to test if there is any change of gene expression pattern in the fps mutants, preliminary test was carried out. The morphology of the 35 days old A. thaliana mutants and control plant (Columbia Ecotype) was recorded and compared. Whole plants, leaves and

flowers of at least eight individual plants were harvested separately. RT-PCR was carried out as mentioned in 3.3 from three types of tissues for genes involved in

terpenoid biosynthesis pathway in Arabidopsis (Figure 13; see Appendix C.). The

intensity of PGR products was compared among mutants and wild type.

FPS1 SQS1 IPP FPS2 SQS2 SQP1 CAS1 + ^^ FPP m^ Squalene ^^Epoxysqualene lij^Cycloartenol •秦Sterols DMAPP ^ ^^ LUP^^r AT5G239^3^Group A & B Jr Triterpenes AT5G44630 Sesquiterpene Tf>te�Pene s e.g. beta-amyrin

Figure 13 A schematic diagram of the MVA pathway showing the major enzymes involved in the pathway and the genes chosen in the preliminary semiquantative RT-PCR study. SQSl: squalene synthasel(AT4G34640); SQS2: squalene synthase2(AT4G34650); SQPl: squalene monoxygneasel(AT5G24150); SQP2: squalene inonoxygnease2(AT5G24140); LUPl:Lupeol synthasel(ATIG78970); LUP2:Lupeol synthase2(ATlG78960); GpA : sesquiterpene synthase for group A sesquiterpene (AT5G23960); GpB: sesquiterpene synthase for group B sesquiterpene (AT5G44630)1

28 3.5 Genechip analysis of fps mutants

Total RNA were extracted from whole plant tissue of fps mutants and

Columbia-0 (Col-0) wild type plants and subjected to genechip (AfFymetrix

Arabidopsis Genome ATHl array) analysis. Each sample consists of pooled RNA

from eight individual plants, and samples were analyzed using two duplicated arrays.

Signal generated by hybridized probes were transferred into numerical data. The data

was normalized by Robust Multichip Analysis (RMA) method (Irizarry et al., 2003).

Each gene was fitted to a linear model, pairwise compared, and the t-statistics and

log-odds of differential expression were moderated by empirical Bayes shrinkage

.through the application of limma package from bioconductor

(http://www.bioco:ndiictor.org). The differentially expressed genes were screen out by

the following filters: 1) log fold change greater than 1.0，2) p-values smaller than

0.05%. Filtered genes were grouped by Gene Ontology (GO) under the categories of

molecular function and biological process.

3.6 Enzyme assays

The enzyme activity of three key pathway enzymes: famesyl diphosphate

synthase (FPS), squalene synthase (SQS), sesquiterpene cyclase were investigated.

The procedures for FPS enzyme assay was described by Masferrer et al, 2002). For

SQS and sesquiterpene cyclase assay, I followed the description from Vogeli &

Chappell，1988). Proteins were extracted from the whole vegetative part offpsUfps2

mutant and wild type Arabidopsis Columbia-0 with corresponding extraction

buffers for different enzymes. The amount of the protein extracted was estimated by

29 Bradford test (Walker, 2002) using bovine serum albumin (BSA) as standard. Known

amount of protein was used in the reaction mixture with radioactive labeled substrate.

Product isolation of sesquiterpene cyclase and farnesyl diphosphate synthase enzyme

assay was done by phase-phase extraction and silica powder, respectively. Thin layer

chromatography (TLC) was performed in SQS enzyme assay to isolate the product.

The product bands (squalene) were scarped and then suspended in gel form

scintillant Hydrofluor (National Diagnostics). The radioactivity of the isolated

product was determined by scintillation counting in LS 6500 Liquid Scintillation

Counter (Beckman Coulter). The activity of the enzyme was thus calculated.

.3.7 Triterpene and sterol analysis of fps mutants

To investigate the downstream metabolites of the MVA pathway, triterpene and

sterol content in wild type vs. fps mutants was studied. Whole vegetative parts of

plants were freeze dried and ground into powder. The powder (120mg) was extracted

in 1ml methanol with 0.005% cis-nerolidol (Fluka) as terpene internal standard for

half hour, followed by ultrasonication extraction at room temperature for 40 minutes.

After centrifUgation for 3 minutes at 8000 rpm, the methanol supernatant extract was

air dried and the residual was dissolved in 400 ^il dichloromethane (DCM). DCM

extrapt was filtered using Acrodisc® Polytetrafluoroethylene (PTFE) 0.45mm

syringe filter (Pall Life Sciences) and introduced to the GC-MS sampler 6890 gas

chromatograph (Agilent) equipped with a HP-5MS (5% phenyl, 95% methyl,

siloxane) column (30 m long, 0.25 mm i.d., 0.25-|im film thickness) coupled with a

5975 mass selective detector (Agilent). The temperature program was set as follows:

a ramp from 80 to 180°C at 4°C per minute, followed by a 5°C per minute ramp from

30 180 to 235°C，a steep ramp from 235 to 300 °C with 25°C per minute and a plateau

at 300°C for 20 minutes (Helium as carrier gas, 2 ml/min). The metabolites were

identified using an HP Chem Station Software based on matching with the National

Institute of standards and technology -mass spectral (NIST-MS) library of mass

spectral data.

3.8 Preparation of carotenoid standards for carotenoid analysis

Carotenoids, the major downstream metabolites in MEP pathway were

investigated. Four regular carotenoids, P-carotene (Wako), lutein (Sigma), and

.chlorophylls a and b (Sigma) (Lichtenthaler, 1987)，were chosen as the standards for

the carotenoid analysis. Standard solution was prepared as previous described (Li et

al., 2005). Standard curve was created using the four corresponding standards to

obtain a R^ value from 0.999 to 1.

3.9 Carotenoids analysis of fps mutants by HPLC

Sample preparation method was modified from previous described (Li et al.,

2005). Sample materials were freeze dried and ground into powder with mortar and

pestle. The powders (0.377 g) were subjected to extraction using 15 ml HPLC graded

ethyl acetate (Labscan) and sonicated for 15 minutes on ice. Sonicated extract was

filtered through 11 \im Whatman cellulose filter paper. The process was repeated

twice. The extracts from two rounds of extraction were pooled together and mixed

vigorously with equal volume of tert-Butyl methyl ether (MtBE) (Sigma) in a

31 separation funnel. An additional 10 ml of Milli-Q water was added and mixed in the separation funnel to facilitate phase-phase separation. The MtBE fraction was collected in a round bottom flask and underwent rotoevaporation at 150mbar, 30 for 30 minutes. The residue was transferred to a 2 ml glass vial and flushed with nitrogen gas to dryness. The non-volatile remains in the glass vial were freeze dried and dissolved in 1 ml ethyl acetate. The extract was then filtered with Acrodisc®

PTFE 0.45mm syringe filter (Pall Life Sciences) and 3 |j1 of the lOX diluted ethyl acetate extract was introduced to HPLC.

The High Performance Liquid Chromatography (HPLC) condition of the analysis was modified from literature (De Las Rivas et al., 1991). The system consists of Waters 2695 Alliance Separations Module and Waters 996 Photodiode

Array Detector (PDA) equipped with Nova-Pak C18 Sentry Guard Column (Waters) and reversed-phase Nova-Pak CI8 column, 150mm x 3.9mm I.D., 60A

(Waters). Mobile phase: (A) Methanol, (B) MilliQ Water, (C) Acetonitrile (ACN), (D)

Ethyl acetate. The solvent gradients were: 90% solvent A, 6% solvent B, and 4% solvent D for 7 minutes followed by 76% solvent A, 4 % solvent B, 10% solvent C and 10% solvent D for 11 minutes, followed by 70.6 % solvent A，1.4% solvent, and

28% solvent D for 7 minutes; and then 61.6% solvent A，1.4% solvent B and 37% solvent D for 12 minutes, followed by equilibration for 4 minutes. The flow rate and detection wavelength were set to 0.6 ml/mins and 450nm with injection volume of 3

|il for each sample.

(

The accuracy and precision of HPLC was determined by extraction efficiency and relative standard deviation (RSD), respectively. All of the extraction efficiency

32 fell within the range of 80-120% and the accuracy was close to the theoretical recovery. RSD in the six replicates was smaller than 7%, indicating the acceptable i precision in the experiments. Extraction efficiency was done by adding fixed amount of standard solution into the sample with known amount of compound corresponding to the standard solution. The extraction efficiency was calculated by the following equation:

Extraction efficiency (%) = (C—A) /BxlOO.

A: Known amount of sample

B: Fixed amount of standard

C: Total amount

3.10 Subcellular localization of FPSl and FPS2 by transient expression

In order to investigate the roles of the two FPS, determination of the subcellular localization of the two FPS is necessary. Because there is no direct evidence explicitly proves the cytosolic localization of the two FPS (Brechbiihler-Bader, 1968;

Hugueney et al., 1996; Flores-Sanchez et al, 2002; Ortiz-Gomez et al, 2006), I examined their subcellular localization via GFP transient expression. Full length

sequence of fpsl and fps2 genes were amplified by PGR (Table 2) using the

Advantage® HF 2 PGR Kit (Clontech). After agarose gel electrophoresis, the PGR

product was recovered using Gel-M Gel Extraction Miniprep System (Viogene). The

recovered fragment was ligated to pMD18-T Simple vector (Takara) and transformed

to E.coli DH5a competent cells by heat shock at 42°C. Transformed cells were

33 screened out and incubated at 37°C for 16 hours on LB plate containing 50 ^ig/ml kanamycin. Three colonies with the positive bands were selected for plasmid DNA extraction using Mini Plus™ Plasmid DNA Extraction System (Viogene) and sent for sequencing.

Plasmids with the right fps sequence and the pBI221 (Clontech) vector bearing green florescent protein (GFP) were double digested by the Xbal and Xhol (or Sail) enzymes (New England Biolabs). The fpsl and fps2 fragment was ligated to the GFP vector with the T4 ligase system (Promega) (Figure 14). The pBI221-FPSl: GFP and pBI221-FPS2: GFP constructs were transformed to the DH5a E. coli. Plasmids were extracted with Mini Plus™ Plasmid DNA Extraction System (Viogene), plasmids was validated again with enzyme digestion. And thus used for the transfection of the 5 days old Arabidopsis suspension cell culture. Confocal microscope were used to detect the fluoresence' of the GFP labeled proteins (Miao & Jiang, 2007).

1 Steps from preparation of protoplast to observation for fluorescent signals was performed by Mr. Wei Wang

34 Table 2 Primer pairs designed for FPSl-GFP construct and FPS2-RFP construct

Gene Locus Restriction Primer Sequence (5，to 3') Ta* Length

site (�C ) (bp)

FPSlAT5G47770 Xbal Forward TCTAGAATGAGTGTGAGTTGTTGTTUM

Xhol Reverse CTCGAGCTTCTGCCTCTTGTAGATC

FPS2 AT4G17190 Xbal Forward TCTAGAATGGCGGATCTGAAATCAA 53.3 1038

Sail Reverse GTCGACCTTCTGCCTCTTGTAG

*: annealing temperature

FPSl-GFP CaMV 35S promotor^^^^^^^^^^^^l NOS

FPS2-GFP CaMV 35S NOS

Figure 14 Fluorescent fusion constructs of the FPSs with Cauliflower Mosaic Virus promoter overproduce the FPS-GFP fusion protein

35 Chapter 4. Results

4.1 Screening offpsl and fps2 homozygous mutants

The homozygous mutant plants were screened out by PCR as described (Figure

15A-C). The wild type plants (WT) only have one PCR product (about 1000 base pair (bp)) because the absent t-DNA insertion (refer to page 27 Figure 12).

Homozygous mutant plants have no PCR product in the LP-RP reaction because the

LP-RP fragments with 4493bp t-DNA insertion is too large to be amplified by regular

PCR Taq polymerase (Figure 12). The homozygous mutant (HM) only has the

BP-RP PCR product, whereas the heterozygous mutants (HZ) have both BP-RP and

LP-RP PCR product (Figure 12). The homozygous mutants were identified by electrophoresis and the homozygosity of their offsprings were further confirmed

(Figure 15D).

The morphology of the Arabidopsis plants were compared (Figure 16). Most of the fps2 mutants (about 60%) have earlier flowering time and smaller leaves compared to the Col-0 wild type. The fpsl mutants have similar morphology as the wild type plant with slightly bigger leaves.

4.2 Validation of fps mutants by RT-PCR and enzyme activity assay

Alignment of fpsl and fpsl was done to confirm the position of fpsl specific

region and the high similarity of their structures (Figure 17) that specific primer

36 design for open reading frame is only available for fpsl. Reduced fpsl gene expression was confirmed (Figure 18). Due to the lack of unique sequence in fps2 gene for specific primer design, enzyme activity assay was used to verify the knockdown mutant of fps2 and the FPS enzyme activities in the fps2 mutant were lowered 2.5 fold than that in the wild type (Figure 19). This further validate the t-DNA insertion exert effect on the reduction of FPS enzyme activity in ips2 mutant.

A) ！EH

BP-RP jjjjjj^^

LP-RP jjjjuiiiiiijiQmmjmimiiimiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiniiiim

fpsl

BP—PtP jjjjjjjjjj^^ LP-I^P iQ2iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiimuiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii

C) ^ ) WT 1 2 3456789 10 BP-RP HHHHE3HHHHC9C9E9B9HHHI

D) Col-0 offsprings fps1 offsprings fps2 offsprings 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 BP-RP HHIHH DHBBH qqqqq LP-RP mmomQ mniiiiim mmniiiiiim

Figure 15 Gel photos of the PCR analysis for mutant screening. The PCR products of A) and B) the fpsl mutants; C) the fps2 mutants; and D) the offsprings, fpsl, SALK一073756 Arabidopsis; fps2, SALK_045864 Arabidopsis', WT/ Col-0, wild type Arabidopsis (ecotype Columbia-0); BP, SALK-LBblprimer; LP, fpsl/ fps2 left primer; RP,fpsl/fps2 right primer. The numbers above the lane denote different individual plants.

37 35 days old Arabidopsis CoFO f^ fps2 _ _ ,__•-〜• - 、..；— — —.- +..,,:+.- — .‘." “― - ： H

��•�//•, .：：. . . ； ：i

；•：零^ / •::: 缚 I

35 days old Arabidopsis Core f^ f^

Figure 16 Morphology of wild type Arabidopsis plants and fps mutants

38 fpalCDS.aef PS2,1, seq IBMHHIIlHHHHBHHiiHHBB 切0 fpalCDS.seFPS2.1. »eqq MSSBStf^^^^^BBK^B^^^BS&IBB^BBflSCBBHB&BBnBM B 訂令 ConafttiauA Coti»«Mus eg ggcc»gacg«c g«ctcg«cc«cc«ccct:tg« ofl»ff fp5lCD3.Be(i •••••{{^••••••••Hi ao fpslCDS.»«j FFS2.1.««q 0 rPS2.1.5eq BBii^HlwIEHrBCI^HBvM Con«*fisu« CooAttoauj am亀 c 觀象fftACtc ctg 亀教ccc巍 eg eg £p

Consfttuiuji 9 c g«ccc a* tCMCCCC cce •cgccz Consensus cccgctgcccg gc tcflct acggc go fpjilCDS.Acq 200 fp»lCDS.»«q Conaenaua a cccgcccccaagtccg* ct eccca ga ceccecce Con««n»u* aaaaccaca tgi^tfftoaAO毳 tgttcccactO»cAtflflg fpaicDs.aea A^^Bili^^^BTcK^Hni 240 fp»lCD3.#e

Conaenstui ogccggcgcACcaMcgocc caagcccacccccccgrcoc Cons«n»xx« tacggtaaa ccfl4 ceace gcga* g

CpalCDS.stq 440 fpslCDS.aeq BBvwsHc^HBIB^^^Bv 314 FP32.i.，eq ConAQnsua ct^gacga ac acgg* uccccgccAc cgccQtgg ca Conaensu» ccrctacAAa««ffct g秦tec a*a00A9 etc應tflgA

fpslCD3.*eq 460 fpalCDS.seq

Consfttisufl ccccff cflgcc aaa cc aggccagrcacgr ccgcc Coaaetiaua cazqaqa aa^gcca gagaagccgac gat BAi^^^^^Bnil^Bc^^HH 520 ConfiftnauA at aa gatgr90ACt:ccacctcgc«acc« acccac廳cro象 Consensua aft g cacca •gc急Mgc孰巍t ca^gcagtacc^Aaac

mmillllimcfiH^l^millllllH seo £p，icDS.，eq mBHMBiHH^I^IBHI ” 4 NI^^H^^B^^^^I^^H Constta«u« cccccuMftgca rcc 9 ga * gcctCMccmtgztga Caaa^nsua c fpsicDs.flftq jHTBOBQI^B^^Bi^^Bi^llB^BbBWBllBBHMBBBi^mJMBWBHI 604740 ConsensuA ccc gccgacccgcttAa qao^c gagcc caaacagcc Figure 17 The alignment of fpsl and fps2 gene sequences. The identical base-pairs are shaded in dark blue shaded while mismatched base-pairs are shaded in light blue. A fpsl specific region is located at the beginning lOObp of the fpsl sequence.

Whole plant cDNA Col-0 fps1 fps2

ORF

FPS2 ORF

Figure 18 Gel photos showing the result of PCR verification of办si mutant No PCR product is detected in the fpsl mutant because of the 4493bp t-DNA insertion in the eion 2 of the mRNA making the amplification by norma里 Taq polymerase impossible. CoM)’ Wild type Arabidopsis (Columbia-0 ecotype); fpsl, SALK一073756 Arabidopsis; Jps2, SALK一045864 Arabidopsis; fpsl Qviijpsl open reading frame PCR fragment;加2 orf: fps2 open reading frame PCR fragment

39 Farnesyl Pyrophosphate Synthase (FPS) activity in fps mutants and Col-0 wild type 0.35 C.3S^205 -—— - « e F 35 •o 0.3 k jn f. r JT J m fc * TJ 驾 r ^ •i 0-25 -4 K1 > ；0.2 gi'-J 'iZ M b r n t5 E k* A "= A bpps i/> t 0.15 C； ^'J -..." g. I r 3 0.120755 f J HHHjjji m "S 0.05 — 0=03663 !i JTM — — • — •— 1 Col-O fpsl .. fps2 Figure 19 Farnesyl diphosphate synthase (FPS) activity in the fps mutants and Col-0 wild type. The activity of FPS showed a 10 fold decrease in the fpsl mutant and 2.5 fold decrease in the fps2 mutant

4.3 Genechip analysis of two fps mutants

4.3.1 Quality control and normalization of microarrav sample

The quality of extracted total RNA was examined using three methods: gel

electrophoresis, optical density (OD) measurement, and RNA degradation plot. The

five clear bands in RNA electrophoresis gel indicate the high purity of RNA (Figure

20). The OD 260/280 ratio of six RNA samples was close to two, suggesting a high

percentage of ribonucleotide. The RNA degradation plot showed that the slopes of

the plot for the six chips are similar, indicating no degradation or manipulation error

in the genechip (Figure 21).

40 Figure 20 Gel photos showing total RNA extracted from leaves of wild type Columbia-0 and fpsl, fps2\ for the extraction and B) 2"** extraction. The high quality of RNA was confirmed by cleaiHSr bands with the large rRNHA subunit (the firs Bt band from th8e top) show s about two times greater intensity than the small subunit (the second band from the top).

RNA degradation plot

o - I 1 1 1 1 1 0 2 4 6 8 10 5' <——> 3' Probe Number

Figure 21 RNA degradation plot for the six genechips showing high quality of RNA by similar slopes of six lines.

41 4.3.2 Normalization and identification of differentially expressed genes

Affymetrix genechip data were normalized using RMA method (Irizarry et al.，

2003). preRMA boxplot (Figure 22A) and RMA normalized boxplot (Figure 22B) showed the distribution of normalized data. By comparison among different groups, top tables of differentially expressed candidate genes were generated and the differentially expressed candidate genes were screened out by the following filters: 1)

Log Fold change > 1.0; 2) p-values < 0.05%.

NormaCzed Data

i ！ j i i i I 1 j 1 i I

; ! ; ; i j J!- i j i j j 1

I • I I I t i I I ) ! I ！! I • - .1..« . ! ~I^ II I J i 'J !! II I —»— \ I - ; { : ; J • iiiiiHKA1CEL HKBI CEL HKCI iCEL iiiiiHKAI CB. HKBI CEL HKCiI CEL Figure 22 Boxplot of the six genechips showing the normalization of the data distribution of each dataset A) PreRMA boxplot have different value of medium and quartile，while B) RMA normalized boxplot excluding those outliers, making the data distribution of each dataset become comparable. HKA: CoW), HKB: fpsl, HKC: fpsl

42 4.3.3 GO annotation of differentially expressed genes in fps mutants

Annotations of genes were obtained from The Arabidopsis Information

Resource (TAIR) database. The differentially expressed (DE) genes were grouped under the GO category of molecular function (Figure 23) and biological process

(Figure 24). In total, 27 genes were up-regulated and 21 genes were down-regulated in the fpsl mutant. In the fps2 mutants, 128 up-regulated genes and 58 down-regulated genes were identified.

Genes involved in enzyme activities and transport functions were the major categories. Apart from the unknown gene (21%), transferase activity account 28% of the 27 upregulated genes in fpsl mutant (Figure 23A). The top ranked molecular functions in fps2 mutant were enzyme activity (22%), unknown molecular function

(14%), other bindings (12%)，transferase activity (11%) (Figure 23C). For both mutants, "enzymes activities" were the major category with 15% and 22% genes fell into this category in the fpsl and fps2 mutants respectively.

For the 21 down-regulated genes in fpsl mutant, the top-ranked GO categories

are: unknown molecular function (34%), transporter activity (13%), other enzyme

activity (13%) and hydrolase activity (8%) (Figure 23B). For the 58 down-regulated

genes in fps2 mutant, the top categories are: unknown molecular function (24%),

other enzyme activity (20%), other bindings (18%) and transferase activity (9%)

(Figure 23D). Enzyme activity remained as the top ranked category in the two

mutants.

43 When grouped under the GO category of biological process, differentially expressed genes were mainly involved in stress response. Out of 48 differentially expressed genes in fpsl mutant, other biological processes (21%), other cellular processes (20%), unknown biological processes (13%), response to abiotic or biotic stimulus (11%) and stress responses were the top-ranking categories (Figure 24A). In fps2 mutant, the differentially expressed 186 genes belong to GO categories of other metabolic processes (18%), other cellular processes (17%), stress responses (14%), unknown biological processes (12%), other biological processes (11%), and response to abiotic or biotic stimulus (10%) (Figure 24B).

44 A) Molecular functions of upregulated genes in fpsl mutant

/ \ atrsnsfersM activky / •unknown molaaitvfunctions / : oth«r •nzymt KtMtv / FJIHk •�™ /�.'.-- - •Wntf.activrty / 讯 •profinbindini fl^H^H^^^nl^^ • othftr molccUar functions .transcriptionf^oractWirv �^•ONAorRKAbimlini y^Pffij^ ^^ Bnucleotid^bindint »transporttr activity 漏 r鲁 cttptorbindingoractivlty • stnictum motecUftKtMty

B) Molecular functions of downregulated genes in fpsl mutant

: - unknown mol^Urfunrtons

I雅 • transfers activity •transcription Htctor actMty

r* ？ Vjg^^^^ ^^^B^^Sj^^H^S •DNAorRKAbifidint

V jMBSMF ^^^^^^^^^^^^ m oth«r molccuisr functions ^^^^^^^^^ “ r«c*ptorbindingor activity ^^^Hk •structural molKul*activity

45 C) Molecularfunctions of upregulated genes in fps2 mutant

y/^ ' y、 • oth«r•nzym«actMcy

/ ** • unknown moltoiter functions

I •tr»nsf«ras« activity

/ : Bothermolcciisrfurctions

S;. •transport.factivity • hydralM acti^ty • transcription factor •ctivity 攝 JJB BDNAorRNAtainding

\ ^^^^^^^^^^ bineSinf \ ^^HWli ."ucLiccidbindne

VV^y^ • structural moteculcxtwtty

D) Molecularfunctions of downregulated genes in fps2 mutant

Z unknown molecular functions / •othftr.niymtftctMty

j transfsrastt activfty

/ othtr mol«cUar functtom

•rvydrolMSCtWity • transcription fMor activity a kinm activity

\ •nucl«ic»cid bindns

\ ^^^^^^^^^^^^ • transporter activity

Nv ^^^^^^^^^ VrftCftptorbindinc or Activity

^Vs^ • struaural mol«cii*actMty

Figure 23 Pie charts showing the distribution of the differential expressed genes grouped according to their GO molecular functions for A) upregulated genes in the fpsl mutant; B) downregulated genes in the fpsl mutant; C) upregulated genes in the fpsl mutant; and D) downregulated genes in the fpsl mutant

46 Biological process of the differentiatty expressed gene in fpsl mutant A) IX

flB^^^^^^PHr^. •otn«rm«tadoRcprocissts / * \ • oth«r cfttkilar procttius / mttRknowft biotOficslprocMMS / • rv^nst to abiotic or bbltcscirnuhis rciponsft to stress

BgthtrbioloiicslpfOGVSsts •trviscription ，料：二一

"I ？• cftU organizKion and bbgin«siK ^^^^^^^^^^^^^ « •l^ctron trmport or »n»rgy pathMrtys m DNA or RNA mstaboKsm

B) Biological process of the differentiatty expressed gene in fps2 mutant

1%

S^^I^H^^^Hj^^^ • other metaboKc processes •，• other cellulsr processes / • response to stress / 4X 叙 unknown biological processes j ^^^^^^^^^^^^^^^^^^^^^ • other bk>l(^al processes / response abiotic or bioOc stimulus • transport ^ • ceU organization and biogenesis etectron transport or energy pathw^ \ Ig-^j® • ProtemmetBboiIsm

V- : 践鄉釣 ^ .c^—nta, proems \ 、‘ / • DNA or RNA metabolism \ .”..議豪顯 ：海•：考

Figure 24 Pie charts showing the distribution of the difTerential expressed genes grouped according to their GO biological processes for A) the fpsl mutant; and B) the Jps2 mutant

47 4.3.4 Genes participate in stress and defense response were differentially

expressed in both fpsl and fpsl mutants

The commonly modulated genes in the two mutants are listed in table 3. Most of them are involved in stress and defense response. Both mutants have nearly two folds increase in glutaredoxin GRX480; thiol-disulfide exchange intermediate expression

(Table 9). GRX480 is a member of the glutaredoxin family and the key factor for the crosstalk of salicylic acid / jasmonic acid pathway (Ndamukong et al., 2007;

Koomneef & Pieterse, 2008). These pathways are important components of stress response in plants. Decreased levels of monothiol glutaredoxin family (GRXS) were found in the fps2 mutant (Table 4). Particularly, GRXCll (AT3G62950), GRXS6

(AT3G62930)，GRXS8 (AT4G15660), GRXS4 (AT4G15680), GRXS3 (AT4G15700),

GRXS5 (AT4G15690) and GRXSl (ATIG03020) were down regulated 1.4 to 1.7 fold in the fps2 mutant. Interestingly, the transcript of another glutaredoxin increased for nearly two folds.

As the counterpart of glutaredoxin, thioredoxin gene family genes were changed as well (Table 5). The thioredoxin-dependent peroxidase 2 (TPX2) (AT1G65970) was up-regulated by 2.4 folds, whereas three H-type thioredoxins including ATH7

(AT1G69880), TRX5 (AT1G45145), and a protein similar to TRX2 (AT3G53220), were up-regulated for 1.6，1.2 and 0.5 fold respectively.

48 Table 3 Genes that were regulated in both fpsl and fps2 mutants

Locus Log Fold Change Annotation Biological Process &

fpsl fps2 / Molecular Function

AT1G28480 1.883537 1.631759 GRX480; thiol-disulfide exchange intermediate electron transport, jas

AT4G11650 3.182016 6.078548 OSMOTIN 34 defense response to b

AT1G73260 2.313541 4.891249 trypsin and protease inhibitor family protein Unknown/ endopeptic

AT3G01420 1.406242 4.147479 ALPHA-DOXl cell death, response tc

acid stimulus/ lipoxyj

AT3G55970 2.178525 3.494942 20G-Fe(II) oxygenase family protein NA / iron ion binding

AT2G39030 2.350627 2.41775 GNAT family protein metabolic process/ N-

AT3G49620 2.228318 2.341394 DIN 11 (DARK INDUCIBLE 11) aging, cellular starvati

AT2G19970 1.703106 1.957941 pathogenesis-related protein Unknown/ unknown

AT5G05600 1.667361 1.582335 20G-Fe(II) oxygenase family protein NA / iron ion binding.

AT1G13650 -1.81366 -1.52464 similar to 18S pre-ribosomal assembly protein Unknown/ unknown gar2-related

ATIG59930 -1.14934 -1.7057 similar to unknown protein (TAIR:AT1G59920.1) Unknown/ unknown

叙:NA : not available in database

49 Table 4 Differentially regulated GRX in fps mutants

Locus Annotation fpsl :WT logFC P.Value fps2 :WT logFC P.Value

AT1G77370 GRXC3 in^ iii^ 0.576432 0.032684

AT1G28480 GRXC9 1.883537 0.002243 1.631759 0.004173

AT3G62950 GRXCll insig. insig. -1.69171 0.000371

AT1G03020 GRXSl insig. insig. -1.37586 0.000405

AT5G18600 GRXS2 insig. insig. -0.72921 0.041011

AT4G15700 GRXS3 insig. insig. -1.48577 0.009872

AT4G15680 GRXS4 insig. insig. -1.53317 0.008087

AT4G15690 GRXS5 insig. insig. -1.44184 0.00468

AT3G62930 GRXS6 insig. insig. -1.64846 0.009834

AT4G15660 GRXS8 insig. insig. -1.57553 0.011503 insig. : insignificant

Table 5 Differentially expressed TRX in fps mutants

Locus Annotation fpsl :WT logFC P.Value fi}s2 :WT logFCP.Value

AT5G42980 TRX3 hii^ iiiiij -0.42917 0.0491

AT1G45145 TRX5 insig. insig. 1.207763 0.02081

AT3G53220 similar to TRX2 0.575547 0.00635 insig. insig.

AT1G76080 similar to TRX4 -0.34768 0.03462 insig. insig.

AT1G69880 similar to ATH7 insig. insig. 1.630881 0.01303

AT5G16400 TRXF2 insig. insig. -0.63396 0.02688

AT1G65970 TPX2 insig. insig. 2.406465 0.03384

insig.: insignificant

50 4.3.5 Genes in the plastidial pathway were down-regulated

A list of terpenoid pathway genes was examined for their expression pattern

(Figure 25) (Table 6). Most of the genes involved in MEP pathway and its downstream genes were only slightly changed. These genes include: IPP2, GGPSl and PSYl leading to carotenoid biosynthesis. Genes in the cytosolic MVA pathway did not show significant changes, except for SQPl (Table 6).

•Sip•“ •灣譯 •」」』'.丨：丨！狐円 SSSS

•hih^^hP^HH mmiiQ^j^llpiiin

mA^f^ at 一 ^SA^^H

•HESfiSlHI BBBlBSWBBt^EBB atg4 ASfTii-H ^mmSrn,^ ^ygB^BHBSm^lB™ BIBKBIhihsbs BHHSPJPPHH •BiiBiB •Ka"丨欄 HpM“+5i jHii^slpiii

^^ 33 HB&BiiyiiSlS^uHil HZj^ESSEEI^ffi^ • I lliil 丨丨II laflWl irnii • WBBpiBI •MaaMMP|jjjjjl||p

•TnilTlllfW ^^丨丨；丨：疆^ 圓fjl^ll

Sb^SSiS^SS mmsm

Figure 25 Schematic diagram of the terpenoid biosynthesis pathway in Arabidopsis thaliana showing the genes (black words) and metabolites (boxes) involved in MVA (orange boxes) and MEP (green boxes) routes and their downstream pathways (blue boxes).

51 Table 6 Summary of Log2 fold change in expression level of terpenoid pathway genes

Locus Annotation fpsl :WT logi AT4G15560 , -0.35125 AT5G62790 DXR insig. AT5G60600 ISPG insig. AT3G02780 IPP2 insig. AT4G36810 GGPSl -0.47369 AT3G14510 similar to GGPS3 insig. AT5G24150 SQP1,1 -1.23187 AT5G24160 SQP1,2 insig. AT4G37760 similar to XFl (squalene epoxidase 1) -0.59065 AT1G74010 similar to SS2 (strictosidine synthase 2) insig. AT1G74020 Strictosidine synthase 1 precursor (SSI) insig. AT3G51450 similar to YLS2 (yellow-leaf-specific gene 2), strictosidine synthase 1.111148 AT2G41290 similar to strictosidine synthase family protein -0.75399 AT5G17230 Phytoene synthase, chloroplast precursor (PSYl) -0.63315 AT1G20330 SMT2 -0.315431 insig.: insignificant LogFC: log fold change

52 4.4 Effects of FPS mutations on pathway enzymes

Sesquiterpene cyclase activity increased 2.6 and 2.4 folds in fpsl and fps2 mutant respectively (Figure 26). Squalene synthase showed no significant activity change in the mutants (Figure 27). One of the dominant sesquiterpenes in

Arabidopsis is the defense plant volatile beta-caryophyllene. I hypothesize that

inhibition of FPS cause a stress like condition, which promotes the increased

sesquiterpene cyclase to facilitate the production of beta-caryophyllene as part of the

stress response.

Sesquiterpene cyclase activity in fps mutants and Col-0 wild type

• 1.4 1 1.315597 I ••• 1.2315

！r::=t=i= i"? 0.8 ^B • • S^ Qg ^^m •SesCyc

I卜 J—一―I I

‘i 二 广―—•—-——• •

Col-O fpsl fp$2 Figure 26 Sesquiterpene cyclase activity in the fps mutants and Col-0 wild type. The sesquiterpene activity showed a 2.6 fold increase in the fpsl mutant and 2.4 folds increase in the fps2 mutant

53 Squalene Synthase (SQS) activity in fps mutants and Col-0 wild type

每 IS 17.62198 1 H D~

“：滑一-f：^^

11"= y ==置 I—-

s 4 i ^

12——id • •—— Q —..... WBsmMSm. — hHHHII flHHHHI Col-0 fpsl fps2

Figure 27 Squalene synthase (SQS) activity in the fps mutants and Col-0 wild type. There is no significant change in the SQS activity among the 2 fps mutants and Col-0 wild type plant.

54 4.5 Effects of fps mutants on terpenoids and sterol metabolism

Nerolidol was used as the internal control in each sample (Figure 28). Triterpene and sterol profiles in the fps mutants were found similar to that in the wild type, except for the sulforaphane nitrile, which is related to the induction of thioredoxin reductase, was decreased in the fps mutants (Figure 29).

4.6 Comparison on carotenoids and chlorophyll contents

Compared to wild type, higher chlorophylls (both a and b) and p-carotene content were detected in fpsl mutant. Decreased lutein, chlorophyll b and a and slightly increased chlorophyll a were observed in fpsl mutant (Figure 30 and 31).

55 95

• ...."",，...... • . . . UU.bC UU.Ud UUbl UUUl UUb v-u,,

Zl I s ” 0000001

M Olf

'''' '0000091

ji 0000002 8

9 OOOOOK

ooraotK

ooooosc

oooooov

•000009，

.0000009 e (a o pouvpmq^ I . •..…•沾• . "丨 4 ...山. •.叫丨. •.叫I 比... .UU.b6 .UU.UC . UUbt UUUl UUl. V•^上 '"""•^Tni^irf p『“i| V '21. oi S p ooooof 9 II 000009 z Q 000008

•0000001

丨J ^ -OOOOOZL H -ooooon

•0000091

•0000081

•0000002

•0000022

•OOOOOfrZ

•0000092

•OOOOOK

OOOOOZE (v ^ laouepunqv ^bundancel 04 C) ’7000000

6500000 3

6000000

5500000' 7

4500000 8 4000000 5

3500000

3000000

2500000 11 2000000 6 10

1500000. 2 4 丨 12

1000000

500000 1 3 L.. .i_.. ,.|i,..L,…tiUJM^^^^M^!^- Timo-A mm 70 nn mnn innn snnn Knnn

Figure 28 GC-MS profile of A) Col-0 wild type, B) fpsl mutants and C) fps2 mutants. l.SuIforaphane nitrile, 2.Nerolidol, 3.cis-pinene, 4.n-Hexadecanoic acid, S.Phyto, 6.a-Linolenic acid, 7.2-palmitoylgIycerol, 8. Methyl linolenate, 9.0ctadecanoic acid, 2-hydroxy-l-(hydroxymethyl)ethyl ester, 10.Tetratetracontane, 11.vitamin E, 12.campesterol, 13.stigmasterol, 14.p-sitosterol

57 Relative corrected peak areas of fps mutants and Col

9

8

8 « - 18 ® 16 a01 . T T 1 2 * , § 01 UId J tc

*/ / // ^///// z

Peaks

Figure 29 Relative corrected peak areas of the fps mutants and CoI-0 wild type with Nerolidol as identified compounds do not show significant changes (95% confidence by t-test) except the suli asterisk) in the mutants.

58 A) "H 1 ‘

o.ie-

0.16- 0.14-

0.12-

？ 0.10-

0.08- 2 3 0.08^ 0.04- I 4

iljo. '4:00 '6.00 aoo" la'oo 12.00 u'oo ie'oo iaoo . 20.00 • 22.00 .24'oo . x.'oo . k'oo . ao'oo . izoo . moo . km . xoo . 4aoo _ 4200 Mhtles

B) i 1 “ 3

020-

3 015r 112 � i°" H

000- L 1 ^^ LA /V A . \

2.00' '4.00' 6.00' 800' io'oo i2'od ‘ iVoo ‘ is'oo ‘ ie w ‘ io'oo ‘ 22'od ‘ w'oo ‘ zs'oo “ k'oo ‘ ao.'oo ‘ a'oo ‘ w'oo ‘ ie'oo ‘ SB'oo ‘ io'oo ‘ 42'oo Mnuies

1.20-j c) 1 1.00- 3 O.BO-

严 2

0.40- 4

L^^A. I 丨 'i00' '4.00' Voo '8.00' io'oo i2'o6 ‘ i4.'00 ‘ i8.'o6 ‘ is.'oo ‘ k'oo ‘ 22'o6 ‘ 24.'oo ‘ x'oo ‘ k'oo ‘ X.'oo ‘ k'oo ‘ MM ‘ 38.'00 ‘ 38'00 ‘ itt'oo ‘ «'oo Mnutes

Figure 30 HPLC profile of A) Col-0 wild type, B)fpsl mutant and C) fps2 mutant Arabidopsis at 450nm. 1. lutein, 2. chlorophyll b，3. chlorophyll a, 4.p-carotene

59 Carotenoids and chlorophyll content in fps mutants and Col-0 wild type

18000 -|

16000 ^

14000 I .000 • H___ § 10000 — ^H 者 8000 BI^^H M^H •附 I • • _ ••

:—一 mp'-m •- lutein chorophyll b chlorophyll a beta- carotene

Compounds Figure 31 Cartenoids content in the fps mutants and Col-0 wild type plant All of the changes in carotenoids content among the mutants and wild type plant are significant (95% confidence by t-test). The lutein contents in both mutants are decreased while chlorophyll a and b, beta-carotene contents showed dramatic increases in the fpsl, only chlorophyll a increased in the fps2.

GO 4.7 Subcellular localization of FPSl and FPS2

Fluorescence of both fps GFP fusion proteins was detected in cytosol, with certain amount of diffusion into the nucleus (Figure 32). This result provides some direct evidence that both FPSl and FPS2 are localized in the cytosol.

Figure 32 Fluorescent microscopic images of A) FPSl-GFP; and C) FPSl-GFP visualized using a confocal laser-scanning microscope and phase-contrast microscopic images of the cell expressing B) FPSl-GFP; and D) FPSl-GFP

61 Chapter 5. Discussion

Cytosol t • Plasttd ‘ 1 S^utants (MVA pathway) . ^ (MEPpathHd^) ；

s邮'曲丨 p Acetyl-CoA 广 v ^APmuvate i FPSactivity^l iw^^J^L-™., / \ „ Z V^^TE= / \ y .onalecp^'^^*?^^™ SJL se^ne 众里 。•；

^^^ , stiictosidine j H^^^H

Salicyciicactd ； W^^BK/^^^ jBW^ 丨 r with GRX increas^N卜Kilj^BBMUHH^—

=1 备 \ G^^ TRXinl^ PDF1.2 gefie ^^^^ ^ ^ X . sulforaphane nitnle4, PmHHPI^ sulforaphane

Figure 33 Summary of the changes in gene expression level, enzyme activities and metabolites of the two fps mutants in the present study. The genes with an arrow next to it are differentially expressed in at least one of the mutants and the direction denote increase (upward) and decrease (downward) of the expression level. The color of gene name indicates it is differentially expressed in fpsl (red),j5?s2 (blue) or both mutants (purple).

This study aims at understanding the roles of two FPSs in the terpenoid biosynthesis. There is no previous report on Arabidopsis FPS mutants using comprehensive transcriptome profiling approach. Results of the present study suggested additional crosstalks between MVA and MEP routes, and unravel possible regulatory mechanisms in terpenoid biosynthesis.

62 Previous studies showed that when enzymes situated at the upstream of IPP were inhibited, vigorous metabolic profile changes were resulted while transcripts of most terpenoids pathway genes remain unchanged (Laule et al., 2003). In the present

study, similar phenomena were observed as well. Most of the terpenoids and sterol

pathway genes remain unchanged or only showed slight changes. The transcripts of

SQP, an enzyme important for sterol biosynthesis, were altered in both mutants.

However, the sterol contents in both mutants were similar to that in the wild type (95

% confidence proved by t-test). This suggests a lack of direct correlation between the

transcripts and metabolites in terpenoid biosynthesis. Our finding further confirmed

that post transcriptional regulation may play important roles in controlling metabolic

flux in terpenoid pathway.

FPS2 isoform was originally treated as a redundant enzyme for the FPSl, as

both of them have the same function in vivo (Cunillera et al., 1996). However, the

differential expression pattern of the two FPSs, with fpsl being expressed throughout

the plant but fps2 is preferentially expressed in the inflorescence (Cunillera et al.,

2000)，leads to suspect that fpsl and fps2 may have different roles in the terpenoid

metabolism. The present comprehensive survey of the transcriptome revealed that

more genes were modulated in the fps2 mutant compared to the fpsl mutant (186

genes in fps2 vs 48 genes in fpsl) and sqpl gene expression level increases in fps2

while decreases in fpsl. These all indicate the distinctive roles of these two FPS

genes. Based on their expression patterns, fpsl is suggested to be involved in the

primary metabolism and sterol biosynthesis while fps2 is suspected to be participated

in the secondary metabolism for plant defense during flowering Cunillera et al.,

63 2000). Yet the FPS IS overexpressor study suggested that FPS IS is not a limiting factor for sterol biosynthesis (Masferrer et al., 2002). The relationship between fpsl and fps2 remains controversial.

The crosstalk between the cytosolic and plastidial compartments was suggested previously (Laule et al, 2003). Inhibition of the HMGR in the MVA pathway resulted in significantly changed metabolites resulted from the MEP pathway. In the current stuffy, perturbation of the MVA pathway downstream gene fps led to changes in gene transcripts and metabolites in the plastidial pathway. This

provides clear evidence that crosstalk exists between the cytosol and plastid during

terpenoid biosynthesis.

This study provides the first evidence that the crosstalk might occur at points

besides IPP as well. Previous studies suggest that the common precursor IPP shared

by both MVA and MEP pathways is the point of crosstalk (Laule et al., 2003). In the

present study, perturbation of the MVA pathway was made at a point downstream of

IPP and metabolites produced from FPS remained unchanged. Interestingly, the MEP

pathway genes and the downstream metabolites were altered. This indicates that EPP

is not the only point of crosstalk between the two compartments.

The transcript of glutaredoxin GRX480, a key regulator in the crosstalk of

salicylic acid (SA) and jasmonic acid (JA) pathway (Ndamukong et al., 2007)，was

found to be increased by two folds in both of the fpsl and fps2 mutants. While the

transcripts of seven other glutaredoxin family genes (GRXCll, GRXS6, GRXS8,

GRXS4, GRXS3, GRXS5 and GRXSl) were reduced for 1.4 to 1.7 folds in the fps2

64 mutant. This result suggested that fps2 is more closely related to the glutaredoxin

system, and stronger cellular oxidative stress may have occurred in the fps2 mutant.

Although the actual roles of the glutaredoxin and thioredoxin are not clearly

demonstrated, it is known that both are reducing systems involved in oxidative stress

response (Femandes & Holmgren, 2004). In the fps2 mutant, seven glutaredoxin

genes were down-regulated while three thioredoxin genes (the antioxidant TPX2

(Thioredoxin-dependent peroxidase 2) and two h-type thioredoxin (thioredoxin

H-type 8 and thioredoxin H-type 5) were up-regulated. Thioredoxin is recognized as

the dominant reducing system in oxidative stress response and glutaredoxin as a

supplement (Rouhier et al., 2004). From the observation in the fps mutants, it appears

that the thioredoxin system may be activated to reduce the stresses caused by fps

gene knockdown to maintain the normal cellular processes. Therefore, the

compensating glutaredoxin might become less active.

Similar to other Brassicaceae members, Arabidopsis thaliana possesses

glucosinolates, which is thought to be released from the vacuole in defense condition

and will be hydrolyzed by myrosinase to form sulforaphane and sulforaphane nitrile

(Bradbume & Mithen，2000; Matusheski & Jeffery，2001; Wittstock & Halkier，2002;

Zhao et al, 2008). The number of reduced thioredoxin was controlled by the inducer

of the thioredoxin reductase, sulforaphane (Hasten et al., 2002). The results of

present study show that the sulforaphane nitrile level, which can reflect the level of

sulforaphane, was reduced in the fps2 mutant. Since all experimental procedures are

carrying out under cold ambient temperature in organic solvent while all samples

were treated in the same way, it was unlikely to have hydrolysis reaction or

65 statistically significant reduction of the compound during experimental procedures.

Thus, it could be hypothesized that the decreased level of sulforaphane may serve to avoid the over-reduced environment in the cell.

The independent behaviors of carotenoids biosynthesis and the plastidial pathway genes are possibly due to the post-transcriptional control as well. With the

MEP pathway genes transcripts slightly decreased (Figure 31), it is interesting that dramatically decreased lutein and increased beta-carotene and chlorophylls content were observed. At the point of lycopene, the pathway separates into two branches, leading to lutein and beta-carotene production respectively. A decreased expression level of the upstream genes of the phytyl pyrophosphate, which contributes to part of the skeleton of chlorophylls, is observed. Therefore, it would be expected that the level of chlorophyll would be decreased accordingly. It could, therefore, presume that negative feedback mechanism may exist to overproduce chlorophyll (Terry &

Kendrick, 1999; Eckardt, 2009) in the fps mutants. The tetrapyrrole biosynthesis pathway (upstream of the chlorophyllide a) is possibly responsible for the production of chlorophylls and increased chlorophyll level in the fps mutants in spite of reducing expression of the upstream genes of the phytyl pyrophosphate.

66 Chapter 6. Conclusion

Using integrated transcriptomics and targeted metabolite analyses on

Arabidopsis fps knockdown mutants, the present study demonstrates that the fpsl and fps2 have distinctive roles in modulating terpenoid biosynthesis. More genes were

differentially expressed in the fps2 mutant than that in the fpsl mutant. Interestingly,

the most affected genes in the two mutants belong to the glutaredoxin and

thioredoxin family, which are key reducing systems in response to oxidative stress.

This suggests that perturbation of fps gene function may result in stress condition in

plants, which triggers the glutaredoxin and thioredoxin systems. The fact that no

direct correlation between gene transcripts and metabolites was found in terpenoid

biosynthesis suggests the importance of post-transcriptional controls on the terpenoid

biosynthesis. The present study provides new insights into the regulatory

mechanisms of terpenoid metabolism and generates new hypothesis for further

investigation.

Furthermore, the present study demonstrates that perturbation of cytosolic MVA

pathway gene, fps, at the downstream of IPP caused significant changes in

metabolites produced from the plastidial MEP pathway. This indicates crosstalks

between the two compartments may exist at multiple points apart from the previously

suggested IPP. Subsequent studies are prompted to further elucidate other crosstalk

points, which may open new avenue for efficient terpenoid metabolic engineering.

67 Reference

Aharoni, A., Giri, A.P., Deuerlein, S.，Griepink, R, de Kogel, WJ.，Verstappen, F.W.,

Verhoeven, H.A., Jongsma, M.A., Schwab, W., & Bouwmeester，H.J. 2003.

Terpenoid metabolism in wild-type and transgenic Arabidopsis plants. Plant

Cell, 15(12), 2866-2884.

Baldwin, I.T.，Halitschke, R., Paschold, A., von Dahl, C.C.，& Preston，C.A. 2006.

Volatile signaling in plant-plant interactions: "talking trees" in the genomics

era. Science, 311(5762), 812-815.

Hasten, G.P.，Bao，Y., & Williamson, G 2002. Sulforaphane and its glutathione

conjugate but not sulforaphane nitrile induce UDP-glucuronosyl transferase

(UGTlAl) and glutathione transferase (GSTAl) in cultured cells.

Carcinogenesis, 23(8)，1399-1404.

Baxter, C.J.，Redestig, H.，Schauer, N.，Repsilber, D., Patil, K.R., Nielsen, J.，Selbig,

J., Liu, J., Femie, A.R., & Sweetlove, L.J. 2007. The metabolic response of

heterotrophic Arabidopsis cells to oxidative stress. Plant Physiol, 143(1)，

312-325.

Berlin，N., & Staudt，M. 1996. Effect of water stress on monoterpene emissions from

young potted Holm oak (Quercus ilex L.) trees. Oecologia, 107，456-462.

Bouwmeester, H.J. 2006. Engineering the essence of plants. Nat Biotechnol, 24(11)，

1359-1361.

Boyes，D.C., Zayed, A.M., Ascenzi, R.，McCaskill, A.J.，Hoffinan, N.E., Davis, K.R.,

& Gorlach, J. 2001. Growth stage-based phenotypic analysis of Arabidopsis: a

model for high throughput functional genomics in plants. Plant Cell, 13(7)，

1499-1510.

68 Bradbume, R.P., & Mithen，R. 2000. Glucosinolate genetics and the attraction of the

aphid parasitoid Diaeretiella rapae to Brassica. Proc Biol Sci, 267(1438)，

89-95.

Brechbuhler-Bader, S., Coscia, C. J., Loew, P., v. Sczepanski, C. & Arigoni，D. 1968.

The chemistry and biosynthesis of loganin. J. Chem. Soc. Chem. Commun, (3)，

136-137.

Buchanan, B.B.，Gruissem, W., & Jones, R.L. 2000. Biochemistry & Molecular

Biology of Plants. Rockville，Md: American Society of Plant Physiologists.

1251-1268 pp.

Chang, M.C., Eachus, R.A., Trieu, W., Ro, D.K., & ICeasling，J.D. 2007. Engineering

Escherichia coli for production of functionalized terpenoids using plant P450s.

Nat Chem Biol, 3(5), 274-277.

Chappell, J. 1995a. The Biochemistry and Molecular Biology of Isoprenoid

Metabolism. Plant Physiol, 107(1), 1-6.

Chappell, J. 1995b. Biochemistry and molecular biology of the isoprenoid

biosynthetic pathway in plants. Annu Rev Plant Physiol Plant Mol Biol, 46，

521-547.

Cunillera, N., Arro, M.，Delourme, D.，Karst, R, Boronat, A., & Ferrer，A. 1996.

Arabidopsis thaliana contains two differentially expressed

famesyl-diphosphate synthase genes. J Biol Chem, 271(13), 7774-7780.

Cunillera, N.，Boronat, A., & Ferrer，A. 1997. The Arabidopsis thaliana FPSl gene

generates a novel mRNA that encodes a mitochondrial famesyl-diphosphate

synthase isoform. J Biol Chem, 272(24)，15381-15388.

Cunillera, N.’ Boronat, A., & Ferrer，A. 2000. Spatial and temporal patterns of GUS

expression directed by 5' regions of the Arabidopsis thaliana farnesyl

69 diphosphate synthase genes FPSl and FPS2. Plant Mol Biol, 44(6)，747-758.

Czechowski, T.，Stitt, M.，Altmann, T.，Udvardi, M.K., & Scheible, W.R. 2005.

Genome-wide identification and testing of superior reference genes for

transcript normalization in Arabidopsis. Plant Physiol, 139(1), 5-17.

De Las Rivas，J., Milicua, J.，& Gomez, R. 1991. Determination of carotenoid

pigments in several tree leaves by reversed-phase high-performance liquid

chromatography. J Chromatogr, 585, 168-172.

De Moraes, C.M.，Mescher, M.C., & Tumlinson，J.H. 2001. Caterpillar-induced

nocturnal plant volatiles repel conspecific females. Nature, 410(6828),

577-580.

Dixon, R.A. 2001. Natural products and plant disease resistance. Nature, 411(6839),

843-847.

Douglas, J.M.，& Rodney，C. 1995. Terpenoid Metabolism. Plant Cell, 7，1015-1026.

Dubey, V.S.，Bhalla, R.，& Luthra, R. 2003. An overview of the non-mevalonate

pathway for terpenoid biosynthesis in plants. JBiosci, 28(5), 637-646.

Eckardt, N.A. 2009. Anew chlorophyll degradation pathway. Plant Cell. 21(3), 700.

Efferth, T. 2006. Molecular pharmacology and pharmacogenomics of artemisinin and

its derivatives in cancer cells. Curr Drug Targets, 7(4)，407-421.

Eisenreich, W.，Menhard, B.，Hylands, P.J., Zenk, M.H., & Bacher，A. 1996. Studies

on the biosynthesis of taxol: the taxane carbon skeleton is not of mevalonoid

origin. Proc Natl Acad Sci USA, 93(13)，6431-6436.

Femandes, A.P., & Holmgren, A. 2004. Glutaredoxins: glutathione-dependent redox

enzymes with functions far beyond a simple thioredoxin backup system.

Antioxid Redox Signal, 6(1)，63-74.

Flores-Sanchez, I.J.，Ortega-Lopez, J., del Carmen Montes-Horcasitas, M.，&

70 Ramos-Valdivia，A.C. 2002. Biosynthesis of sterols and triterpenes in cell

suspension cultures of Uncaria tomentosa. Plant Cell Physiol, 43(12)，

1502-1509.

Fujino, G, Noguchi，T.，Takeda, K.，& Ichijo，H. 2006. Thioredoxin and protein

kinases in redox signaling. Semin Cancer Biol, 16(6)，427-435.

Govindarajan, P., V. Arun, N.，& Pundi’ N.R. 2007. Drugs and drug targets against

malaria. Curr Sci, 92(11).

Hemmerlin, A., Gerber, E.，Feldtrauer, J.R, Wentzinger, L., Hartmann, M.A., Tritsch,

D., Hoeffler, J.R, Rohmer, M.’ & Bach，T.J. 2004. A review of tobacco BY-2

cells as an excellent system to study the synthesis and function of sterols and

other isoprenoids. Lipids, 39(8)，723-735.

Hemmerlin, A., Gerber, E., Hartmann, M.A., Tritsch, D.，Crowell, D.N., Rohmer, M.,

& Bach，D.J. 2006. Tobacco BY-2 Cells: From Cellular Dynamics to Omics.

Biotech in Agri and Forestry, 58, 242-272.

Hemmerlin, A., Hoeffler， J.R, Meyer, O.， Tritsch, D.， Kagan, LA.,

Grosdemange-Billiard, C.，Rohmer, M.，& Bach，T.J. 2003. Cross-talk between

the cytosolic mevalonate and the plastidial methylerythritol phosphate

pathways in tobacco bright yellow-2 cells. J Biol Chem, 278(29)，

26666-26676.

Holmgren, A. 1989. Thioredoxin and glutaredoxin systems. J Biol Chem, 264(24),

13963-13966.

Hugueney, P., Bouvier, F., Badillo, A.，Quennemet, J., d'Harlingue, A., & Camara，B.

1996. Developmental and stress regulation of gene expression for plastid and

cytosolic isoprenoid pathways in pepper fruits. Plant Physiol, 111(2),

619-626.

71 Hunter, W.N. 2007. The non-mevalonate pathway of isoprenoid precursor

biosynthesis. J Biol Chem, 282(30), 21573-21577.

Irizarry, R.A.，Bolstad, B.M.，Collin, F.，Cope, L.M., Hobbs, B.，& Speed, T.R 2003.

Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res, 31(4)，

el5.

Jordan, M.A., & Wilson，L. 2004. Microtubules as a target for anticancer drugs. Nat

Rev Cancer, 4(4)，253-265.

Kang, H.W., Cho, Y.G, Yoon, U.H., & Eun, M.Y. 1998. A Rapid DNA Extraction

Method for RFLP and PGR Analysis from a Single Dry Seed. Plant Mol Biol

Rep, 16，323-399.

Kessler, A., & Baldwin, I.T. 2001. Defensive function of herbivore-induced plant

volatile emissions in nature. Science, 291(5511), 2141-2144.

Kingston, D.G 2009. Tubulin-Interactive Natural Products as Anticancer Agents (1).

J Nat Prod.

Koomneef，A.，& Pieterse，C.M. 2008. Cross talk in defense signaling. Plant Physiol,

146(3), 839-844.

Lange, B.M., & Ghassemian, M. 2003. Genome organization in Arabidopsis thaliana:

a survey for genes involved in isoprenoid and chlorophyll metabolism. Plant

Mol Biol, 51(6), 925-948.

Lange, B.M.，Wildung, M.R., McCaskill, D.，& Croteau, R. 1998. A family of

transketolases that directs isoprenoid biosynthesis via a

mevalonate-independent pathway. Proc Natl Acad Sci USA, 95(5),

2100-2104.

Laule, O., Furholz, A., Chang, H.S.，Zhu, T.，Wang, X.，Heifetz, P.B., Gruissem, W.，

& Lange, M. 2003. Crosstalk between cytosolic and plastidial pathways of

72 isoprenoid biosynthesis in Arabidopsis thaliana. Proc Natl Acad Sci USA,

100(11)，6866-6871.

Li, H.，Tyndale, S.T., Heath, D.D., & Letcher，R.J. 2005. Determination of

carotenoids and all-trans-retinol in fish eggs by liquid

chromatography-electrospray ionization-tandem mass spectrometry. J

Chwmatogr B Analyt Technol Biomed Life Sci, 816(1-2)，49-56.

Lichtenthaler, H.K. 1987. Chlorophylls and carotenoids: pigments of photosynthetic

biomembranes. Method Enzymol, 148’ 350-382.

Lichtenthaler, H.K. 1999. The 1 -Deoxy-D-Xylulose-5-Phosphate Pathway of

Isoprenoid Biosynthesis in Plants. Annu Rev Plant Physiol Plant Mol Biol, 50,

47-65.

Lichtenthaler, H.K., Schwender, J., Disch, A., & Rohmer，M. 1997. Biosynthesis of

isoprenoids in higher plant chloroplasts proceeds via a

mevalonate-independent pathway. FEES Lett, 400(3)，271-274.

Lixin, Z.，& Arnold, L.D. 2005. Natural products: drug discovery and therapeutic

medicine. Totowa, New Jersey: Humana press. 210-212 pp.

Long, B.H., Carboni, J.M., Wasserman, A.J.，Cornell, L.A., Casazza, A.M., Jensen,

P.R.，Lindel，T.，Fenical, W., & Fairchild，C.R. 1998. Eleutherobin, a novel

cytotoxic agent that induces tubulin polymerization, is similar to paclitaxel

(Taxol). Cancer Res, 58(6)，1111-1115.

Loreto, R, Ciccioli，P., Cecinato, A., Brancaleoni, E.，Frattoni, M.，& Tricoli, D. 1996.

Influence of Environmental Factors and Air Composition on the Emission of

[alpha]-Pinene from Quercus ilex Leaves. Plant Physiol, 110(1), 267-275.

Lunn, J.E. 2007. Compartmentation in plant metabolism. J Exp Bot, 58(1), 35-47.

Manzano, D.，Busquets，A., Closa, M.，Hoyerova, K.，Schaller, H.，Kaminek, M.，

73 Arro, M.，& Ferrer，A. 2006. Overexpression of famesyl diphosphate synthase

in Arabidopsis mitochondria triggers light-dependent lesion formation and

alters cytokinin homeostasis. Plant Mol Biol, 61(1-2), 195-213.

Manzano, D.，Femandez-Busquets, X., Schaller, H.，Gonzalez, V.，Boronat, A., Arro,

M., & Ferrer，A. 2004. The metabolic imbalance underlying lesion formation

in Arabidopsis thaliana overexpressing famesyl diphosphate synthase (isoform

1S) leads to oxidative stress and is triggered by the developmental decline of

endogenous HMGR activity. Planta, 219(6)，982-992.

Martin, D.，Piulachs, M.D.，Cunillera, N.，Ferrer, A., & Belles’ X. 2007.

Mitochondrial targeting of famesyl diphosphate synthase is a widespread

phenomenon in eukaryotes. Biochim Biophys Acta, 1773(3)，419-426.

Masferrer, A., Arro, M.，Manzano, D.，Schaller, H.，Femandez-Busquets, X.，

Moncalean, P., Fernandez, B.，Cunillera, N., Boronat, A., & Ferrer，A. 2002.

Overexpression of Arabidopsis thaliana famesyl diphosphate synthase (FPSIS)

in transgenic Arabidopsis induces a cell death/senescence-like response and

reduced cytokinin levels. Plant J, 30(2)，123-132.

Matusheski, N.V.，& Jeffery，E.H. 2001. Comparison of the bioactivity of two

glucoraphanin hydrolysis products found in broccoli, sulforaphane and

sulforaphane nitrile. JAgric Food Chem, 49(12)，5743-5749.

Mayrhofer, S., Teuber, M.，Zimmer, I•，Louis, S., Fischbach, R.J.，& Schnitzler，J.R

2005. Diurnal and seasonal variation of isoprene biosynthesis-related genes in

grey poplar leaves. Plant Physiol, 139(1), 474-484.

Mentewab, A., & Stewart, C.N., Jr. 2005. Overexpression of an Arabidopsis thaliana

ABC transporter confers kanamycin resistance to transgenic plants. Nat

Biotechnol, 23(9), 1177-1180.

74 Miao, Y.，& Jiang，L. 2007. Transient expression of fluorescent fusion proteins in

protoplasts of suspension cultured cells. Nat Protoc, 2(10), 2348-2353.

Ndamukong, I.，Abdallat, A.A., Thurow, C.，Fode, B.，Zander, M.，Weigel, R.，&

Gatz，C. 2007. SA-inducible Arabidopsis glutaredoxin interacts with TGA

factors and suppresses JA-responsive PDF 1.2 transcription. Plant J, 50(1),

128-139.

Ortiz-Gomez, A” Jimenez, C.，Estevez, A.M., Carrero-Lerida, J., Ruiz-Perez, L.M.,

& Gonzalez-Pacanowska, D. 2006. Famesyl diphosphate synthase is a

cytosolic enzyme in Leishmania major promastigotes and its overexpression

confers resistance to risedronate. Eukaryot Cell, 5(7), 1057-1064.

Penuelas, J., & Llusi‘ J. 1999. Seasonal emission of monoterpenes by the

Mediterranean tree Quercus ilex L. in field conditions: relations with

photosynthetic rates, temperature and volatility. Physiologia Plantani, 105,

641-647.

Porat, A., Lillig, C.H., Johansson, C., Femandes, A.P., Nilsson, L.，Holmgren, A., &

Beckwith, J. 2007. The reducing activity of glutaredoxin 3 toward cytoplasmic

substrate proteins is restricted by methionine 43. Biochemistry, 46(11),

3366-3377.

Porter, J.W., & Spurgeon，S.L. 1981. Biosynthesis of Isoprenoid Compounds. New

York: Wiley.

Reinhold, C.，Niczyporuk, S.，Beran, K.C., & Jahns，P. 2008. Short-term

down-regulation of zeaxanthin epoxidation in Arabidopsis thaliana in response

to photo-oxidative stress conditions. Biochim Biophys Acta, 1777(5), 462-469.

Ro, D.K.，Paradise, E.M., Ouellet, M., Fisher, K.J., Newman, K.L., Ndungu, J.M.,

Ho, K.A., Eachus, R.A.，Ham, T.S., Kirby, J., Chang, M.C., Withers, S.T.,

75 Shiba, Y.，Sarpong, R.，& Keasling，J.D. 2006. Production of the antimalarial

drug precursor artemisinic acid in engineered yeast. Nature, 440(7086)，

940-943.

Rodriguez-Concepcion, M., & Boronat, A. 2002. Elucidation of the methylerythritol

phosphate pathway for isoprenoid biosynthesis in bacteria and plastids. A

metabolic milestone achieved through genomics. Plant Physiol, 130(3)，

1079-1089.

Rohmer, M.，Knani, M.，Simonin, P., Sutter, B.，& Sahm，H. 1993. Isoprenoid

biosynthesis in bacteria: a novel pathway for the early steps leading to

isopentenyl diphosphate. Biochem J, 295 (Pt 2)，517-524.

Rouhier, N.，Gelhaye，E.，& Jacquot，J.P. 2004. Plant glutaredoxins: still mysterious

reducing systems. Cell Mol Life Sci, 61(11)，1266-1277.

Sallaud, C.，Rontein, D.，Onillon, S.，Jabes, F.，Duffe, P., Giacalone, C.，Thoraval, S.，

Escoffier, C.，Herbette，G，Leonhardt, N.，Causse, M., & Tissier，A. 2009. A

novel pathway for sesquiterpene biosynthesis from Z,Z-famesyl

pyrophosphate in the wild tomato Solanum habrochaites. Plant Cell, 21(1),

301-317.

Schmelz, E.A., LeClere, S.，Carroll, M.J.，Albom, H.T.，& Teal, RE. 2007. Cowpea

chloroplastic ATP synthase is the source of multiple plant defense elicitors

during insect herbivory. Plant Physiol, 144(2)，793-805.

Schnee, C.，Kollner, T.G, Held, M.，Turlings, T.C., Gershenzon, J., & Degenhardt，J.

2006. The products of a single maize sesquiterpene synthase form a volatile

defense signal that attracts natural enemies of maize herbivores. Proc Natl

Acad Sci USA, 103(4)，1129-1134.

Schultz, J.C. 2002. Biochemical ecology: how plants fight dirty. Nature, 416(6878),

76 267.

Schwender，J., Seemann, M.，Lichtenthaler, H.K., & Rohmer，M. 1996. Biosynthesis

of isoprenoids (carotenoids, sterols, prenyl side-chains of chlorophylls and

plastoquinone) via a novel pyruvate/glyceraldehyde 3-phosphate

non-mevalonate pathway in the green alga Scenedesmus obliquus. Biochem J’

316 (Pt 1)，73-80.

Steeghs, M.，Bais, H.P., de Gouw, J.’ Goldan, P., Kuster, W.，Northway, M., Fall, R.’

& Vivanco, J.M. 2004. Proton-transfer-reaction mass spectrometry as a new

tool for real time analysis of root-secreted volatile organic compounds in

Arabidopsis. Plant Physiol, 135(1), 47-58.

Taiz, L., & Zeiger，E. 2002. Plany Physiology. Sunderland, MA: Sinauer Associates,

Inc.

Terry, M.J.，& Kendrick’ R.E. 1999. Feedback inhibition of chlorophyll synthesis in

the phytochrome chromophore-deficient aurea and yellow-green-2 mutants of

iomdXo. Plant Physiol, 119(1), 143-152. van Klink, J., Becker, H.，Andersson, S.，& Boland，W. 2003. Biosynthesis of

anthecotuloide, an irregular sesquiterpene lactone from Anthemis cotula L.

(Asteraceae) via a non-farnesyl diphosphate route. Org Biomol Chem, 1(9),

1503-1508.

Vogeli，U.，& Chappell, J. 1988. Induction of Sesquiterpene Cyclase and Suppression

of Squalene Synthetase Activities in Plant Cell Cultures Treated with Fungal

Elicitor. Plant Physiol, 88(4)，1291-1296.

Walker, J.M. 2002. The Protein Protocols Handbook. New Jersey: Humana Press.

15-22 pp.

Weech, M.H., Chapleau, M.’ Pan, L.，Ide, C.，& Bede，J.C. 2008. Caterpillar saliva

77 interferes with induced Arabidopsis thaliana defence responses via the

systemic acquired resistance pathway. J Exp Bot, 59(9), 2437-2448.

WHO. 2003. Assessment of the safety of artemisinin compounds in pregnancy. In

Geneva: World Health Organization

WHO. 2006. Guidelines for the Treatment of Malaria. In Geneva: World Health

Organization.

Withers, S.T., & Keasling, J.D. 2007. Biosynthesis and engineering of isoprenoid

small molecules. Appl Microbiol Biotechnol, 73(5), 980-990.

Wittstock, U., & Halkier, B.A. 2002. Glucosinolate research in the Arabidopsis era.

Trends Plant Sci, 7(6), 263-270.

Xiao, S.H. 2005. Development of antischistosomal drugs in China, with particular

consideration to praziquantel and the artemisinins. Acta Trop, 96(2-3)，

153-167.

Zhao, Z.，Zhang, W.，Stanley, B.A., & Assmann，S.M. 2008. Functional proteomics

of Arabidopsis thaliana guard cells uncovers new stomatal signaling pathways.

Plant Cell. 20(12), 3210-3226.

78 Appendices

Appendix A. Primers designed for homozygous mutant screening for fps mutants

Primer Sequence (5，to 3，) fn? LP+RP BP + RP

(�C ) Product Product

SALK LBbl border primer GCGTGGACCGCTTGCTGCAACT ^

SALK_045864 LP primer TTTCCATTCAAAGACACGCTC 60.24 955 bp 430-730bp

SALK_045864 RP primer GAATCGAAGGAGTACCCGTTC 59.95

SALK_073576 LP primer AAAACCATTGCATCAGGTCAG 59.99 1007 bp 467-767bp

SALK_073576RP primer GACCTTCTTCATGACCCTTCC 59.93

":melting temperature

LP: left primer; RP: right primer

79 Appendix B. Primer pairs designed for fps ORE, common sequence and fpsl specific n

Gene Locus Remarks Direction Sequence (5，to 3，）

J^l AT5G47770 O^ Forward ACATCATTTGCATCTGAGAAGTC]

Reverse TGCTTGGATTGCTTTACTTTGGTG

fps2 AT4G17190 ORF Forward ATGGACAACTCTGTCACACGCCG^

Reverse TTTAGCACTGCTTGAATTGCTTTAi

fpsl AT5G47770 Specific region Forward GGAGAAGAAGGAGGAATATGAGl

Reverse TGGATACGACGACGATAGAGAC

fpsl AT5G47770 Common region Forward TGCTGATCCTGAGACACTTG andfps2 AT4G17190 Reverse GTTCGGCTTTACCATAGTTCTC

*: annealing temperature

80 Appendix C. Primer pairs designed for studying expression level of the downstream gen

The tested genes included famesyl diphosphate synthase 1 (FPSl) (AT5G47770), famesyl diphosphate sy (AT4G34640), squalene synthase 2 (SQS2) (AT4G34650), squalene monoxygnease 1 (SQPl) (AT5G2415 synthase 1 (LUPl) (AT1G78970), lupeol synthase 2 (LUP2) (AT1G78960), At5G23960 and At5G4462 internal control for the normalization of reverse transcription polymerase chain reaction (RT-PCR) (Czec Gene Locus Primer Sequence (5，to 3，) UBC21 At5G25760 Forward CAGTCTGTGTGTAGAGCTATCATAGCAT Reverse GATTCCCTGAGTCGCAGTTAAGA FPSl AT5G47770 Forward CTTCGCAATCACATCCACA Reverse AGCAACGCACAAGCAACA FPS2 AT4G17190 Forward TGGCAAGATAGGGACAGACA Reverse GGTTCGGCTTTACCATAGTTC SQSl AT4G34640 Forward ATTCGCTGGCCGAATCG Reverse AGACCACGAGGGATAAATTGCA SQS2 AT4G34650 Forward CTTGGCACCGAGCTTCGTA Reverse CAACAGTATCAAGAGCTCGGAGAA SQPl AT5G24150 Forward TTCACAATGGCCGATTCGT Reverse AGGCGCACATTGGGAAGA SQP2 AT5G24140 Forward CTTCTCGCCTTTGTTCT Reverse GTGCGTCTATGTCCTCC AT5G23960 Forward CGCTTTCTCACGTTCTCATTTAGA Reverse TCTTGATGCGAATTGCGAGAT AT5G44630 Forward CGCTTGCGATACTTATGGTTCA Reverse AATCTGGATCCCATCTTTCCAA LUPl AT1G78970 Forward TGTCAATGCGTGGGAGC Reverse AAGAGCAAACCAAGTAGCGTA LUP2 AT1G78960 Forward TACGCTACTTGGTTTGCTCTGA Reverse TGTTCCCTTCTAATGGTATGTATCTC *: annealing tenqjerature

81 Appendix D. Annotations of differentially expressed genes in fpsl mutant

logFC P.Value Locus Annotation from The Arabidopsis Information Resource (TAIR)

3.1820160.032703 AT4G11650ATOSM34 (OSMOTIN 34)

2.964299 0.031084 AT5G42900 similar to unknown proteinAT4G33980.1

2.350627 0.048197 AT2G39030 GCN5-related N-acetyltransferase (GNAT) family protein

2.313541 0.040976 ATI G73260 trypsin and protease inhibitor family protein / Kunitz family protein

2.228318 0.002692 AT3G49620 DINl 1 (DARK INDUCIBLE 11); oxidoreductase

2.178525 0.005043 T3G55970 oxidoreductase, 20G-Fe(II) oxygenase family protein

FAMT (FARNESOIC ACID CARBOXYL-O-METHYLTRANSFERASE); 1.948267 0.000949 AT3G44860 S-adenosylmethionine-dependent methyltransferase/ famesoic acid 0-methyltransferase

1.948267 0.000949 AT3G44870 S-adenosyl-L-methionine:carboxyl methyltransferase family protein

1.883537 0.002243 AT1G28480 GRX480; thiol-disulfide exchange intermediate

-1.81366 0.003959 ATI G13650 similar to 18S pre-ribosomal assembly protein gar2-related

1.764343 0.037987 AT3G07650 C0L9 (CONSTANS-LIKE 9); transcription factor/ zinc ion binding

1.703106 0.002355 AT2G19970 pathogenesis-related protein, putative

1.667361 0.007835 AT5G05600 oxidoreductase, 20G-Fe(II) oxygenase family protein

1.64957 0.028641 AT1G76790 0-methyltransferase family 2 protein

1.628714 0.004074 AT5G67080 MAPKKK19 (Mitogen-activated protein kinase kinase kinase 19); kinase

-1.55678 0.020185 AT2G41230 unknown protein

1.54699 0.022352 AT2G24850 TAT3 (TYROSINE AMINOTRANSFERASE 3); transaminase

1.513835 0.00769 AT3G27210 Identical to Uncharacterized protein At3g27210 (Y-2)

-1.47978 0.018367 AT2G15020 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G64190.1)

1.424894 0.023116 AT5G67480 BT4 (BTB AND TAZ DOMAIN PROTEIN 4); protein binding / transcription regulator

ATGSTU17/ERD9/GST30/GST30B (EARLY-RESPONSIVE TO DEHYDRATION 9); -1.41692 0.006174 AT1G10370

glutathione transferase

1.406242 0.039261 AT3G01420 ALPHA-DOXl (ALPHA-DIOXYGENASE1)

1.405027 0.047097 AT5G13220 JAS1/JAZ10/TIFY9 (JASMONATE-ZIM-DOMAIN PROTEIN 10)

-1.39449 0.002316 AT4G27030 small conjugating protein ligase

-1.23187 0.005172 AT5G24150 SQPl (Squalene monooxygenase 1)

1.230614 0.042556 AT5G57630 CIPK21 (CBL-INTERACTING PROTEIN KINASE 21); kinase

1.197344 0.00341 AT5G12340 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G28190.1)

-1.1926 0.014646 AT3G21670 nitrate transporter (NTP3)

-1.18816 0.01854 AT3G01550 triose phosphate/phosphate translocator, putative

-1.18264 0.007857 AT4G24700 unknown protein

1.171911 0.025389 AT1G28330 DRMl (DORMANCY-ASSOCIATED PROTEIN 1)

-1.16963 0.004887 AT2G19650 DCl domain-containing protein

82 -1.16239 0.039757 AT3G56290 similar to hypothetical protein [Vitis vinifera] (GB:CAN75527.1)

-1.14934 0.005648 AT1G59930 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G59920.1)

1.120686 0.017633 ATI G51760 IAR3 (lAA-ALANINE RESISTANT 3); metallopeptidase

1.120686 0.017633 ATIG51780 ILLS (lAA-leucine resistant (ILR)-like gene 5); metallopeptidase

-1.11422 0.015231 AT5G62430 CDFl (CYCLING DOF FACTOR 1); DNA binding / protein binding / transcription factor

1.111148 0.038231 AT3G51450 strictosidine synthase family protein

-1.10713 0.021117 AT5G35490 unknown protein

-1.10563 0.012195 AT3G21420 oxidoreductase, 20G-Fe(II) oxygenase family protein

-1.10413 0.012993 AT2G04039 similar to unnamed protein product [Vitis vinifera] (GB:CA041070.1)

-1.06604 0.006762 AT4G11460 protein kinase family protein

FPSl (FARNESYL DIPHOSPHATE SYNTHASE 1); dimethylallyltranstransferase/ 1.06588 0.00196 AT5G47770 geranyltranstransferase

1.041885 0.032474 AT3G49120 ATPCB/ATPERX34/PERX34/PRXCB (PEROXIDASE 34); peroxidase

1.041885 0.032474 AT3G49110, ATPCA/ATPRX33/PRX33/PRXCA (PEROXIDASE 33); peroxidase

-1.04151 0.044648 AT3G44990 XTR8 (xyloglucan:xyloglucosyl transferase 8); hydrolase, acting on glycosyl bonds

1.02634 0.001504 AT5G18470 curculin-丨ike (mannose-binding) lectin family protein

-1.02258 0.000148 AT4G38860 auxin-responsive protein, putative

-1.01841 0.023643 AT2G21320 zinc finger (B-box type) family protein

-1.01211 0.000345 AT5G19730 pectinesterase family protein

1.006761 0.00179 AT2G27310 F-box family protein

83 Appendix E. Annotations of differentially expressed genes in fps2 mutant

logFC P.Value Locus Annotation from The Arabidopsis Information Resource (TAIR)

6.078548 0.002565 AT4G11650 ATOSM34 (OSMOTIN 34)

4.994898 0.001686 AT3G60140 DIN2 (DARK INDUCIBLE 2); hydrolase, hydrolyzing 0-glycosyl compounds

4.891249 0.002216 ATI G73260 trypsin and protease inhibitor family protein / Kunitz family protein

4.400317 0.000987 AT5G43580 serine-type endopeptidase inhibitor

4.147479 0.00046 AT3G01420 ALPHA-DOXl (ALPHA-DIOXYGENASE 1)

3.654687 0.000526 AT3G09220 LAC7 (laccase 7); copper ion binding / oxidoreductase

3.494942 0.000621 AT3G55970 oxidoreductase, 20G-Fe(II) oxygenase family protein

3.487299 0.002377 AT2G37750 unknown protein

3.361468 0.013645 AT2G46750 FAD-binding domain-containing protein

3.198549 0.000546 AT5G63600 flavonol synthase, putative

3.162459 0.030539 AT5G39520 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G39530.1)

2.895901 0.003646 AT3G04320 endopeptidase inhibitor

2.895901 0.003646 AT3G04330 trypsin and protease inhibitor family protein / Kunitz family protein

2.809133 0.021563 AT5G64100 peroxidase, putative

2.756428 0.02669 AT3G12500 ATHCHIB (BASIC CHITINASE); chitinase

2.610478 0.005236 AT1G80160 丨actoylglutathione lyase family protein / glyoxalase I family protein

2.514176 0.008247 ATI G61820 BGLU46; hydrolase, hydrolyzing 0-glycosyl compounds

2.461591 0.005691 AT2G43510 ATTIl (ARABIDOPSIS THALIANA TRYPSIN INHIBITOR PROTEIN 1)

SRGl (SENESCENCE-RELATED GENE 1); oxidoreductase, acting on paired donors,

2.428038 0.010115 ATIG17020 with incorporation or reduction of molecular oxygen, 2-oxoglutarate as one donor, and

incorporation of one atom each of oxygen into both donors

2.41775 0.044091 AT2G39030 GCN5-related N-acetyltransferase (GNAT) family protein

2.406465 0.033836 AT1G65970 TPX2 (THIOREDOXIN-DEPENDENT PEROXIDASE 2

2.406465 0.033836 AT1G60740 peroxiredoxin type 2, putative

ATPTl (PHOSPHATE TRANSPORTER 1); carbohydrate transmembrane transporter/ 2.396639 0.000112 AT5G43350 phosphate transmembrane transporter/ sugar:hydrogen ion symporter

APT1/PHT1;2/PHT2 (PHOSPHATE TRANSPORTER 2); carbohydrate transmembrane

2.396639 0.000112 AT5G43370 transporter/ inorganic phosphate transmembrane transporter/ phosphate transmembrane

transporter/ sugarhydrogen ion symporter

2.380759 0.007728 AT1G62760 invertase/pectin methylesterase inhibitor family protein

2.341394 0.002167 AT3G49620 DINl 1 (DARK INDUCIBLE 11); oxidoreductase

-2.14034 0.026711 AT1G65390 ATPP2-A5; carbohydrate binding

2.137804 0.002952 AT2G44010 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT3G59880.1)

84 2.125564 0.01249 AT1G55020 LOXl (Lipoxygenase 1); lipoxygenase

-2.11758 0.035457 AT1G01190 CYP78A8 (cytochrome P450, family 78, subfamily A, polypeptide 8); oxygen binding

2.104792 0.04889 AT2G32660 disease resistance family protein / LRR family protein

2.073733 0.025146 AT1G61810 BGLU45; hydrolase, hydrolyzing 0-glycosyl compounds

ARR6 (RESPONSE REGULATOR 6); transcription regulator/ two-component response -2.0659 0.004381 AT5G62920

regulator

2.004497 0.042662 AT1G68620 hydrolase

1.989452 0.045826 ATIG09500 cinnamyl-alcohol dehydrogenase family / CAD family

1.981107 0.03019 AT5G24160 squalene monooxygenase 1,2 / squalene epoxidase 1 ^ (SQPl ,2)

1.957941 0.001266 AT2G19970 pathogenesis-related protein, putative

-1.94983 0.022627 AT2G18210 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G36500.1)

1.945865 0.020569 AT5G44380 FAD-binding domain-containing protein

1.855274 0.040382 AT2G29470 ATGSTU3 (GLUTATHIONE S-TRANSFERASE 21); glutathione transferase

1.851602 0.018755 AT2G46680 ATHB-7 (ARABIDOPSIS THALIANA HOMEOBOX 7); transcription factor

1.805458 0.025734 AT3G14060 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G54120.1)

-1.80181 0.008308 AT2G21650 MEE3 (maternal effect embryo arrest 3); DNA binding / transcription factor

1.785326 0.0097 AT2G26020 PDF 1.2b (plant defensin 1.2b)

ATPDR9/PDR9 (PLEIOTROPIC DRUG RESISTANCE 9); ATPase, coupled to 1.777782 0.001053 AT3G53480 transmembrane movement of substances

1.7636 0.04287 ATI G15380 lactoylglutathione lyase family protein / glyoxalase I family protein

1.750313 0.000101 AT2G17500 auxin efflux carrier family protein

1.74023 0.028288 AT5G44420 PDF1.2 (Low-molecular-weight cysteine-rich 77)

1.730429 0.039561 AT2G44790 UCC2 (UCLACYANIN 2); copper ion binding

ATPDR12/PDR12 (PLEIOTROPIC DRUG RESISTANCE 12); ATPase, coupled to 1.716938 0.041204 ATI G15520 transmembrane movement of substances

1.708166 0.004187 AT1G64660 ATMGL; catalytic/ methionine gamma-lyase

-1.7057 0.001005 AT1G59930 similar to unknown protein [Arabidopsis thaliana] (TAIR: ATI G59920.1)

1.696841 0.021448 AT1G52890 ANAC019 (Arabidopsis NAC domain containing protein 19); transcription factor

-1.69171 0.000371 AT3G62950 glutaredoxin family protein

1.688071 0.016125 AT2G28110 FRA8 (FRAGILE FIBERS); transferase

xyloglucan :xyloglucosyl transferase, putative / xyloglucan endotransglycosylase, putative -1.67954 0.011 AT4G37800

/ endo-xyloglucan transferase, putative

-1.64846 0.009834 AT3G62930 glutaredoxin family protein

1.631759 0.004173 AT1G28480 GRX480; thiol-disulfide exchange intermediate

1.630881 0.013029 AT1G69880 ATH8 (thioredoxin H-type 8); thiol-disulfide exchange intermediate

1.625883 0.039541 AT2G47190 MYB2 (myb domain protein 2); DNA binding / transcription factor

1.622082 0.011681 AT4G35060 heavy-metal-associated domain-containing protein / copper chaperone (CCH)-related

1.62006 0.024021 AT1G05340 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT2G32210.1) 85 1.614536 0.017389 AT2G29460 ATGSTU4 (GLUTATHIONE S-TRANSFERASE 22); glutathione transferase

-1.61275 0.030661 AT4G10270 wound-responsive family protein

ATGSTU24 (ARABIDOPSIS THALIANA GLUTATHIONE S-TRANSFERASE 1.599528 0.001593 ATI G17170 (CLASS TAU) 24); glutathione transferase

1.597171 0.03722 AT2G18050 HIS 1-3 (HISTONE HI-3); DNA binding

inositol-3-phosphate synthase, putative / myo-inositol-1 -phosphate synthase, putative / -1.58721 0.018748 AT5G10170

MI-l-P synthase, putative

1.582335 0.009665 AT5G05600 oxidoreductase, 20G-Fe(II) oxygenase family protein

1.579564 0.027074 AT3G53980 protease inhibitor/seed storage/lipid transfer protein (LTP) family protein

-1.57553 0.011503 AT4G15660 glutaredoxin family protein

-1.57356 0.021634 AT5G22390 similar to unnamed protein product [Vitis vinifera] (GB:CA048728.1)

1.567748 0.025526 ATI G13300 myb family transcription factor

1.535397 0.038344 AT5G07440 GDH2 (GLUTAMATE DEHYDROGENASE 2); oxidoreductase

-1.53317 0.008087 AT4G15680 glutaredoxin family protein ‘

-1.52667 0.007337 AT3G05320 similar to unnamed protein product [Vitis vinifera] (GB:CA044440.1)

similar to 18S pre-ribosomal assembly protein gar2-related [Arabidopsis thaliana] -1.52464 0.00814 ATI G13650

(TAIR: AT2G03810.4)

-1.52111 0.009649 AT1G73602 CPuORF32 (Conserved peptide upstream open reading frame 32)

-1.52111 0.009649 AT1G73600 phosphoethanolamine N-methyltransferase

FPS2 (FARNESYL DIPHOSPHATE SYNTHASE 2); dimethylallyltranstransferase/ 1.515266 0.013422 AT4G17190 geranyltranstransferase

1.512632 0.030832 AT3G61190 BAPl (BON ASSOCIATION PROTEIN 1)

1.504884 0.026042 AT1G23120 major latex protein-related / MLP-related

1.502652 0.044924 AT1G54020 myrosinase-associated protein, putative

1.486601 2.95E-05 AT5G62165 AGL42 (AGAMOUS LIKE 42); transcription factor

-1.48577 0.009872 AT4G15700 glutaredoxin family protein

1.479122 0.029108 AT4G25000 AMYl/ATAMYl (ALPHA-AMYLASE-LIKE); alpha-amylase

1.471652 0.032737 AT1G06180 ATMYB13 (myb domain protein 13); DNA binding / transcription factor

UGT73B3 (UDP-GLUCOSYL TRANSFERASE 73B3); UDP-glycosyltransferase/ 1.470683 0.046642 AT4G34131

abscisic acid glucosyltransferase/ transferase, transferring hexosyl groups

UGT73B2; UDP-glucosyltransferase/ UDP-glycosyltransferase/ flavonol 1.470683 0.046642 AT4G34135 3 -0-glucosyltransferase

1.468575 0.006749 AT4G33150 LKR (SACCHAROPINE DEHYDROGENASE)

1.456689 0.036849 AT4G37010 caltractin, putative / centrin, putative

1.450582 0.005645 AT1G78780 pathogenesis-related family protein

1.445741 0.007957 AT3G10450 SCPL7; serine carboxypeptidase

-1.44184 0.00468 AT4G15690 glutaredoxin family protein

1.436968 0.026992 AT2G31945 similar to unknown protei86 n [Arabidopsis thaliana] (TAIR:ATIG05575.1) meprin and TRAP homology domain-containing protein / MATH domain-containing -1.43405 0.00826 AT4G01390 protein

GLP9 (GERMIN-LIKE PROTEIN 9); manganese ion binding / metal ion binding / 1.432677 0.034865 AT4G14630 nutrient reservoir

PGDH (3-PHOSPHOGLYCERATE DEHYDROGENASE); phosphoglycerate 1.415762 0.017136 ATI G17745 dehydrogenase

1.39987 0.001719 AT5G12340 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G28190.1);

1.390072 0.031667 ATI G18980 germin-like protein, putative

1.379746 0.006765 AT5G57910 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G30630.1)

IMS2/MAM-L/MAM3 (METHYLTHIOALKYMALATE SYNTHASE-LIKE); -1.379 0.018638 AT5G23020

2-isopropylmalate synthase/ methylthioalkylmalate synthase

-1.37586 0.000405 AT1G03020 glutaredoxin family protein

1.359087 0.036821 AT3G14680 CYP72A14 (cytochrome P450, family 72, subfamily A, polypeptide 14); oxygen binding

1.341058 0.036622 AT2G24850 TAT3 (TYROSINE AMINOTRANSFERASE 3); transaminase

1.339205 0.00748 AT2G03090 ATEXPA15 (ARABIDOPSIS THALIANA EXPANSIN A15)

-1.33853 0.023252 AT4G30110 HMA2 (Heavy metal ATPase 2); cadmium-transporting ATPase

-1.32918 0.002655 AT4G11460 protein kinase family protein

-1.32753 0.001038 AT3G61210 embryo-abundant protein-related

1.316338 0.02986 ATI G74020 SS2 (STRICTOSIDINE SYNTHASE 2); strictosidine synthase

-1.309 0.00415 AT3G46880 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G59080.1)

xyloglucan :xyloglucosyl transferase, putative / xyloglucan endotransglycosylase, putative 1.308651 0.026343 AT3G48580 / endo-xyloglucan transferase, putative

-1.30631 0.001466 AT4G19530 disease resistance protein (TIR-NBS-LRR class), putative

1.301812 0.005641 AT5G63580 flavonol synthase, putative

-1.29434 0.017774 AT5G27290 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G54680.1)

-1.29046 0.013277 AT5G62280 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT2G45360.1)

1.281991 0.015435 AT3G49120 ATPCB/ATPERX34/PERX34/PRXCB (PEROXIDASE 34); peroxidase

1.281991 0.015435 AT3G49110 ATPCA/ATPRX33/PRX33/PRXCA (PEROXIDASE 33); peroxidase

1.276189 0.04016 AT3G61930 unknown protein

1.270337 0.00119 AT4G01430 nodulin MtN21 family protein

1.26632 0.031719 AT4G37990 ELI3-2 (ELICITOR-ACTIVATED GENE 3)

-1.25047 0.013172 AT5G01050 laccase family protein / diphenol oxidase family protein

-1.25047 0.013172 AT5G01040 LAC8 (laccase 8); copper ion binding / oxidoreductase

1.246969 0.017565 AT5G61820 similar to MtN19-like protein [Pisum sativum] (GB:AAU 14999.2)

1.246521 0.010702 AT1G59860 17.6 kDa class I heat shock protein (HSP17.6A-CI)

1.246521 0.010702 AT1G07400 17.8 kDa class I heat shock protein (HSP17.8-C1)

1.245732 0.044253 AT5G06740 丨ectin protein kinase family protein

1.234069 0.040391 AT4G17215 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G47635.1) 87 -1.23405 0.026999 AT5G12050 similar to unnamed protein product [Vitis vinifera] (GB:CA045643.1)

1.226312 0.02738 AT5G12880 proline-rich family protein

1.222604 0.028809 AT3G54040 photoassimilate-responsive protein-related

-1.21932 0.048157 AT1G67910 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G24577.1)

1.218383 0.001958 AT1G52410 TSAl (TSK-ASSOCIATING PROTEIN 1); calcium ion binding / protein binding

1.216573 0.048345 AT5G44480 DUR (DEFECTIVE UGE IN ROOT); catalytic

-1.21058 0.026053 AT5G54720 ankyrin repeat family protein

1.210003 0.022132 AT3G59010 pectinesterase family protein

1.207763 0.020805 ATI G45145 ATTRX5 (thioredoxin H-type 5); thiol-disulfide exchange intermediate

1.206989 0.014428 AT2G41380 embryo-abundant protein-related

1.20497 0.0079 AT3G56170 CAN (CA-2+ DEPENDENT NUCLEASE); nuclease

epsin N-terminal homology (ENTH) domain-containing protein / clathrin assembly -1.2028 0.001016 AT1G33340

protein-related

1.20079 0.000865 AT5G44610 MAP 18 (MICROTUBULE-ASSOCIATED PROTEIN 18); microtubule binding

-1.20027 0.02029 AT4G17810 nucleic acid binding / transcription factor/ zinc ion binding

-1.19767 0.015792 AT5G35490 unknown protein

-1.19581 0.012233 AT4G12320 CYP706A6 (cytochrome P450, family 706, subfamily A, polypeptide 6); oxygen binding

-1.19581 0.012233 AT4G12310 CYP706A5 (cytochrome P450, family 706, subfamily A, polypeptide 5); oxygen binding

ALDH7B4 (ALDEHYDE DEHYDROGENASE 7B4); 3-chloroallyl aldehyde 1.194294 0.043464 AT1G54100 dehydrogenase

1.183024 0.011204 AT5G66400 RAB18 (RESPONSIVE TO ABA 18)

1.182117 0.001631 AT3G12700 aspartyl protease family protein

1.175987 0.028982 ATI G25530 lysine and histidine specific transporter, putative

1.175318 0.01519 AT5G67080 MAPKKK19 (Mitogen-activated protein kinase kinase kinase 19); kinase

-1.16458 0.001201 AT1G76130 AMY2/ATAMY2 (ALPHA-AMYLASE-LIKE 2); alpha-amylase

-1.16271 0.162139 AT3G26960 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G41050.1)

Identical to RING-H2 finger protein ATLIA (ATLIA) [Arabidopsis Thaliana] -1.1617 0.022432 AT1G04360

(GB:P93823);

1.159693 0.07363 AT2G47180 similar to ATGOLS2 (ARABIDOPSIS THALIANA GALACTINOL SYNTHASE 2),

1.159666 0.220814 AT4G32870 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT2G25770.2)

1.15927 0.008646 AT3G20340 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G21920.1)

1.157631 0.114573 AT1G67030 Identical to Zinc finger protein 6 (ZFP6) [Arabidopsis Thaliana] (GB:Q39265)

Identical to Homeobox protein knotted-l-like 1 (KNATl) [Arabidopsis Thaliana] 1.157265 0.046975 AT4G08150

(GB:P46639;GB:Q8S3L9;GB:Q8S3M0)

1.156179 0.103633 AT3G02240 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT3G02242.1)

-1.1549 0.068095 AT5G59130 similar to ATSBT4.12, subtilase [Arabidopsis thaliana] (TAIR:AT5G59090.3)

-1.15099 0.055168 ATCGOlllO Identical to NAD (ndhH) [Arabidopsis Thaliana] (GB:P56753)

-1.14773 0.365549 AT5G24110 Identical to Probable WRKY transcription factor 30 (WRKY30) [Arabidopsis Thaliana] 88 (GB:Q9FL62)

1.146922 0.151324 AT5G38780 similar to methyltransferase-related [Arabidopsis thaliana] (TAIR:AT5G38100.1)

similar to SUSl (SUCROSE SYNTHASE 1), UDP-glycosyltransferase/ sucrose synthase 1.146902 0.118544 AT4G02280 [Arabidopsis thaliana] (TAIR:AT5G20830.2)

similar to FAD-binding domain-containing protein [Arabidopsis thaliana] 1.143775 0.031184 AT2G34810 (TAIR:AT1G30720.1)

similar to TOO, P-P-bond-hydrolysis-driven protein transmembrane transporter 1.142828 0.054049 AT4G03320 [Arabidopsis thaliana] (TAIR:AT1G04940.1)

Identical to Peroxidase 21 precursor (PER21) [Arabidopsis Thaliana] 1.137369 0.145785 AT2G37130 (GB:Q42580;GB:Q43733 ；GB:Q93YM9)

similar to disease resistance-responsive family protein / dirigent family protein 1.136144 0.160754 AT1G64160 [Arabidopsis thaliana] (TAIR:AT4G23690.1)

Identical to Two-component response regulator ARR16 (ARR16) [Arabidopsis Thaliana] -1.13471 0.023771 AT2G40670

(GB:Q9SHC2) ..

-1.13145 0.00589 AT1G75250 similar to myb family transcription factor [Arabidopsis thaliana] (TAIR:AT1G19510.1);

similar to END04 (ENDONUCLEASE 4), T/G mismatch-specific endonuclease/

1.130446 0.11548 ATI G11190 endonuclease/ nucleic acid binding / single-stranded DNA specific

endodeoxyribonuclease [Arabidopsis thaliana] (TAIR:AT4G21585.1)

Identical to Non-symbiotic hemoglobin 2 (AHB2) [Arabidopsis Thaliana] -1.1291 0.110966 AT3G10520 (GB:024521);

similar to protein phosphatase 2C, putative / PP2C, putative [Arabidopsis thaliana] -1.12825 0.152118 AT1G07160 (TAIR:AT2G30020.1)

similar to ANAC048 (Arabidopsis NAC domain containing protein 48), transcription 1.12667 0.178793 AT1G02220

factor [Arabidopsis thaliana] (TAIR:AT3G04420.1)

1.126591 0.041416 AT5G40730 Identical to Arabinogalactan peptide 24 precursor (AGP24) [Arabidopsis Thaliana]

1.126224 0.111984 AT3G48020 similar to F-box family protein-related [Arabidopsis thaliana] (TAIR: AT5G62860.1)

-1.12464 0.056249 AT3G48720 similar to transferase family protein [Arabidopsis thaliana] (TAIR:AT5G63560.1)

similar to RNA polymerase II mediator complex protein-related [Arabidopsis thaliana] 1.123034 0.006101 AT1G26665 (TAIR:AT5G41910.1)

1.122109 0.173559 AT5G65300

-1.12148 0.000614 AT5G67370 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G 11840.1)

similar to trehalose-6-phosphate phosphatase, putative [Arabidopsis thaliana] -1.12026 0.12159 AT2G22190 (TAIR:AT4G39770.1)

similar to glycosyl hydrolase family 1 protein [Arabidopsis thaliana] 1.120044 0.124304 AT2G44480 (TAIR:AT2G44450.1);

Identical to Cytochrome P450 71A12 (CYP71A12) [Arabidopsis Thalianal 1.117436 0.22654 AT2G30750 (GB:049340)

-1.11534 0.070936 AT3G48100 Identical to Two-component response regulator ARR5 (ARR5) [Arabidopsis Thaliana]

89 (GB:Q9SB04;GB :080364;GB:Q9SBH9)

similar to amino acid transporter family protein [Arabidopsis thaliana] -1.11055 0.007087 AT5G02180 (TAIR:AT3G09330.1);

1.108942 0.018245 AT4G39675

Identical to Extensin-3 precursor (EXT3) [Arabidopsis Thaliana] 1.107223 0.252893 AT1G21310 (GB:Q9FS16;GB:Q9FS14;GB:Q9LMP2)

similar to ATMLP-300B (MYROSINASE-BINDING PROTEIN-LIKE PROTEIN-300B) 1.106275 0.166507 AT3G16450 [Arabidopsis thaliana] (TAIR:AT3G16440.1)

similar to UGT76C2 (UDP-glucosyl transferase 76C2), UDP-glycosyltransferase/ 1.101714 0.09327 AT3G11340

transferase, transferring glycosyl groups [Arabidopsis thaliana] (TAIR:AT5G05860.1)

similar to NDA2 (ALTERNATIVE NAD(P)H DEHYDROGENASE 2), NADH -1.10057 0.113643 AT1G07180 dehydrogenase [Arabidopsis thaliana] (TAIR:AT2G29990.1)

Identical to Myrosinase-binding protein-like Atlg52030 (F-ATMBP) [Arabidopsis 1.099603 0.071947 AT1G52030 Thaliana] (GB:Q9SAV 1;GB:023831 ；GB:Q9SSV0;GB:Q9SXH0)

similar to EXS family protein / ERDl/XPRl/SYGl family protein [Arabidopsis thaliana] 1.099577 0.137151 AT1G69480 (TAIR:AT1G26730.1)

Identical to Probable flavin-containing monooxygenase 1 (FMOl) [Arabidopsis -1.09955 0.425532 ATI G19250 Thaliana] (GB:Q9LMA1);

similar to meprin and TRAP homology domain-containing protein / MATH -1.09914 0.000211 AT2G32880

domain-containing protein [Arabidopsis thaliana] (TAIR:AT2G32870.1)

-1 09906 0.083557 AT1G22590 contains domain PTHR11945 (PTHR11945) Identical to Probable WRKY transcription factor 61 (WRKY61) [Arabidopsis Thaliana] 1.097917 0.112794 ATI G18860

(GB:Q8VWV6;GB:Q9M9V5)

Identical to Cytokinin dehydrogenase 4 precursor (CKX4) [Arabidopsis Thaliana]

-1.0967 0.059451 AT4G29740 (GB:Q9FUJ2;GB:Q9SU77); similar to CKX2 (CYTOKININ OXIDASE 2), cytokinin

dehydrogenase [Arabidopsis thaliana] (TAIR:AT2G 19500.1)

similar to CIPK18 (SNFl-RELATED PROTEIN KINASE 3.20)，kinase [Arabidopsis 1.09531 0.215868 AT2G34180 thaliana] (TAIR:AT1G29230.1)

similar to unnamed protein product [Vitis vinifera] (GB:CA042101.1); contains domain -1.09484 0.001375 AT1G29700 SSF56281(SSF56281)

Identical to Heat stress transcription factor B-1 (HSFBl) [Arabidopsis Thaliana] 1.093441 0.16762 AT4G36990 (GB:Q96320;GB:023186)

Identical to UPF0496 protein At3g28290/At3g28300 [Arabidopsis Thaliana] 1.092103 0.399029 AT3G28290 (GB:Q9XF11 ；GB:Q56XE6;GB:Q8GXJ7)

Identical to COBRA-like protein 4 precursor (C0BL4) [Arabidopsis Thaliana] 1.090446 0.207931 AT5G15630 (GB:Q9LFW3 ；GB :Q6ICX3)

Identical to Peroxidase 30 precursor (PER30) [Arabidopsis Thaliana] 1.087856 0.010118 AT3G21770 (GB:Q9LSY7;GB:Q43737;GB:Q96521)

90 1.087833 0.100526 AT3G19660 similar to unnamed protein product [Vitis vinifera] (GB:CA049873.1)

Identical to Putative phosphoethanolamine N-methyltransferase 2 (NMT2) [Arabidopsis -1.08675 0.006551 AT1G48600 Thaliana] (GB:Q944H0;GB:Q9LP63;GB:Q9LP64)

similar to short-chain dehydrogenase/reductase (SDR) family protein [Arabidopsis -1.08366 0.018314 AT5G02540 thaliana] (TAIR:AT2G37540.1)

similar to photoassimilate-responsive protein-related [Arabidopsis thaliana] 1.083175 0.09925 AT5G52390 (TAIR:AT3G54040.1)

1.077696 0.459435 AT1G07050 similar to CIL [Arabidopsis thaliana] (TAIR:AT4G25990.1)

1.077149 0.273242 AT1G15010 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT2G01300.1)

similar to oxidoreductase, 20G-Fe(II) oxygenase family protein [Arabidopsis thaliana] 1.076572 0.015178 AT2G38240 (TAIR:AT5G05600.1)

1.075626 0.033238 AT2G30550 similar to lipase class 3 family protein [Arabidopsis thaliana] (TAIR:AT1G06800.1)

similar to FAD-binding domain-containing protein [Arabidopsis thaliana] 1.075552 0.396226 AT1G30720 (TAIR:AT1G30730.1) ‘ Identical to 5'-adenylylsulfate reductase 2, chloroplast precursor (APR2) [Arabidopsis -1.07522 0.00117 AT1G62180

Thaliana] (GB:P92981;GB:004215;GB:004583;GB:022554;GB:Q38947;GB:Q541D4)

1.070665 0.039283 AT2G47550 similar to ATPMEPCRB, pectinesterase [Arabidopsis thaliana] (TAIR:AT4G02330.1

-1.06979 0.032149 AT4G39770 similar to trehalose-phosphatase [Arabidopsis thaliana] (TAIR:AT2G22190.1)

similar to acid phosphatase class B family protein [Arabidopsis thaliana] -1.067 0.014558 AT5G44020 (TAIR:AT1G04040.1)

similar to ATGSTU13 (GLUTATHIONE S-TRANSFERASE 12)，glutathione transferase 1.064258 0.103903 AT1G69920 [Arabidopsis thaliana] (TAIR:AT1G27130.1)

Identical to 1 -aminocyclopropane-1 -carboxylate oxidase homolog 1 [Arabidopsis 1.061228 0.049947 AT1G06620 Thaliana] (GB:Q84MB3;GB:Q9SHK4)

1.059651 0.03595 AT1G23040 similar to proline-rich family protein [Arabidopsis thaliana] (TAIR:AT1G70990.1)

similar to YLS2 (yellow-leaf-specific gene 2), strictosidine synthase [Arabidopsis 1.05929 0.280634 AT3G51440 thaliana] (TAIR:AT3G51430.1)

similar to ATPTR2-B (NITRATE TRANSPORTER 1), transporter [Arabidopsis thaliana] -1.05881 0.002569 AT2G02020 (TAIR:AT2G02040.1)

Identical to Nuclear transcription factor Y subunit A-2 (NFYA2) [Arabidopsis Thaliana] 1.057554 0.155191 AT3G05690 (GB:Q9M9X4;GB:023631)

Identical to Beta-glucosidase homolog precursor (BGl) [Arabidopsis Thaliana] 1.055437 0.021128 AT1G52400

(GB:Q9SE50;GB:Q93V63)

1.053324 0.115674 AT3G23090 similar to WDLl (WVD2-LIKE 1) [Arabidopsis thaliana] (TAIR:AT3G04630.2)

-1.05331 0.000508 AT1G43650 similar to nodulin MtN21 family protein [Arabidopsis thaliana] (TAIR:AT5G64700.1)

1.052628 0.059027 AT2G37760 similar to aldo/keto reductase family protein [Arabidopsis thaliana] (TAIR: AT2G37790.1)

Identical to Glutamate decarboxylase 1 (GADl) [Arabidopsis Thaliana] 1.052112 0.05755 AT5G17330 (GB:Q42521;GB:Q9FFH9)

91 Identical to Photosystem I P700 chlorophyll a apoprotein Al (psaA) [Arabidopsis -1.05159 0.060491 ATCG00350

Thaliana] (GB:P56766;GB:Q31852)

1.051342 0.04603 AT1G02470 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT1G02475.1)

Identical to Nuclear transcription factor Y subunit A-3 (NFYA3) [Arabidopsis Thaliana] 1.049724 0.141117 AT1G72830 (GB:Q93ZH2;GB:023632;GB:Q9SSP3)

1.048267 0.322821 AT2G35980 similar to NHL2 (NDRl/HINl-like 2) [Arabidopsis thaliana] (TAIR:AT3G11650.1)

1.046598 0.367264 AT1G73010 similar to phosphoric monoester hydrolase [Arabidopsis thaliana] (TAIR:AT1G17710.1)

similar to RLM1 (RESISTANCE TO LEPTOSPHAERIA MACULANS 1), ATP binding I -1.04653 0.005499 AT1G56510

protein binding / transmembrane receptor [Arabidopsis thaliana] (TAIR:ATIG64070.1)

similar to FAD-binding domain-containing protein [Arabidopsis thaliana] 1.046516 0.437619 AT1G26410 (TAIR:AT1G26380.1) similar to mitochondrial substrate carrier family protein [Arabidopsis thaliana] -1.04589 0.118724 AT5G26200 (TAIR:AT1G72820.1)

similar to ATFRUCT5 (BETA-FRUCTOFURANOSIDASE 5)，hydrolase, hydrolyzing 1.043069 0.367502 AT5G11920 0-glycosyl compounds / levanase [Arabidopsis thaliana] (TAIR:AT1G55120.1)

-1.03902 0.107007 AT1G02813 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G02360.1)

Identical to Cysteine-rich receptor-like protein kinase 21 precursor (CRK21) -1.03845 0.0663 AT4G23290

[Arabidopsis Thaliana] (GB:Q3E9X6;GB:065480;GB:Q8LPL4)

1.037915 0.121368 AT3G25640 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G23100.1)

1.037447 0.056547 AT5G53870 similar to plastocyanin-like domain-containing protein [Arabidopsis thaliana]

1.036022 0.045802 AT3G55910 -1.03582 0.001402 ATI G19510 similar to myb family transcription factor [Arabidopsis thaliana] (TAIR:ATIG75250.2);

similar to cyclopropane fatty acid synthase-related [Arabidopsis thaliana] -1.03572 0.157444 AT3G23530

(TAIR:AT3G23480.1)

1-033544 0.017915 AT2G25625 similar to unnamed protein product [Vitis vinifera] (GB:CA044157.1)

similar to ATPTR2-B (NITRATE TRANSPORTER 1), transporter [Arabidopsis thaliana] -1.02944 0.001182 AT5G01180

(TAIR:AT2G02040.1)

1 028188 0.031454 AT1G33800 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G09990.1)

1.027839 0.388302 AT4G27652 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT4G27657.1)

similar to CATS (CATIONIC AMINO ACID TRANSPORTER 5), cationic amino acid 1.027422 0.375395 AT4G21120

transmembrane transporter [Arabidopsis thaliana] (TAIR:AT2G34960.1)

-1.02675 0.082327 AT3G10720 similar to pectinesterase, putative [Arabidopsis thaliana] (TAIR:AT5G04970.1)

similar to leucine-rich repeat protein kinase, putative [Arabidopsis thaliana] -1.02567 0.00447 AT3G46370 (TAIR:AT3G46340.1) Identical to Heat shock protein 81-1 (HSP81-1) [Arabidopsis Thaliana] 1.024568 0.123177 AT5G52640 (GB:P27323;GB:Q03930;GB:Q96268;GB:Q9LTF3)

similar to TMAC2 (TWO OR MORE ABRES-CONTAINING GENE 2) [Arabidopsis 1.024252 0.013237 AT3G29575 thaliana] (TAIR:AT3G02140.1)

92 -1.02424 0.069251 AT1G62290 similar to aspartyl protease family protein [Arabidopsis thaliana] (TAIR:AT4G04460.1)

Identical to Glutamate receptor 1.3 precursor (GLR1.3) [Arabidopsis Thaliana] 1.023639 0.154841 AT5G48410 (GB:Q9FH75)

-1.0233 0.005728 AT3G48460 similar to lipase, putative [Arabidopsis thaliana] (TAIR:AT1G28650.1)

-1.01361 0.017953 AT2G04039 similar to unnamed protein product [Vitis vinifera] (GB:CA041070.1)

similar to RGLGl (RING DOMAIN LIGASEl), protein binding / zinc ion binding 1.012751 0.046338 AT5G63970 [Arabidopsis thaliana] (TAIR:AT3G01650.1)

similar to acetylomithine aminotransferase, mitochondrial, putative / acetylomithine

1.012571 0.120143 AT5G46180 transaminase, putative / AOTA, putative / ACOAT, putative [Arabidopsis thaliana]

(TAIR:AT1G80600.1)

-1.01209 0.064822 AT2G15020 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT5G64190.1)

Identical to Two-component response regulator ARR7 (ARR7) [Arabidopsis Thaliana] -1.01202 0.105202 ATI G19050

(GB:Q9ZWS7)

1.01071 0.049048 AT3G14990 similar to DJ-1 family protein [Arabidopsis thaliana] (TAIR:AT1G53280.1)

1.010551 0.12076 AT1G76790 similar to 0-methyltransferase, putative [Arabidopsis thaliana] (TAIR:AT1G21100.1)

1.008934 0.033442 AT5G66170 similar to unknown protein [Arabidopsis thaliana] (TAIR:AT2G17850.1)

similar to UbiE/C0Q5 methyltransferase family protein [Arabidopsis thaliana] -1.00571 0.117955 AT1G69523 (TAIR:AT1G69526.2) Identical to Fasciclin-like arabinogalactan protein 12 precursor (FLA 12) [Arabidopsis 1.003692 0.271019 AT5G60490 Thaliana] (GB:Q8LEE9;GB:Q94EG7；GB:Q9FKJ8)

Identical to 17.4 kDa class I heat shock protein (HSP17.4) [Arabidopsis Thaliana] 1.002616 0.094819 AT3G46230 (GB:P19036;GB:Q540N8;GB:Q94EJ5;GB:Q9LX72)

similar to 0s01g0927300 [Oryza sativa (japonica cultivar-group)] 1.002538 0.034472 AT2G35730 (GB:NP_001045267.1);

1.001283 0.019786 AT3G55890 Identical to Protein yippee-like At3g55890 [Arabidopsis Thaliana] (GB:Q9LY56)

similar to glycogenin glucosyltransferase (glycogenin)-related [Arabidopsis thaliana] -1.00113 0.233807 AT2G35710

(TAIR:AT4G16600.1)

1.001023 0.003276 AT1G08320 similar to bZIP family transcription factor [Arabidopsis thaliana] (TAIR:AT5G06839.1);

1.000994 0.038491 AT5G25770 similar to unnamed protein product [Vitis vinifera] (GB:CA044054.1)

93 Appendix F. Log fold changes of terpenoid pathway genes involved in FPS mutants

Fpsl :WT Fps2 :WT Locus Annotation P.Value adj.P.Val P.Value adj.P.Val logFC logFC

AT5G16440 IPPl 0.14005 0.477893 i 0.043251 0.8221840.983162

AT3G02780 IPP2 -0.12886 0.214587 1 -0.29305 0.023391 0.817061

AT5G47770 FPSl 1.06588 0.00196 1 0.02382 0.899146 0.990951

AT4G17190 FPS2 0.12104 0.776147 1 1.515266 0.013422 0.801339

AT4G36810 GGPSl -0.47369 0.043302 1 -0.15889 0.40722 0.921514

AT2G18620 similar to GGPS1 -0.01892 0.856134 1 0.017269 0.868552 0.98693

AT3G14510 similar to GGPS3 -0.11867 0.269299 1 -0.24814 0.048702 0.825274

AT3G29420 similar to GGPS3 -0.0286 0.784262 1 -0.0203 0.845586 0.984488

AT3G14530 similar to GGPS3 -0.02538 0.898869 1 -0.14374 0.483292 0.934685

AT3G20160 similar to GGPS6 -0.09326 0.404793 1 0.045757 0.673988 0.962536

AT3G32040 similar to GGPP synthetase -0.12681 0.217912 1 0.015914 0.866473 0.986486

AT3G29430 similar to GGPP synthetase -0.02779 0.779613 丨 -0.06056 0.548323 0.948911

AT2G23800 GGPP2 0.019006 0.942235 1 -0.09662 0.71453 0.970145

AT2G18640 GGPP4 -0.08293 0.424788 1 -0.03936 0.697107 0.966488

AT1G49530 GGPP6 0.19783 0.206938 1 0.165974 0.278169 0.888505

AT4G34640 SQSl -0.27438 0.05351 1 -0.10731 0.369608 0.916657

AT4G34650 similar to SQSl -0.1533 0.520516 1 -0.04584 0.844505 0.984224

AT5G24150 SQPl.l -1.23187 0.005172 1 -0.49185 0.11604 0.862445

AT5G24160 SQPl,2 0.492648 0.488327 1 1.981107 0.03019 0.819184

AT5G24140 SQP2 0.065872 0.500129 1 0.06421 0.510498 0.939111

similar to XFl (SQUALENE AT2G22830 0.058483 0.864269 1 -0.71741 0.078828 0.84591 EPOXIDASE 1)

similar to XFl (SQUALENE AT4G37760 -0.59065 0.025154 1 -0.59562 0.02442 0.817061 EPOXIDASE 1)

similar to SQE3 (SQUALENE AT1G58440 -0.17651 0.432143 1 -0.35449 0.147331 0.863364 EPOXIDASE 3)

Dehydrodolichyl diphosphate

AT5G60500 synthase 7 -0.22358 0.261211 1 -0.23256 0.244959 0.884391

UPP synthetase family protein

Dehydrodolichyl diphosphate AT2G17570 0.222934 0.386943 1 0.250308 0.336157 0.912004 synthase 6

94 UPP synthetase family protein

similar to SS2

AT1G74010 (STRICTOSIDINE 0.159753 0.668515 1 0.919779 0.047671 0.825274

SYNTHASE 2)

similar to YLS2

AT3G51420 (yellow-leaf-specific gene 2)， -0.03525 0.886509 1 -0.35985 0.186275 0.875166

strictosidine synthase

similar to YLS2

AT3G51450 (yellow-leaf-specific gene 2)， 1.111148 0.038231 1 0.485877 0.275193 0.888505

strictosidine synthase

similar to YLS2

AT3G51440 (yellow-leaf-specific gene 2)， 0.61687 0.512564 1 1.05929 0.280634 0.888588

strictosidine synthase

similar to strictosidine synthase .. AT3G57030 -0.1671 0.143736 1 -0.02248 0.824703 0.983162 family protein

similar to strictosidine synthase AT3G51430 0.091567 0.834737 1 0.544666 0.248345 0.884391 family protein

similar to strictosidine synthase AT2G41300 -0.06485 0.655986 1 0.054951 0.704962 0.967695 family protein

similar to strictosidine synthase AT3G57010 -0.04906 0.91319 丨 0.510208 0.286927 0.893606 family protein

similar to strictosidine synthase AT1G08470 0.006802 0.9773 1 -0.18569 0.451526 0.929617 family protein

similar to strictosidine synthase AT3G57020 -0.92468 0.116499 1 -0.20966 0.684758 0.964677 family protein

similar to strictosidine synthase AT2G41290 -0.75399 0.029594 1 -0.25485 0.353718 0.913039 family protein

similar to strictosidine synthase AT3G59530 -0.29764 0.318259 1 0.02599 0.926629 0.99186 family protein

Strictosidine synthase 1 AT1G74020 0.400339 0.401129 1 1.316338 0.02986 0.819184 precursor (SSI)

Strictosidine synthase 3 AT1G74000 -0.00945 0.923522 1 0.085891 0.401169 0.921134 precursor (SS3)

similar to ATTPS-CIN

(TERPENE AT4G16740 0.111922 0.807396 1 0.612663 0.218813 0.881699 SYNTHASE-LIKE

SEQUENCE-1,8-CINEOLE)

AT2G24210 similar to ATTPS-CIN 0.068137 0.66503 1 -0.1481 0.363753 0.91582

95 (TERPENE

SYNTHASE-LIKE

SEQUENCE-1,8-CINEOLE)

similar to ATTPS-CIN

(TERPENE AT3G25810 0 1 1 0.030741 0.768239 0.975318 SYNTHASE-LIKE

SEQUENCE-1,8-CINEOLE)

similar to ATTPS-CIN

(TERPENE AT3G25820 -0.05766 0.574693 0.953628 SYNTHASE-LIKE

SEQUENCE-1,8-CINEOLE)

AT5G23960 Group A sesquiterpene cyclase -0.17141 0.187924 1 -0.20918 0.122068 0.862445

AT5G44630 Group B sesquiterpene cyclase -0.2494 0.686899 1 -0.38603 0.53763 0.947108

similar to ACL ..

AT2G23620 (ACETONE-CYANOHYDRIN -0.10336 0.796982 1 0.548168 0.210077 0.880154

LYASE), hydrolase

Phytoene synthase, chloroplast AT5G17230 -0.63315 0.009658 1 -0.41743 0.043432 0.825274 precursor (PSYl) Ent-kaurene synthase A, AT4G02780 -0.11392 0.316512 1 -0.08123 0.463458 0.931773 chloroplast precursor (GA1)

Identical to Ent-kaurene

AT1G79460 synthase B, chloroplast 0.408748 0.454785 0.93019

precursor (GA2)

similar to GA2 (GA

ATI G61120 REQUIRING 2), ent-kaurene 0.095336 0.792778 0.977867

synthase

AT2G07050 CASl -0.17422 0.262128 1 -0.34644 0.053958 0.830487

AT1G78970 similar to ATLUP2 (LUPl) 0.390625 0.585582 1 0.67394 0.361276 0.915193

96 c:,/ ,^ ’ 、、气 V ^ \ V 女、〜 - ‘> ； . . \

..... f

. . <

. ...

. / r •• • ., •

. CUHK Libraries 圏_1111 004659949