Cyanide Metabolism, Postharvest Physiological Deterioration and Abiotic Stress Tolerance in Cassava (Manihot Esculenta Crantz)

Home , Cyanase, Linamarase, Linamarin

Cyanide Metabolism, Postharvest Physiological Deterioration and Abiotic Stress Tolerance in Cassava (Manihot esculenta Crantz)

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Tawanda Zidenga

Graduate Program in Plant Cellular and Molecular Biology

The Ohio State University

2011

Dissertation Committee:

Professor Richard Sayre (Adviser)

Professor Rebecca S. Lamb

Professor Jyan-Chyun Jang

Professor Randall Scholl

Tawanda Zidenga

2011

Abstract

Cassava (Manihot esculenta Crantz) is one of the six most important crops in the world, and an important staple for more than 800 million people. Its tolerance to drought and poor soils make an important food security crop especially in sub-Saharan Africa, while its high starch content has made it attractive as a biofuel feedstock. However, cassava produces potentially toxic levels of cyanogenic glycosides and undergoes rapid postharvest physiological deterioration (PPD) within 72 hours of harvest, shortening its shelf-life. The major cyanogenic glycoside in cassava is linamarin (95%). Linamarin is made from valine in the leaves and translocated to the roots where it is believed to contribute to the reduced nitrogen pool of the plant. Linamarin in the roots is hydrolyzed during mechanical damage by linamarase to acetone cyanohydrin, which breaks down spontaneously into cyanide and acetone at pH > 5.0 or temperatures > 35°C. The first objective of this project was to analyze the activities of enzymes involved in nitrogen and cyanide metabolism in cassava in order to understand the effect of cyanogenic potential on nitrogen metabolism. We compared the activity of nitrate reductase, a key enzyme in nitrogen metabolism, in wild-type lines and transgenic low cyanogen lines. The results show that nitrate reductase activity is 3X higher in low cyanogen plants than in wild-type lines, suggesting that cyanogenesis provides a significant source of reduced nitrogen which has to be compensated for by nitrate reduction in low cyanogen plants. In addition,

β-cyanoalanine synthase (CAS), nitrilase and rhodanese assays confirmed that CAS is the ii key cyanide metabolizing enzyme in cassava roots, with 3X more activity in the roots compared to shoots, while rhodanese had no detectable activity in cassava roots. We concluded from these assays that the β-cyanoalanine synthase pathway is active in cassava roots, and is likely to be the route for assimilation of cyanide into free amino acids. The second objective was to investigate the potential of redirecting cyanide from cyanogenesis in cassava roots to the production of free amino acids by overexpressing two genes involved in cyanide metabolism; CAS and nitrilase 4 (NIT4). Transgenic plants generated using each of these strategies had limitations because of effects on other pathways, which will be discussed. The third objective was to investigate the role of cyanogenesis in oxidative stress and PPD in cassava. PPD in cassava is marked by an initial oxidative burst, followed by vascular discoloration, ultimately rendering the crop unpalatable. Effective transgenic strategies for extending the shelf-life of cassava require an understanding of the causes of the early events, especially the oxidative burst. We show a causal link between cyanogenesis, which occurs upon mechanical damage in cassava, and accumulation of reactive oxygen species which trigger PPD. By measuring

ROS accumulation in transgenic low cyanogen plants and as well as biochemically complementing the low cyanogens by adding 5 mM potassium cyanide, we show that the oxidative burst in cassava roots is cyanogen-induced. In light of these data, we have generated transgenic plants expressing codon-optimized Arabidopsis thaliana Alternate

Oxidase (AOX), a cyanide resistant terminal oxidase in the mitochondrial respiratory chain in plants, as a strategy to control PPD by reducing the cyanide-induced ROS accumulation. The transgenic AOX plants show a 10-14X reduction in ROS

iii accumulation compared to wild type plants, with at least two week extension in shelf– life. The final objective was to test these AOX transgenic lines for tolerance to abiotic stress. The production of ROS is one of the earliest detectable responses of plants to abiotic stresses. Since AOX blocks ROS accumulation, we tested the hypothesis that

AOX overexpression protects cassava plants from abiotic stress. Our results show marked tolerance to waterlogging and salt stress, suggesting AOX as a strategy to extend shelf-life of cassava while providing stress tolerance.

For my mother.

Acknowledgments

I received help and support from many people during the course of this project and it will be impossible to list them all here, so first I must say thanks to everyone I interacted with for the support and encouragement. I am grateful especially to my advisor Dr. Richard

Sayre for the continued guidance throughout the project, for the opportunity to be part of a great program (BioCassava Plus) and for his patience and encouragement; To my committee members; Dr. Rebecca Lamb, Dr. Randy Scholl and Dr. JC Jang for their support, suggestions and willingness to accommodate me in their schedules.

I also want to thank my fellow graduate students in the lab; Anil Kumar, Zoee Perrine and Elisa Leyva-Guerrero with whom I shared many lively discussions. Thanks to the

Sayre lab members past and present who have given their assistance and encouragement;

Uzoma Ihemere, Sathish Rajamani, Vanessa Falcao, Shayani Peris and Narayanan

Narayanan. Thanks also to Hangsik Moon, who was involved in the early part of the

AOX project; Dimuth Siritunga, whose work on cyanogenesis my project rested upon;

Matt Stephens, Mary-Ann Abiado, Shantha Peris and Jennifer Norris, who, at different stages kept the lab working. And to the interns I have worked with during my years in graduate school; Tony Galleinstein, Reid Rice and Solomon Afuape, not only for the help vi they offered in parts of my project, but for challenging me to refine my understanding of biology as I worked with them. I would also like to thank all the people in the facilities I have used especially the greenhouse stuff at Ohio State department of Plant Cellular and

Molecular Biology (PCMB) and the Donald Danforth Plant Science Center (DDPSC),

Microscopy Facilities at Ohio State and DDPSC.

vii

Vita

August 1977 ...... Born, Gweru, Zimbabwe

2001...... B.S. Agriculture, University of Zimbabwe

2003...... Visiting Scholar, The Ohio State University

2004 to present ...... Graduate Research Associate, Department

of Plant Cellular and Molecular Biology,

The Ohio State University

2008 to present ...... Visiting Graduate Student, Donald Danforth

Plant Science Center, St Louis, MO

Fields of Study

Major Field: Plant Cellular and Molecular Biology

viii

Table of Contents

Abstract ...... ii

Acknowledgments...... vi

Vita ...... viii

List of Tables ...... xiv

List of Figures ...... xvi

Chapter 1: Introduction ...... 1

1.1 Cassava and food security ...... 1

1.2 Propagation...... 5

1.3 Cyanogenesis ...... 8

1.3.1 Sources of cyanide in plants ...... 8

1.3.2 Linamarin biosynthesis and breakdown ...... 11

1.3.3 Role of cyanogenic glycosides in plants ...... 13

1.3.4 Translocation of linamarin ...... 14

1.3.5 Dietary cyanide causes health disorders ...... 19

1.4 Postharvest physiological deterioration ...... 23

1.4.1 Cassava response to wounding in comparison with other crops...... 23

1.4.2 Biochemical events during cassava PPD ...... 25

1.4.3 Oxidative damage during PPD ...... 27

1.5 Strategies for controlling PPD...... 33

1.5.1 Storage conditions ...... 33

1.5.2 Cassava processing ...... 34

1.5.3 Plant Breeding ...... 38

1.6 Objectives ...... 38

Chapter 2: The role of cyanogenesis in cassava nitrogen metabolism ...... 41

2.1 Introduction ...... 41

2.2 Experimental procedures ...... 46

2.2.1 Cassava cultivars ...... 46

2.2.2 Tissue culture propagation of plant material ...... 47

2.2.3 Total protein extraction and analysis ...... 47

2.2.4 Activity of β-cyanoalanine synthase in cassava tissue ...... 48

2.2.5 Activity of Rhodanese...... 51

2.2.6 Determination of nitrilase activity ...... 54

2.2.7 Determination of nitrate reductase activity ...... 56

2.2.8 Transformation of Cassava with the β-cyanoalanine synthase and NIT4 ... 58

2.2.9 Plant growth in the greenhouse ...... 66

2.2.10 Measurement of yield parameters ...... 66

2.2.11 Free amino acid extraction and analysis ...... 67

2.2.12 Analysis of IAA ...... 67

2.2.13 Statistical analysis ...... 68

2.3 Results ...... 69

2.3.1 Biochemical analysis of cyanide metabolism in cassava ...... 69

2.3.2 Expression of β-cyanoalanine synthase in cassava ...... 77

2.3.3 Expression of Nitrilase 4 in cassava ...... 93

2.4 Discussion ...... 107

Chapter 3: The mechanism and control of postharvest physiological deterioration in cassava ...... 116

3.1 Introduction ...... 116

3.2 Experimental procedures ...... 119

3.2.1 Tissue culture propagation of plant material ...... 119

3.2.2 Detection of Reactive Oxygen Species by H2DCF-DA ...... 119

3.2.3 Detection of hydrogen peroxide in cassava roots ...... 120

3.2.4 Inhibition of NADPH oxidase and mitochondrial ETC ...... 120

3.2.5 Design of alternative oxidase construct for cassava transformation ...... 121

3.2.6 Regeneration and transformation of Cassava ...... 124

3.2.7 Cassava transformation using embryo cotyledons ...... 124

3.2.8 Induction of embryogenesis ...... 124

3.2.9 Co-cultivation with Agrobacterium ...... 124

3.2.10 Regeneration of plant tissue ...... 125

3.2.11 Cassava transformation using friable embryogenic callus ...... 125

3.2.12 RT-PCR analysis of transgenic plants ...... 126

3.2.13 Plant growth in the greenhouse ...... 128

3.2.14 Determination of the alternative oxidase capacity in cassava roots ...... 128

3.2.15 Evaluation of postharvest physiological deterioration in cassava roots ... 130

3.2.16 Measurement of PPD using UV fluorescence ...... 130

3.2.17 Measurement of yield parameters ...... 130

3.2.18 Stress treatments ...... 131

3.2.19 Statistical analysis ...... 132

3.3 Results ...... 133

3.3.1 The oxidative burst in damaged cassava roots is cyanogen-induced ...... 133

3.3.2 The cyanide-induced oxidative burst in wounded cassava roots is of

mitochondrial origin ...... 138

3.3.3 Generation of AOX transgenic lines ...... 141

xii

3.3.4 Expression of Arabidopsis alternative oxidase prevents ROS accumulation

in cassava roots ...... 143

3.3.5 Transgenic AOX plants have increased alternative oxidase capacity ...... 147

3.3.6 Expression of Arabidopsis alternative oxidase delays post-harvest

physiological deterioration in cassava storage roots ...... 149

3.3.7 Analysis of PPD using UV imaging ...... 154

3.3.8 Yield parameters in AOX lines grown under greenhouse conditions ...... 156

3.3.9 Overexpression of AOX enhances tolerance to abiotic stress ...... 158

3.4 Discussion ...... 166

Chapter 4: Conclusions and Perspectives ...... 176

4.1 Introduction ...... 176

4.2 The role of cyanogens in nitrogen metabolism ...... 177

4.3 The potential role for β-cyanoalanine synthase and cyanide in regulating root

development ...... 185

4.4 A cyanogen-induced oxidative burst initiates postharvest physiological

deterioration (PPD) in cassava ...... 185

4.5 Over-expression of Alternative Oxidase (AOX) extends cassava shelf-life .... 186

4.6 Over-expression of Alternative Oxidase controls abiotic stress ...... 187

References ...... 190

xiii

List of Tables

Table 1.1. Food Sources of Cyanogenic Glycosides ...... 22

Table 1.2. Foods made from cassava...... 35

Table 1.3. Efficiency of common cassava processing techniques ...... 37

Table 2.1. Rhodanese and β-cyanoalanine synthase activity in cassava...... 71

Table 2.2. Effect of CAS overexpression on root development...... 84

Table 2.3. Effect of CAS overexpression on β-cyanoalanine synthase rate...... 86

Table 2.4. Effect of CAS overexpression on yield...... 88

Table 2.5. Effect of CAS overexpression on amino acid content ...... 90

Table 2.6. Effect of CAS overexpression on total protein...... 92

Table 2.7. Branching in Nitrilase plants...... 99

Table 2.8. IAA analysis in wild-type and transgenic cassava roots and leaves...... 101

Table 2.9. Effect of NIT4 overexpression on tuber fresh weight in cassava ...... 103

Table 2.10. Total amino acid and protein analysis in wild-type and NIT4 transgenic lines.

...... 105

Table 3.1. Flourescence in wild-type and Cab1 plants as quantified by ImageJ software.

...... 135

Table 3.2. ROS-induced fluorescence in wild-type and transgenic AOX plants ...... 146

Table 3.3. Delayed PPD in transgenic AOX lines after 21 days discoloration ...... 152

Table 3.4. Measurement of yield parameters in transgenic AOX plants ...... 157 xiv

Table 3.5. Effect of waterlogging on tuber formation...... 161

Table 3.6. Effect of salt stress in wild-type and transgenic cassava plants...... 164

List of Figures

Figure 1.1. Major cassava producing areas in 2006...... 3

Figure 1.2. Cassava crop in a field trial in Puerto Rico...... 5

Figure 1.3. Transverse section of the cassava root ...... 7

Figure 1.4. Cyanogenesis in plants...... 10

Figure 1.5. Linamarin biosynthesis and breakdown ...... 12

Figure 1.6. Current model of linamarin metabolism in cassava...... 16

Figure 1.7. Translocation of linamarin in cassava...... 18

Figure 1.8. Postharvest physiological deterioration in cassava roots ...... 26

Figure 1.9. A balance between ROS production and scavenging is necessary to avoid oxidative stress...... 29

Figure 1.10. The mitochondrial electron transport chain showing location of alternative oxidase...... 32

Figure 2.1. Standard curve used for estimating the concentration of hydrogen sulfide from

β-cyanoalanine synthase assays ...... 50

Figure 2.2. Standard curve used for estimating the concentration of thiocyanate in the rhodanese assay...... 53

Figure 2.3. Standard curve used to determine the amount of ammonia ...... 55

Figure 2.4. Standard curve used for estimating the concentration of nitrite ...... 57

xvi

Figure 2.5. Constructs design for β-cyanoalanine synthase and Nitrilase expression in cassava ...... 60

Figure 2.6. Primers used for checking the presence of β-cyanoalanine synthase (A) and

Nitrilase 4 (B) transgenic lines in putatively transformed cassava tissues...... 65

Figure 2.7. Nitrilase activity in cassava...... 74

Figure 2.8. Analysis of nitrate reductase activity in wild-type (WT) and low cyanogen

(Cab1-1, Cab1-2 and Cab1-3) lines ...... 76

Figure 2.9. Model of cyanide assimilation in plants...... 78

Figure 2.10. Recovery of putatively transgenic plants after transformation with the CAS gene...... 80

Figure 2.11. RT-PCR analysis of CAS expression ...... 81

Figure 2.12. In vitro growth analysis of transgenic β-cyanoalanine synthase plants

(PCAS1-4) and wild type (TMS 60444) plants after 3 weeks ...... 83

Figure 2.13. RT-PCR analysis of transgenic lines expressing Nitrilase 4...... 94

Figure 2.14. Expression of Nitrilase increases cyanoalanine hydrolase activity in cassava roots...... 96

Figure 2.15. Branching in cassava plants expressing the Nitrilase 4 gene ...... 99

Figure 2.16. Relationship between cyanide detoxification and cysteine biosynthesis in plants ...... 110

Figure 2.17. Analysis of amino acid sequence similarity between Arabidopsis NIT4

(AtNIT4) and Zea mays NIT2 (ZmNIT2)...... 112

Figure 3.1. Plasmid map of 3D-AtAox1A CO ...... 123

xvii

Figure 3.2. RT-PCR analysis of the transgenes...... 127

Figure 3.3. Illustration of submergence experiment setup in non-draining trays...... 131

Figure 3.4. Analysis of ROS production in low and high cyanogen plants...... 134

Figure 3.5. Biochemical complementation of low cyanide plants with 5 µM potassium cyanide (KCN) results in increases ROS production ...... 137

Figure 3.6. Inhibition of the plasma membrane NADPH oxidase ...... 139

Figure 3.7. RT-PCR analysis of transgenic AOX lines...... 142

Figure 3.8. Analysis of ROS accumulation in transgenic AOX lines...... 144

Figure 3.9. Alternative capacity in wild-type (WT) and transgenic plants overexpressing

AOX...... 148

Figure 3.10. Delayed PPD in transgenic plants expressing alternative oxidase ...... 150

Figure 3.11. Prolonged shelf-life in transgenic plants overexpressing AOX...... 153

Figure 3.12. Relationship between scopoletin fluorescence and PPD ...... 155

Figure 3.13. Effect of waterlogging on tuber development ...... 160

Figure 3.14. Effect of salt stress in transgenic and wild-type cassava...... 163

Figure 3.15. The mechanism and control of postharvest physiological deterioration in cassava ...... 171

Figure 3.16. Possible interaction of scopoletin and ROS (hydrogen peroxide) in the events leading to vascular discoloration ...... 173

Figure 4.1. Enzymes involved in cyanide metabolism in cassava roots ...... 179

Figure 4.2. Direct conversion of acetone cyanohydrin into ammonia without intermediate cyanide release...... 184

xviii

Chapter 1: Introduction

1.1 Cassava and food security

Cassava (Manihot esculenta Crantz) is a woody shrub of the Euphorbiaceae family grown mainly for its tuberous roots. It is a staple for more than 800 million people in the world and one of the six most important crops in the world (Lebot, 2009). In regions with erratic rainfall and limited input use, cassava is an important food security crop (Nweke et al., 2002; Ihemere et al. 2008). For this reason, cassava‟s importance as a food security crop has been recognized even in countries in which it is not the main staple. For example, in Malawi, cassava has been promoted for its value in supplementing maize in the event of crop failure (Rusike et. al., 2010). In addition, cassava roots can be left in the soil for up to 36 months, providing a household food bank in times of adversity (Lebot,

2009).

Africa accounts for most of the cassava production in the world with Nigeria being the major producer (Figure 1.1). In China and Thailand, cassava is increasingly being produced as an industrial crop. There is growing interest in the use of cassava as a biofuel feedstock. For example, the Guangxi Zhuangzu autonomous region in southern China planned to expand its ethanol production from 139 million liters in 2007 to 1.27 billion liters in 2010 using cassava as the primary feedstock (Drapcho et al., 2008). However,

Commercial production of cassava is impaired by several constraints, including susceptibility to the African Cassava Mosaic Virus (ACMV), low protein content, short shelf life and the presence of potentially toxic levels of cyanogenic glycosides (McMahon et al. 1995; Wenham, 1995; Siritunga and Sayre, 2003; Ihemere et al. 2008; Lebot, 2009).

These concerns, and the potential role of cassava in alleviating chronic malnutrition, have led to an increased interest in cassava improvement research utilizing biotechnology tools. Previously considered a neglected crop, cassava has recently received attention from researchers and donor agencies. Several foundations have supported research aimed at improving cassava for farmers in the tropics. One such program is the Bill and Melinda

Gates supported BioCassava Plus project, which focuses on improving various aspects of cassava using biotechnology, including iron and zinc bioavailability, protein and vitamin content, and reduction in postharvest physiological deterioration (BioCassava Plus, http://biocassavaplus.org).

Cassava Production Cassava Million metric tonnes metric Million 0

Brazil India Nigeria GhanaAngola Thailand Vietnam Indonesia Tanzania Mozambique Congo (ex Zaire) B

Figure 1.1. Major cassava producing areas in 2006. (A) by region as percentage of total production and (B) by country in million metric tons (B). Created from data presented in

Lebot, 2009.

Cassava belongs to the Dicotyledon family Euphorbiaceae, which is characterized by lactiferous vessels composed of secretory cells (Lokko et al. 2007). A shrub growing up to 1-4 m in height, cassava is monoecious, bearing both the pistillate and staminate flowers on the same plant (Alves, 2002; Lokko et al. 2007). Pistillate flowers open before staminate flowers of the same inflorescence. The resulting outcrossing in cassava leads to extremely heterozygous gene pools (Nassar, 2007). The cassava root is the major harvested part of the plant, although in some parts of Africa leaves are also consumed as food. The cassava tuberous root is actually not a tuber but a true root which cannot be used for vegetative propagation (Alves, 2002). The mature root (Figure 1.3) is made up of bark (periderm), peel (cortex) and parenchyma, which is the edible portion of the fresh root (Alves, 2002; Wheatley and Chuzel, 1993). Extremely thin vascular bundles ramify through the ﬂesh (Lebot, 2009). The periderm can be removed easily by scratching while the peel can be removed by peeling (Lebot, 2009). Dry matter is distributed between the leaves and roots depending on the stage of growth. In the first 60 to 75 days after planting, dry matter mostly accumulates in the leaves. After the fourth month, accumulation shifts to roots (Ihemere et al., 2008).

Figure 1.2. Cassava crop in a field trial in Puerto Rico.

1.2 Propagation

Cassava is mostly propagated via stem cuttings, although it can form seeds which are used in breeding programs. Plants started from true seeds take longer to develop and they are smaller and less vigorous than plants from stem cuttings (Alves, 2002). The advantage of using stem cuttings is that farmers do not have to reserve part of the yield

(storage roots) for propagation. The disadvantage of stem propagation is the limited multiplication rate between generations which is at least an order of magnitude less that if 5 seeds were used for multiplication. In addition, stems, being clonal, can transmit viruses to the next generation.

Stems selected for use in propagation should be sufficiently lignified, with no branching tips (Lebot, 2009). When grown from seed, cassava develops a taproot, which becomes

ﬁbrous and tuberous. When grown from cuttings, adventitious roots develop from the base of the cuttings within one week, into a ﬁbrous root system (Alves, 2002; Lebot,

2009).

A cassava plant can produce, on average, 10-20 cuttings, each 20-30cm long (Lebot,

2009). The International Center for Tropical Agriculture (CIAT) and the International

Institute of Tropical Agriculture (IITA) have developed methods for rapid propagation of planting material. These include a system based on two-node cuttings planted ﬂat, 1cm deep, in a well-drained substrate, under a screen to preserve sufﬁcient moisture (Lebot,

2009). Young shoots sprouting from these cuttings are then separated as soon as they have two leaves (8–10 cm long) and are placed in plastic tubes of 2 cm in diameter, ﬁlled with boiled water. They are transplanted into the ﬁeld after they develop roots. Another rapid propagation system is based on the use of ministem cuttings, small stem pieces, each with one or more nodes (Otoo, 1996). Sprouting takes place in well- drained nursery beds near a water source or in perforated black polythene bags filled with garden soil.

A method for accelerating propagation used in research facilities is the use of in vitro micropropagation of sub-nodal stems on Murashige and Skoog medium (Murashige and

Skoog, 1962). Meristem culture of cassava, developed by Kartha (1974) has been used as

6 a virus decontamination system to provide clean starting material for farmers (Ihemere,

2003). Tissue culture regenerated plantlets can be widely distributed in culture tubes but require an acclimatization phase during transfer from in vitro conditions to greenhouse conditions (Lebot, 2009).

Figure 1.3. Transverse section of the cassava root showing the periderm (bark), the peel

(cortex) and the parenchyma. Modified from Lebot, 2009. The parenchyma is the edible portion of the root.

1.3 Cyanogenesis

1.3.1 Sources of cyanide in plants

The potential toxicity of cassava is well known in cassava producing areas and has long been documented. In the Voyage of the Beagle (1839), Charles Darwin wrote of cassava in Brazil:

“Every part of this plant is useful: the leaves and stalks are eaten by the horses, and the roots are ground into a pulp, which, when pressed dry and baked, forms the farinha, the principal article of sustenance in the Brazils. It is a curious, though well-known fact, that the expressed juice of this most nutritious plant is highly poisonous. A few years ago a cow died at this Fazênda, in consequence of having drunk some of it.”

The toxicity is due to cassava‟s potential to release of cyanide upon degradation of cyanogenic glycosides, a group of widely occurring natural substances that on hydrolysis yield a ketone or aldehyde, a sugar, and the highly toxic cyanide ion

(Shibamoto and Bjeldanes, 1993; McMahon et al. 1995). Cyanide is known to be a potent cellular toxin. It achieves this toxicity by forming a very stable complex with active site metals (e.g., Fe and Mg) in target enzymes, inhibiting vital functions such as respiration, carbon fixation, and nitrate reduction (McMahon et al. 1995; Yip and Yang,

1998). All plants release cyanide as a byproduct of the ethylene biosynthesis pathway

(Peiser et al., 1984; Dong et al. 1992). Cyanogenic plants like cassava, bitter almonds and rubber have an additional (and major) source of cyanide: the breakdown of cyanogenic

8 glycosides (Figure 1.4) (Poulton, 1990). Cyanogenic glycosides are present in more than

2500 plant species (Vetter, 2000). A number of agronomically important crops including cassava, bitter almonds, rubber, lima beans and sorghum are cyanogenic (Table 1.1;

Shibamoto and Bjeldanes, 1993). Owing to the potential health problems associated with dietary exposure to cyanide, one of the major goals of research on cyanogenic crops is the generation of transgenic plants with reduced levels of cyanogens. Such projects require an understanding of the physiology and biochemistry of cyanogenesis in cassava. Recent work has generated a lot of information in this area (McMahon et al., 1995; Andersen et al., 2000; Siritunga and Sayre 2003; Siritunga et al. 2004).

The major cyanogenic glycoside in cassava is linamarin (95 %), with lotaustralin comprising the remaining portion. Cyanogenesis involves hydrolysis of linamarin to hydrogen cyanide (Poulton, 1990; McMahon et al. 1995). This occurs upon tissue disruption, when linamarin, localized in the vacuole, is brought into contact with linamarase, an enzyme found in the cell wall and laticifers (Mkpong et al., 1990; Elias et al. 1997). Cultivars with less than 100 mg linamarin/kg fresh weight are described as

„sweet‟ while cultivars with 100–500 mg linamarin/kg are „bitter‟ (Wheatley et al.,

1993). Cyanide released from cassava is potentially toxic in high amounts. The minimal lethal dose for humans is 0.5 – 3.5 mg per kg body weight (Shibamoto and Bjeldanes,

1993).

Figure 1.4. Cyanide release in plants. All plants release hydrogen cyanide (HCN) as a byproduct of ethylene biosynthesis. Cyanogenic plants release cyanide from the breakdown of cyanogenic glycosides. (ACC= 1-aminocyclopropane-1-carboxylic-acid).

1.3.2 Linamarin biosynthesis and breakdown

Linamarin is synthesized from the amino acid valine (Figure 1.5). The first dedicated step

(hydroxylation of valine to N-hydroxy valine) in this process is catalyzed by two very similar cytochrome P450s, CYP79D1 and CYP79D2 (Anderssen et al. 2000). Linamarin is produced from the glycosylation of acetone cyanohydrin by UDPG-glycosyltransferase

(Figure 1.5). Linamarin is stored in the vacuoles of all plants tissues while linamarase is found in the cell wall and laticifers (reviewed in McMahon et al., 1995). This compartmentalization means cyanogenesis occurs only during tissue disruption when the enzyme and substrate mix. Linamarin hydrolysis by linamarase yields acetone cyanohydrin and glucose (Figure 1.5) and the acetone cyanohydrin will spontaneously decompose to yield cyanide and acetone at pH > 5.0 or temperatures > 35°C.

Acetone cyanohydrin can also be broken down by the enzyme hydroxynitrile lyase

(HNL), but this enzyme is expressed only in leaves and stems and not in roots (White et al., 1998). Linamarin, acetone cyanohydrin and hydrogen cyanide are collectively called cyanogens (Lebot, 2009).

Figure 1.5. Linamarin biosynthesis and breakdown. Linamarin is synthesized from valine by two cytochrome P450s CYP79D1 and CYP79D2. Breakdown of linamarin releases cyanide. After Siritunga and Sayre, 2003.

1.3.3 Role of cyanogenic glycosides in plants

Cyanogenic plants release sufficient amounts of hydrogen cyanide to provide protection against herbivores and pathogens (Nahrstedt, 1985; Jones, 1998). For example, the larvae of Hypera postica was found to prefer the acyanogenic leaﬂets of white clover

Trifolium repens (Ellsbury et al., 1992) suggesting that the levels of cyanogenic glycosides could be selected in cassava to improve resistance to insect pests.

The efficacy of cyanogens as defense compounds has, however, been recently questioned

(Gleadow and Woodrow, 2002; Moller, 2010). Cyanogenic glycosides are not effective against all herbivores, and not all cyanogenic plants release enough cyanide to be considered toxic (Gleadow and Woodrow, 2002). In addition, some herbivores and pathogens have coevolved with plants to tolerate or even take advantage of cyanogenic glycosides (Moller, 2010). The bamboo lemur, which feeds mainly on the cyanogenic giant bamboo, has been found to consume 12 times above the lethal dose of most animals

(Glander et al., 1989).

Apart from their defense role, cyanogenic glycosides have been proposed to be transportable forms of reduced nitrogen (Selmar et al. 1988; Poulton, 1990). This was demonstrated by Selmar et al. (1988) in the rubber tree (Hevea brasiliensis). In the seed,

90% of the cyanogenic glycoside content is stored in the endosperm as linamarin. During germination, linamarin is converted into its diglycoside linustatin which is then transported over long distances to the seedling where it is assimilated into asparagine and aspartic acid in a reaction catalyzed by β-cyanoalanine synthase. 13

β-cyanoalanine synthase is a pyridoxal phosphate-dependent enzyme that converts cysteine and cyanide to hydrogen sulﬁde and cyanoalanine (Garcia et al. 2010).

Cyanoalanine is further converted by a nitrilase to asparagine, aspartate and ammonia

(Figure 1.6; Blumenthal et al., 1968; Elias et al, 1997; Hatzfeld et al. 2000; Lai et al.

2009) or conjugated to gamma-glutamyl-β-cyanoalanine (Yang, 1989). This pathway results in the assimilation of cyanide into primary metabolism and is available in all higher plants thus far examined (Blumenthal et al., 1968; Miller and Conn, 1980).

β-cyanoalanine synthase is active in all plants tested, since all higher plants produce cyanide as a byproduct of ethylene biosynthesis (Miller and Conn, 1980; Peiser et al.

1984). In addition, its activity correlates with cyanogenesis, with cyanogenic plants showing higher activities compared to non-cyanogenic plants (Miller and Conn, 1980).

Cyanide may also be detoxified via the enzyme rhodanese, a thiosulfate sulfurtransferase in the reaction

However, the role of rhodanese in cyanide detoxification in plants is controversial because it occurs less commonly in plants and its activity in plants does not seem to correlate with levels of cyanogens (Chew, 1973; Miller and Conn, 1980).

1.3.4 Translocation of linamarin

The discovery of genes encoding the cytochrome P450s (CYP79D1 and CYP79D2) that catalyze the ﬁrst-dedicated step in linamarin synthesis (Andersen et al., 2000) made

14 it possible to design a transgenic approach to reduce cyanogenesis in cassava. Siritunga and Sayre (2003, 2004) designed antisense constructs aimed at inhibiting the expression of CYP79D1 and CYP79D2 genes. In one construct, they used a leaf-specific cab1 promoter while in another they used a tuber-specific patatin promoter to express antisense constructs targeting knock down of CYP79D1 and CYP79D2 expression. Thus, they generated transgenic plants with linamarin biosynthesis inhibited in the leaves and the roots respectively. Cassava lines in which linamarin biosynthesis was inhibited in the roots had wild-type linamarin levels in the roots while those in which biosynthesis was inhibited in the leaves had a 99% reduction of linamarin in the roots (Siritunga and Sayre,

2003). This provided confirmation for the leaf as the primary site of synthesis for root linamarin. Previously, labeling studies of cassava by Bediako et al. (1981) with 14 C- valine showed that the primary site of linamarin synthesis is the leaf, while the roots and lower stem had minimal ability to incorporate labeled valine. Also, Ramanujam and

Indira (1984) showed a 13-fold decrease in root cyanogens when girdling was performed on the cassava stem. Makame et al. (1987) performed a series of grafting experiments between roots and shoots of low- and high-cyanogenic cultivars of cassava and the results showed that, at least in part, cyanogen accumulation in the roots was due to the transportation of cyanogens from the leaves. Taken together, the above studies confirm that linamarin is made in the leaves and transported to the roots (Figure 1.6).

Figure 1.6. Current model of linamarin metabolism in cassava. Linamarin is made in the leaves and transported to the roots where it is either stored or deglycosylated by the β- glucosidase linamarase to acetone cyanohydrin. Acetone cyanohydrin is unstable and can spontaneously decompose to release cyanide and a ketone. The cyanide can be assimilated into amino acids and protein via a pathway involving the enzyme β- cyanoalanine synthase and nitrilase.

The „linustatin pathway‟ describes the conversion of linamarin to its diglucoside linustatin, the form in which it is translocated in the phloem of some but not all cyanogenic plants (Selmar et al. 1988). This pathway has been demonstrated in rubber tree (Selmar, 1993). Some evidence for the translocation of linustatin instead of linamarin in cassava has been reported (Selmar, 1994). The proposed pathway involves conversion of linamarin to linustatin in the leaves, which is then transported via the apoplast (Figure

1.7). Linustatin cannot be deglycosylated by linamarase, thus it can pass through the apoplasm during phloem loading without being hydrolyzed (Selmar, 1994). After transport to the roots, linustatin is hydrolyzed by diglucosidases to linamarin (Selmar,

1994; Figure 1.7). Hydrolysis of linustatin can be achieved either „sequentially‟ or

„simultaneously‟ (Selmar, 2010; Figure 1.7). A simultaneous deglycosidase produces gentibiose and acetone cyanohydrin while a sequential deglycosidase converts linustatin to linamarin and then to acetone cyanohydrin (Selmar et al. 1988; Selmar, 2010).Based on this model, Selmar (1994) proposed that the overexpression of simultaneous diglucosidases would reduce accumulation of linamarin (thus reducing cassava toxicity) while enhancing protein content.

Figure 1.7. Translocation of linamarin in cassava. This pathway proposes that linamarin is converted to linustatin in the leaves, which is then transported via the apoplast. After transport to the roots, linustatin is hydrolyzed by diglucosidases to linamarin. Hydrolysis of linustatin can be achieved either „sequentially‟ or „simultaneously.‟ A simultaneous deglycosidase produces gentibiose and acetone cyanohydrin while a sequential deglycosidase converts linustatin to linamarin and then to acetone cyanohydrin. After

Selmar (1994). However, linustatin has not been detected in cassava.

1.3.5 Dietary cyanide causes health disorders

Exposure to cyanide can lead to health problems. Symptoms of acute cyanide poisoning include mental confusion, muscular paralysis, and respiratory distress (Shibamoto and

Bjeldanes, 1993). Cyanide disrupts cellular respiration by inhibiting cytochrome oxidase.

Common cyanide antidotes include use of nitrites and thiosulfate, and administration of hydroxocobalamin (Shepherd and Velez, 2008). Nitrite or nitrite esters such as amylnitrite, convert hemoglobin (Fe2+) to methemoglobin (Fe3+) which binds to cyanide forming non-toxic cyanomethemoglobin, thus allowing cellular respiration to proceed.

Final detoxification of the cyanide is facilitated by administration of thiosulfate, which, in a reaction catalyzed by rhodanese, reversibly combines with extracellular cyanide to form thiocyanate, which can be excreted in the urine (Shibamoto and Bjeldanes, 1993;

Shepherd and Velez, 2008). This reaction involves sulfane sulfur as the rate-limiting co- factor of rhodanese (Tor-Agbidye et al., 1999). In addition, hydroxocobalamin, an endogenous vitamin B12 precursor, binds cyanide to its cobalt moiety in equimolar amounts forming non-toxic cyanocobalamin, which can be excreted in the urine

(Shepherd and Velez, 2008). It has a greater affinity for cyanide than cytochrome oxidase.

Chronic, low-level dietary cyanide exposure via consumption of improperly processed cassava has been associated with the development of several health disorders. These include tropical ataxic neuropathy (TAN) which is characterized by optic atrophy, ataxia

(inability to coordinate muscle movements), and a paralytic disorder known as Konzo

(Osuntokun 1991; Rosling et al. 1992; Tylleskar et al. 1992; Adamolekun, 2010).

Individuals with these disorders have very low concentrations of sulfur amino acids in the blood and elevated levels of plasma thiocyanate (Shibamoto and Bjeldanes, 1993;

Adamolekun, 2010). Thiocyanate is a known goitrogen, thus goiter is also a common health problem associated with improperly processed cassava (Shibamoto and Bjeldanes,

1993; Vanderpas, 2006).

The specific etiological mechanism of konzo remains unknown. Besides conversion to thiocyanate by rhodanese, cyanide may be converted to cyanate, which in turn, is converted by the cysteine-containing enzyme cyanase to ammonia and bicarbonate

(Schultz, 1949). Prolonged cyanate treatment has been reported to induce peripheral neuropathy in humans and rodents, and spastic paraparesis in macaques (Tor-Agbidye et al., 1999). Tor-Agbidye et al. 1999 found an increase in plasma cyanate in rats deficient in the sulfur amino acids, L-cystine and L-methionine, along with a strong correlation between blood cyanide and plasma cyanate concentration. These data are consistent with the hypothesis that cyanate is an important mediator of chronic cyanide neurotoxicity during protein-calorie deﬁciency (Tor-Agbidye et al. 1998, 1999; Spencer, 1999).

Available sulfur is preferentially used in cyanide detoxification even during protein malnutrition (Tor-Agbidye et al., 1999; Adamolekun, 2010). In addition, the enzyme rhodanese has widespread specificity with respect to both the sulfur donor and acceptor.

Thus, sulfur compounds other than sulfur-containing amino acids may function as sulfur donors (Westley, 1981). One such compound is thiamine. Adamolekun (2010) proposes that konzo is not due to cyanide toxicity specifically, but due to inactivation of thiamine

20 that occurs when the sulfur in thiamine is used in cyanide detoxification due to shortages of sulfur amino acids.

While the etiology is not fully defined, it is clear that Konzo is caused by cyanide.

Symptoms of konzo subside when patients are placed on cyanide-free diets and recur when traditional eating habits are resumed. Cassava must therefore be carefully processed to reduce cyanogens to safe levels (Tables 1.2 and 1.3). Most common processing techniques do efficiently reduce cyanogens to safe levels (Table 1.3). However, there are cases in which processing has not efficiently reduced cyanogens to safe levels. Nyirenda et al. (2010) found the cyanogenic potential in processed cassava in parts of Malawi and

Zambia to be at least five times higher than the recommended levels. Complete detoxification of cassava root depends on effective plant cell disintegration (grating, crushing, etc.), followed by heating or drying (Spencer, 1999).

Table 1.1. Food Sources of Cyanogenic Glycosides and Amount of HCN Produced.

Source: Shibamoto and Bjeldanes, 1993

Plant Amount of HCN Glucoside

(mg/100g)

Bitter almonds 250 Amygdalin

Cassava root 53 Linamarin

Sorghum (whole plant) 250 Dhurrin

Lima bean 10-312 Linamarin

1.4 Postharvest physiological deterioration

Cassava has the shortest shelf-life of any tuber crop (Ghosh et al., 1988). This is due to a rapid postharvest deterioration process which renders the root unpalatable within 72 hours of harvest. This rapid process is associated with mechanical damage which is inevitable during harvesting operations (Booth, 1976). It progresses from the site of damage, eventually causing vascular streaking followed by general discoloration of the vascular parenchyma (Figure 1.8A). Estimated losses due to PPD in cassava range from

5-25% (Wenham, 1995).

1.4.1 Cassava response to wounding in comparison with other crops

PPD in cassava is considered to be due to a failure in the wound healing response of the plant (Wenham, 1995). In most plants, tissue damage results in a cascade of responses leading to the formation of a protective layer around the damaged site. For example, the hypersensitive response (HR) in plants on pathogen attack is characterized by the rapid death of the first infected host cells to restrict of pathogen spread (Bowles, 1990; Chong et al., 1999). Response to stress in plants has also been associated with increased ethylene production (Decoteau, 2005). In cassava, no experimental evidence has been found to confirm the role of ethylene in PPD (Wenham, 1995). Preharvest pruning, which is effective in suppressing PPD, had no significant influence on ethylene production

23 following injury and the application of endogenous ethylene was not found to affect wound responses (Hirose, Data and Quevedo, 1984).

Common wound responses in plants involve several events (Bennett and Wallsgrove,

1994; Bowles, 1990; Wenham, 1995; Han et al., 2001). First, signals that act locally on the wound site (such as jasmonic acid, abscisic acid, salicylic acid and hydrogen peroxide) are produced. This is followed by the production of defensive enzymes and molecules for protection against pathogens or the effects of wounding, including lytic enzymes (glucanase and chitinase) that attack components of microbial cell walls, and phenolics, that can act as antimicrobials (phytoalexins) or anti-oxidants (Bowles, 1990).

Thirdly, wound repair includes the synthesis of suberin and lignin from phenolic components, callose synthesis, the insolublisation of hydroxyproline-rich glycoproteins by H2O2, and the formation of a wound meristem (Beeching et al. 1994; Wenham, 1995).

This repair leads to the sealing of the wound, the inhibition of the production of the signals triggering the wound response, and a return of the plant to normal development

(Uritani, 1999).

In contrast to other tuberous crops such as potato, the formation of lignin in mechanically damaged cassava roots is poor under ambient conditions (Uritani, 1999). Coumarins such as scopoletin, which causes UV fluorescence in cut cassava, are absent in the fresh plant and accumulate only upon damage (Wenham, 1995; Uritani, 1999). For this reason, it has been assumed that scopoletin is involved in PPD in cassava (Buschmann et al. 2000;

Blagbrough et al. 2010).

1.4.2 Biochemical events during cassava PPD

Researchers at the University of Bath have dissected many biochemical events occurring during cassava PPD (Buschmann et al. 2000; Han et al. 2001; Reilly et al., 2004). An increase in the production of hydrogen peroxide has been observed in the first 15 minutes after mechanical injury. Phenylalanine ammonia-lyase (PAL) activity has been found to increase during PPD (Wenham, 1995). PAL genes, the key entry enzyme to general phenylpropanoid metabolism, have been cloned from cassava, and two family members are expressed during PPD (Tanaka et al. 1983; Buschmann et al. 2000).

Several secondary metabolites also accumulate during PPD. These include hydroxycoumarins (e.g. scopoletin), flavan-3-ols and diterpenes (Rickard, 1981; 1985;

Tanaka et al., 1983; Wheatley, 1982; Wheatley and Schwabe, 1985). Scopoletin is mainly responsible for the fluorescence in the storage parenchyma observed after cutting cassava

(Figure 1.8B). Its content peaks within 24 hours of injury and prior to the development of the visual symptoms of physiological deterioration (Wheatley and Schwabe, 1985).

Furthermore, the accumulation of scopoletin is more pronounced in cassava cultivars that are more susceptible to PPD, decreasing after 6 days due to metabolism of scopoletin to an insoluble blue-black product by peroxidation (Buschmann et al. 2000, Reilly et al.

2004, and Blagbrough et al. 2010).

Figure 1.8. Postharvest physiological deterioration in cassava roots. PPD is characterized by vascular discoloration (A) and has been associated with scopoletin, which can be analyzed by fluorescence (B).

1.4.3 Oxidative damage during PPD

Reilly et al. (2004) reported an oxidative burst 15 minutes after mechanical injury in cassava roots. Harvested cassava roots were injured by cutting into slices and superoxide was detected by the formation of a blue precipitate with nitro blue tetrazolium while hydrogen peroxide was detected by formation of a brown precipitate with 3, 3- diaminobenzidine (Reilly et al. 2004). The oxidative burst (measured as superoxide and hydrogen peroxide) was detected within 15 minutes of injury, declining to low levels 8-

10 hours later (Reilly et al. 2004). In addition, the activities of enzymes that modulate reactive oxygen species (such as catalase and peroxidase) have been seen to increase during PPD (Beeching et al., 1998). Varieties less susceptible to PPD have also been shown to have high activities of catalase, suggesting that their long shelf life may be due to efficient scavenging of hydrogen peroxide (Reilly et al., 2004).

1.4.3.1 Reactive oxygen species in plant stress response

Oxygen has been described as a double-edged sword; on one hand it is necessary for the production of energy while on the other it is capable of generating superoxide and other reactive oxygen species (Miwa et. al. 2008). The production of reactive oxygen species

(ROS) is an unavoidable consequence of aerobic respiration. A one electron reduction of oxygen yields superoxide, which can spontaneously dismutate to produce hydrogen peroxide in the presence of water (Miwa et. al. 2008).

Plants can produce ROS when damaged (such as during harvesting cassava) or attacked by pathogens resulting in an oxidative burst (Jones and Smirnoff, 2005; Reilly at al.,

2004). ROS production can have a direct role in defense, for example, by damaging the pathogen and enabling peroxidative cross-linking of cell wall proteins and polysaccharides (Smirnoff and Jones, 1995). ROS also initiates signal transduction cascades that activate defense-related gene expression along with varied input from other signaling molecules, such as salicylic acid (SA), nitric oxide (NO) and jasmonic acid

(JA). However, ROS can also directly provoke cellular damage by rapidly oxidizing cellular components, including lipids (Triantaphylide`s et al., 2008). Thus, plants have several mechanisms for scavenging reactive oxygen species to prevent cellular damage, such as the enzymes dismutase (SOD), catalase, and ascorbate peroxidase. SOD accelerates the decomposition of superoxide to hydrogen peroxide, which is removed by catalase or ascorbate peroxidase (Sairam and Tyagi, 2004; Miwa et. al. 2008). A delicate balance between ROS production and scavenging is necessary to avoid oxidative stress in plants (Figure 1.9).

Figure 1.9. A balance between ROS production and scavenging is necessary to avoid oxidative stress.

1.4.3.2 Alternative oxidase prevents ROS accumulation in plants

The cyanide-resistant alternative oxidase (AOX) is a nuclear-encoded protein found in mitochondria of all higher plants (Albury et al., 2009). AOX is encoded in a multigene family with differential expression of the isozymes under various circumstances (Albury et al., 2009). Considine et al. (2002) identiﬁed two subgroups: AOX1, present in all

Angiosperms, and AOX2 identiﬁed only in dicots. It has been proposed that AOX1 is induced by stress, whereas AOX2, which appears to be the major form in many dicots, is usually constitutive or developmentally regulated (Considine et al. 2002). Arabidopsis for example contains four AOX1 genes (a–d) and one AOX2 gene (Rassmuson et al.,

2009). In Arabidopsis, AOX1a is induced by stress treatments (Clifton et al. 2005).

The AOX pathway branches from the mitochondrial electron transfer pathway at the level of the ubiquinone pool and couples the oxidation of ubiquinol to the four-electron reduction of oxygen to water (Figure 1.10). In plants, the AOX is a homodimeric protein, associated with the matrix side of the inner mitochondrial membrane (Maxwell et al.

1999). However, the occurrence of the AOX is not restricted to plants (Vanlerberghe and

McIntosh, 1997). The bloodstream form of Trypanosoma brucei, the parasitic protist that causes African sleeping sickness, depends entirely on the alternative pathway as it lacks functional cytochromes (Chaudhuri and Hill 1996; Vanlerberghe and McIntosh, 1997).

In the alternative oxidase pathway for oxygen reduction, the transfer of electrons from ubiquinone to oxygen does not contribute a transmembrane potential, thus two energy coupling sites are lost. The AOX pathway therefore has potential energetic consequences for cellular metabolism (Moore and Siedow 1991; Vanlerberghe and

McIntosh, 1997). Growing evidence suggests that AOX may play a signiﬁcant role in cell adaptation under stress (Polidorosa et al., 2009). Using transgenic tobacco suspension cells with constitutively high expression of AOX and others with suppressed levels,

Maxwell and colleagues (1999) demonstrated an inverse relationship between the abundance of AOX within the mitochondria and the level of ROS found in the plant cell.

Various conditions enhance AOX activity, including disruption of the cytochrome pathway (Maxwell et al. 1999), abiotic and biotic stresses such as chilling (Vanlerberghe and McIntosh, 1992), drought (Bartoli et al. 2005), pathogen attack (Lennon et al. 1997) and wounding (Hiser and McIntosh 1990). By using reductant in excess of either the cytochrome pathway capacity or the rate of ATP use, AOX prevents excessive reduction of upstream components of the mitochondrial electron transfer chain, which could result in disruption of glycolysis and the tricarboxylic acid (TCA) cycle under conditions of high substrate availability (Lambers, 1982; Maxwell et al., 1999). Thus, AOX activity enables high turnover rates of carbon and operation of TCA cycle under stress conditions, although with a cost in ATP synthesis (Mackenzie and McIntosh 1999).

The fact that AOX is cyanide-resistant and prevents ROS accumulation makes it a possible candidate in preventing the oxidative burst induced by mechanical damage in cassava.

Figure 1.10. The mitochondrial electron transport chain showing location of alternative oxidase.

The transport chain supporting oxidative phosphorylation branches at ubiquinone. Electrons ﬂow from reduced ubiquinone (QH2) through the cytochrome pathway or to alternative oxidase

(AOX, red). I, NADH-ubiquinone oxidoreductase; II, succinate - ubiquinone oxidoreductase; III, ubiquinol-cytochrome c oxidoreductase; IV, cytochrome c oxidase; V, ATP synthase; AOX, alternative oxidase; c, cytochrome c; QH2, ubiquinol (reduced ubiquinone); ND ex/in, NAD(P)H dehydrogenases, respectively external/internal. Source: Juszczuk and Rychter 2003. Reproduced with permission from Acta Biochimica Polonica.

1.5 Strategies for controlling PPD

1.5.1 Storage conditions

The most effective method for controlling cassava PPD is oxygen exclusion (Reilly et al.,

2004; Lebot, 2009). Consequently, cassava for exports markets such as Europe and the

USA is treated with wax. Roots dipped in ordinary paraffin wax for 45 seconds at 90-

95°C can be stored for a month (Lebot, 2009).

At the farm level, PPD can also be delayed by manipulating storage conditions. The simplest method that many farmers use is in-ground storage, i.e. harvesting the crop when it is needed. This is made possible by the fact that cassava does not have a distinct period of physiological maturity (Wenham, 1995). The harvesting window is thus flexible between 6-24 months. This strategy has several disadvantages. First, prolonged in-ground storage may increase chances of pathogen infection (Westby, 2002). At the same time the roots become woody and there can be loss of flavor (Lancaster and Coursey, 1984).

Lastly, in-ground storage locks up land that can otherwise be productively used.

Storage in clamp silos, where roots are piled up on a layer of straw in conical heaps weighing between 300 and 500 kg and covered with straw and soil with openings left for ventilation, has been found to be effective for four weeks (Rickard and Coursey, 1981;

Westby, 2002). The system is labor-intensive and requires some expertise to set up.

Box storage, with sawdust or coconut husk is also effective for four weeks (Westby,

2002). Moisture content of the sawdust requires careful control; too much moisture

33 promotes fungal growth while too little hastens deterioration. Lining the crates with plastic foil prevents drying out of the sawdust resulting in a storage period of 4–8 weeks

(Rickard and Coursey, 1981).

Cassava roots can also be stored in polyethylene films for up to three weeks with microbial and fungicidal treatments (Lebot, 2009). This method is limited in use by the cost of polyethylene bags and fungicide.

Low temperature storage is another strategy used to delay deterioration in cassava, although it is seldom practical for smallholder farmers. The most favorable temperature for storing fresh cassava is 3°C. At this temperature, the total weight loss after 14 days was 14% and was 23% after 4 weeks (Rickard and Coursey, 1981). Alternatively, roots, or pieces of root, can be stored frozen. Freezing changes the texture making it somewhat spongier, but the flavor is preserved (Rickard and Coursey, 1981). Roots stored at low temperature deteriorate faster when taken back to room temperature. Storage methods that cut off oxygen, such as storing in a water bath, are also practiced (Plumbey and

Rickard, 1991).

1.5.2 Cassava processing

The processing of cassava helps to overcome problems with postharvest physiological deterioration, while also reducing the amount of cyanogens in the roots. Common processing techniques include peeling, grating, drying and fermentation (Westby, 2002).

Foods that are made from some of these processing methods are show in Table 1.2 and

Table 1.3.

Table 1.2. Foods made from cassava. (Sources: Lancaster et al., 1982; Omole, 1977;

Balagopalan, 2002; Dufour, 2007; Ihemere et al., 2008; Lebot, 2009).

Name Preparation Region Fufu Fermentation of roots in a pot for 4–7 days, boiling, and West Africa pounding into a sticky dough which is eaten with vegetable soup supplemented with fish and meat Dumby Pounding boiled roots in mortar; supplemented with vegetables Liberia or meat Farina Prepared by removing the skin of the cassava root and grating. South America and The mash is then depulped, sieved, and roasted in a slow fire. It West Indies can then be stored for several months and can be eaten on its own or in combination with other foods Cassava Prepared by blending cassava flour, groundnut flour, and macaroni wheat semolina in the ratio 60:12:15. It is enriched with 12% protein. The food is used to feed children because of its high protein content Gari Cassava roots are skinned, grated, and dewatered in sacks made Ghana, Nigeria, from jute fibers and allowed to ferment for 2–4 days. This is Guinea, Togo, and followed by sieving to remove fibers from the roots. It is then Benin fried in shallow iron pans and stirred continuously until it becomes dry and crisp. Palm oil is added sometimes during frying to prevent burning and also as a source of carotene in the food. It is prepared in boiling water to make a thick paste Casabe The roots are scraped to remove the bark (but not the inner Caribbean Islands (cassava peel) washed and grated into a fine mash. The grated mash is bread) separated into three fractions: liquids, starch and fiber by washing (with water and extracted juices) and squeezing the mash in a basketry strainer. The starch is allowed to settle out of the wash water, and both starch and fiber are allowed to ferment, at least overnight, but preferably for 48hr. The fiber is then dewatered and lightly toasted, mixed with the starch, and the mixture cooked on a griddle as flat bread. Landang fresh cassava roots are squeezed to remove the juice; the pulp is Philippines (cassava made into pellets called landang rice)

Fermentation of cassava roots under water is conducted in many cassava producing countries in Africa. A variety of products are produced (Table 1.1 and Table 1.2). Roots are soaked in water for 3–5 days during which time they soften. This causes lactic acid fermentation and reduces pH to 4 (Lebot, 2009; Oyewole and Odunfa, 1988). Pre-mold fermented roots are inserted during grating to act as starter colonies. Heap fermentation, achieved by heaping peeled roots and leaving them to ferment naturally, is practiced in Tanzania (Ndunguru et al., 1999), Uganda and Mozambique (Essers, 1995).

Table 1.3. Efficiency of common cassava processing techniques in reducing cyanogens.

Tabulated from information presented by Dufour, 2007

Type of Percent reduction in Number of days Reference

processing cyanogens after taken to reduce

processing cyanogens

Farina 99% 8 days Dufour, 1989

Gari 93% 4 days Vasconcelos et al.

1990

Casabe 97% 2 days Dufour, 1989

Sun dried flower 66% 17 days Mlingi and

Bainbridge,1994

Heaped 95% 5 days Essers et al. 1995

fermented flour

1.5.3 Plant Breeding

Breeding for longer shelf life has been hampered by a number of factors, such as the influence of environmental growth conditions and associated pre-harvest stress, limited genetic variability and a persistent but moderate negative correlation between low deterioration rate and high dry matter content (Wenham, 1995; Sanchez et al., 2005). In addition, cassava roots are not organs of propagation, thus selection for postharvest conservation may not provide a selective advantage (Wenham, 1995).

However, cassava‟s highly differentiated gene pools and the large percentage of dominant/recessive gene action loci make it highly heterotic (Fregene and Puonti-

Kaerlas, 2002). The use of molecular markers for germplasm assessment is being employed to take advantage of this.

Using 101 clones, Sanchez et al. (2005) showed a negative correlation between PPD and carotenoid content, suggesting that carotenoids may help delay PPD. Morante et al.

(2010) also identified genotypes with delayed PPD. The availability of these new sources of tolerance should make possible the identification of molecular markers linked to PPD

(Morante et al., 2010).

These studies have provided information on the nature of PPD, allowing for the possibility of using biotechnology tools for solving the problem of PPD in cassava.

1.6 Objectives

Cassava produces potentially toxic levels of cyanogenic glycosides and has the shortest shelf life of any tuber crop due to rapid postharvest physiological deterioration.

The first goal of the project was to redirect (by genetic manipulation) cyanide metabolism in cassava towards increased root protein accumulation. This is based on the hypothesis that cyanogenic glycosides are a transportable form of reduced nitrogen in plants.

Cyanide can be detoxified via the enzyme rhodanese or assimilated via β-cyanoalanine synthase into amino acids and ammonia (Figure 1.5). The following hypotheses were proposed

1. β-cyanoalanine synthase assimilatory pathway is the preferred pathway for

cyanide metabolizing enzyme in cassava roots

2. Expression of β-cyanoalanine synthase in cassava roots increases free amino acid

pool sizes and total protein in cassava roots

3. Expression of Nitrilase 4 gene, which metabolizes β-cyanoalanine to asparagine,

increases free amino acid pool sizes and total protein in cassava roots.

Biochemical assays indicated that β-cyanoalanine synthase is the key enzyme

metabolizing cyanide in cassava roots. Expression of β-cyanoalanine synthase in

cassava roots, while increasing free amino acid and total protein in cassava, seemed

to be detrimental, however, to overall plant growth and survival. Expression of

Nitrilase 4 resulted in disruption of auxin metabolism. These results suggest further

studies are required in defining the cyanide assimilation pathway in cassava roots, as

well as its relation to auxin metabolism.

The second goal of this project was to define the mechanism of postharvest physiological deterioration and to find a transgenic strategy to delay PPD. The following hypotheses were proposed:

1. Cyanogenesis, initiated by tissue disruption in cassava roots, causes the oxidative

burst which leads to vascular discoloration.

2. Expression of Arabidopsis alternative oxidase gene (AOX1A) reduces

accumulation of reactive oxygen species (ROS) in cassava

3. Reduction in ROS accumulation leads to delayed PPD.

Using transgenic low cyanogen plants (Siritunga and Sayre, 2003) and wild type plants, as well as biochemical complementation with potassium cyanide, we show that the oxidative burst in ruptured cassava is induced by cyanogens. By expressing AOX in cassava roots, we were able to block accumulation of ROS in the roots, which resulted in a two to three week delay of PPD.

Having generated transgenic plants expressing alternative oxidase, we also considered the possibility that they could offer tolerance to abiotic stress, such as waterlogging and salinity. Thus, we exposed the plants to waterlogging and salt stress and measured the response. Our results indicate the potential of AOX as a strategy to control abiotic stress in cassava plants.

Chapter 2: The role of cyanogenesis in cassava nitrogen metabolism

2.1 Introduction

Hydrogen cyanide (historical common name Prussic acid) is a colorless extremely poisonous chemical with a bitter almond-like odor that some people cannot smell due to a genetic trait. It is one of the most abundant molecules in space and is a central molecule in prebiotic chemistry as it is thought to have been involved in the synthesis of nucleic acid bases and amino acids (Miyakawa et al. 2002). Cyanide generally occurs in nature in inorganic (hydrogen cyanide) or organic form (nitriles) (O‟Reilly and Turner, 2003).

All plants release hydrogen cyanide as a byproduct of the ethylene biosynthesis pathway

(Peiser et al., 1984; Dong et al. 1992). In addition, cyanogenic plants like cassava, bitter almonds and rubber have the potential to produce large amounts of cyanide from the breakdown of cyanogenic glycosides (Poulton, 1990). In cassava, the major cyanogenic glycoside is linamarin, which is stored in vacuoles of all plant tissues (McMahon et al.

1995). Cyanide is released from linamarin upon tissue disruption, e.g. during harvest- induced mechanical damage or chewing by insects, which facilitates hydrolysis of linamarin to acetone cyanohydrin by the cell-wall localized β-glucosidase, linamarase

(Mkpong et al., 1990; McMahon et al. 1995). Acetone cyanohydrin will spontaneously decompose to cyanide and acetone at pH > 5.0 or temperatures > 35°C, or is broken down by the enzyme hydroxynitrile lyase (HNL), which is expressed only in leaves and stems and not in roots (White et al., 1998). Cyanide is an inhibitor of cytochrome c 41 oxidase in the mitochondrial electron transport chain. Thus, cyanide accumulation is potentially toxic. However, cyanide can also be a substrate for amino acid synthesis in plants. The enzyme β-cyanoalanine synthase (CAS) catalyzes the reaction between cyanide and cysteine to form β-cyanoalanine and hydrogen sulfide. β-cyanoalanine is further converted by a nitrilase to asparagine which can be deaminated to form aspartate and ammonia (Blumenthal et al., 1968; Elias et al, 1997; Hatzfeld et al. 2000; Lai et al.

2009) or conjugated to the peptide gamma-glutamyl-β-cyanoalanine by the action of gamma glutamyl transferase which is also converted to asparagine by asparaginase

(Blumenthal et al. 1968; Yang, 1989). This detoxification pathway results in the assimilation of cyanide into primary metabolism and is present in all higher plants thus far examined (Blumenthal et al., 1968; Miller and Conn, 1980). CAS is localized in mitochondria, effective for removal of cyanide to protect the oxidative phosphorylation process (Maruyama et al. 1998; Lai et al. 2009). The optimal pH for CAS activity is around 8.5, which is also the pH in the matrix of mitochondria (Maruyama et al. 1998;

Lai et al. 2009). Cytosolic CAS activity has also been reported and attributed to cysteine synthase which catalyzes the formation of cysteine from O-acetyl-L-serine and H2S

(Hasegawa et al. 1995; Maruyama et al. 1998). In addition to its contribution to amino acid synthesis by metabolizing cyanide, cytosolic CAS may be involved in protecting cytosolic enzymes such as nitrate reductase, catalase and ribulose bisphosphate carboxylase from cyanide toxicity (Solomonson, 1981).

Β-cyanoalanine produced by the activity of CAS is metabolized by a nitrilase. Nitrilases catalyze the hydrolysis of nitrile compounds (R-CN) to their corresponding carboxylic

42 acid and ammonia (O‟Reilly and Turner, 2003). Nitrile compounds are common in the environment as intermediates in chemical biosynthesis and degradation and as contaminants from industrial and farming processes (Legras et al. 1990; Howden at al.

2009). In plants, nitriles are found in a number of pathways including the biosynthesis of cyanolipids and cyanogenic glycosides, the breakdown of glucosinolates, and the detoxiﬁcation of cyanide (Halkier and Gershenzon, 2006; Legras et al., 1990). While showing significant similarities at the amino acid and protein structure level, nitrilases exhibit specific substrate specificities (Oreilly and Turner, 2003).

For example, the four nitrilase genes identified in Arabidopsis comprise two subgroups and encode proteins with different catalytic properties (Bartel and Fink,

1994; Piotrowski et al., 2001; Vorwerk et al., 2001). One group comprises only a single gene, NIT4, with orthologues in rice, Lotus japonicus and tobacco (Dohmoto et al., 1999; Dohmoto et al., 2000; Tsunoda and Yamaguchi, 1995). The NIT4 orthologues of tobacco and Arabidopsis thaliana encode β cyano-L-alanine- hydratases/nitrilases involved in cyanide detoxification (Piotrowski et al., 2001). The second subgroup of Arabidopsis nitrilases encompasses the highly similar isozymes

NIT1, NIT2 and NIT3 (Kutz et al., 2002). Members of this subfamily are responsible for the metabolism of nitriles originating from glucosinolate breakdown (Vorwerk et al.,

2001). They convert indole-3-acetonitrile to the plant hormone indole-3-acetic acid

(IAA) (Bartling et al., 1992; Bartling et al., 1994; Schmidt et al., 1996). This reaction has been proposed to play a role in particular physiological situations, such as embryo

43 development and maturation, rather than for the general IAA supply of the plant (Muller and Weiler, 2000; Vorwerk et al., 2001).

Cyanogenic glycosides have been proposed to act as transportable forms of reduced nitrogen in some plants (Selmar et al. 1988; Selmar, 1993; McMahon et al., 1995;

Siritunga and Sayre, 2003). Nartey (1968) demonstrated that cassava seedlings could assimilate cyanide into amino acids. Analysis of cassava seeds revealed no linamarin and little free valine available for linamarin synthesis. Also, growing cassava seedlings accumulated asparagine and glutamine rather than linamarin, suggesting that asparagine, not linamarin, is the major transportable form of reduced nitrogen in cassava seedlings

(Nartey, 1968). In the rubber tree (Hevea brasiliensis), linamarin is converted into its diglucoside linustatin during germination, which is transported to the seedling where it is assimilated into asparagine and aspartic acid by β-cyanoalanine synthase and nitrilase

(Selmar et al. 1988).

A second cyanide detoxifying enzyme is rhodanese, a thiosulfate sulfurtransferase which catalyzes the reaction between thiosulfate and cyanide to give thiocyanate and sulfite

(Hatzfeld and Saito, 2000). Cyanide detoxification by rhodanese is more prominent in mammals where the thiocyanate is excreted in urine (Shibamoto and Bjeldanes, 1993).

The use of thiosulfate as a cyanide antidote is based on the activation of rhodanese

(Shibamoto and Bjeldanes, 1993; Shepherd and Velez, 2008). In plants, however, a relationship between rhodanese and cyanogenesis has not been firmly established (Chew,

1973; Miller and Conn, 1980; Hatzfeld and Saito, 2000). No correlation between cyanogenic potential and rhodanese activity was found in studies with cyanogenic and

44 acyanogenic genotypes of Lotus corniculatus and Trifolium repens (Chew, 1973; Kakes and Hakvoort, 1992). Miller and Conn examined the activity of rhodanese in 16 species

(cyanogenic and non-cyanogenic) and detected activity in only 4 species, leading to the conclusion that rhodanese occurs less commonly in plants. On the other hand, Shirai and

Kurihara (1991) detected activity in all the plants they tested, with higher values in cyanogenic (mean activity 55.9 nmol/mg protein/min) compared to non-cyanogenic plants (mean activity 11.6 nmol/mg protein/min). They concluded that rhodanese may indeed function in cyanide detoxification. These results have not been supported by other work in plants (Selmar, 2010). While some early work reported rhodanese activity to be higher in cyanogenic plants, careful analysis indicated artifacts in the assay procedures

(Selmar, 2010). Since the reaction occurs spontaneously, correct analysis of rhodanese cyanide detoxification activity requires careful control. Recent studies have shown that rhodanese activity in plants is relatively low in comparison to β-cyanoalanine synthase

(Shirai and Kurihara, 1991; Kakes and Hakvoort, 1992; Nambisan and Sundaresan, 1994;

Elias et al., 1997). For example, Shirai and Kurihara (1991) compared the mean activities for rhodanese in their assays for cyanogenic (55.9 nmol/mg protein/min) and non- cyanogenic plants (mean 11.6 nmol/mg protein/min) to the mean activities of β- cyanoalanine synthase (720 nmol/mg protein/min and 281.5 nmol/mg protein/min in cyanogenic and non-cyanogenic plants respectively). These observations suggest that cyanogenic plants more actively assimilate cyanide derived from cyanogenesis than detoxify it by rhodanese.

The goal of this work was to determine the contribution of cyanogens to reduced nitrogen assimilation and explore the feasibility of engineering cassava plants for enhanced incorporation of cyanide into free amino acids and possibly protein. We investigated the enzymes involved in detoxifying or assimilating cyanide, namely rhodanese and β- cyanoalanine synthase. We carried out enzyme assays to determine the relative activity of these enzymes in cassava roots (in vivo). Two enzymes are involved in cyanide assimilation via the β-cyanoalanine synthase pathway: β-cyanoalanine synthase and nitrilase 4 (NIT4). We investigated whether the β-cyanoalanine synthase pathway contributes significantly towards amino acid pool sizes from the metabolism of linamarin.

To achieve this, we expressed, separately, Arabidopsis β-cyanoalanine synthase and NIT4 under the control of the tuber-specific patatin promoter and measured changes in free and total amino acids. Our results provide insights into the role for cyanogenic glycosides in nitrogen metabolism in the roots.

2.2 Experimental procedures

2.2.1 Cassava cultivars

Cassava cultivar Manihot Columbia 2215 (MCol 2215) was used for assays of β- cyanoalanine synthase and rhodanese activity. Transformation work was done using cultivar TMS 60444, selected due to ease of transformation using both friable embryogenic callus and embryo cotyledons. Comparative assays with transgenic lines

46 were performed using TMS 60444 as the wild-type control. Transgenic lines previously generated for low cyanogenesis, Cab1-1, Cab1-2 and Cab1-3 with MCol as the background (Siritunga and Sayre, 2003) were also used in this study.

2.2.2 Tissue culture propagation of plant material

Cassava plants were propagated in vitro on the Murashige and Skoog (MS) basal medium

(Murashige and Skoog, 1962) supplemented with Gamborg vitamins (Gamborg et al.,

1968) and 2% sucrose. In vitro plants were propagated in growth chambers at 28°C with

16 hours of light and 8 hours of darkness. Micropropagation of plant materials was done once every 5-8 weeks depending on the requirements of specific experiments.

2.2.3 Total protein extraction and analysis

Total protein was measured using the Bradford assay according to the supplier‟s

(Invitrogen, www.invitrogen.com) instructions. Protein was extracted from root and leaf tissue of cassava using the following extraction buffer (unless otherwise described): 50 mM Tris-HCl (pH 8.5), 5 mM dithiothreitol (DTT) and 1mM EDTA. Extraction buffer was used at a ratio of 5 mL of buffer per gram fresh tissue.

For leaf tissue, leaves were ground in liquid nitrogen to a fine powder before adding the buffer. Tuberous roots were blended together in the buffer in the Magic Bullet MB1001

47 blender (Homeland Houseware LLC) for 30 seconds at 4°C. The ground extract was passed through four layers of cheesecloth and centrifuged at 21000 g for 10 minutes in the Hermle Z233 MK-2 Refrigerated Microcentrifuge (Labnet International Inc., NJ).

The supernatant was used as the crude extract and measured for protein using the

Bradford reagent with bovine serum albumin (BSA) as the standard. To measure protein concentrations of crude extracts, 100 µL of extract were added to 3 mL of Bradford reagent and absorbance at 595 nm was read after 10 minutes. The concentration was calculated using a standard curve, based on known concentrations of bovine serum albumin (BSA) up to 1 mg/mL.

2.2.4 Activity of β-cyanoalanine synthase in cassava tissue

The activity of β-cyanoalanine synthase was determined using the method described by

Goudey et al. (1989) with some modifications. The method measures the formation of hydrogen sulfide by formation of methylene blue (described below). Cassava tissue was ground in liquid nitrogen using a motor and pestle and extracted in a buffer containing 50 mM Tris-HCl, 5.0 mM dithiothreitol (DTT), 5.0 mM phenylmethylsulfonyl fluoride

(PMSF) and 1.0 mM EDTA at pH 8.0. The extract was filtered through four layers of cheesecloth to remove debris. To 500 µL of substrate solution (10 mM L-cysteine and 10 mM sodium cyanide in 50 mM Tris buffer pH 8.0), 100-200 µg of crude protein extract was added to make a total reaction volume of 1.0 mL. The optimum concentration of enzyme used was determined after testing different amounts of wild-type plant enzyme 48 extract. The reaction was carried out for 10 min (determined after testing 5 to 30 minute time points) and stopped by adding 0.1 mL of 30 mM ferric chloride in 1.2N HCl followed by 0.1 mL of 20 mM N,N dimethyl-p-phenylenediamine sulfate in 7.2 N HCl.

The mixture was closed and shaken, and left at room temperature. A blue color develops, showing the presence of sulfide. Absorbance was measured at 640 nm after 10 min.

Concentration of sodium sulfide was estimated from the absorbance by using a standard curve prepared by adding 0.1 mL 30 mM ferric chloride in 1.2 N hydrochloric acid (HCl) per mL of total reaction mixture followed by 0.1 mL of 20 mM N, N dimethyl-p- phenylenediamine sulfate in 7.2 N HCl to 1mL of known concentrations of sodium sulfide (Figure 2.1).

120

100 R² = 0.9898

Hydrogen sulfide concentration in 60 µM

0 0 0.5 1 1.5 2 OD at 640 nm

Figure 2.1. Standard curve used for estimating the concentration of hydrogen sulfide from

β-cyanoalanine synthase assays. The standard curve was prepared by adding 0.1 mL 30 mM ferric chloride in 1.2 N hydrochloric acid (HCl) per mL of total reaction mixture followed by 0.1 mL of 20 mM N,N dimethyl-p-phenylenediamine sulfate in 7.2 N HCl to

1mL of known concentrations of sodium sulfide and measuring absorbance at 640 nm after 10 minutes.

2.2.5 Activity of Rhodanese

The activity of rhodanese was assayed colorimetrically by quantifying the amount of thiocyanate produced in the reaction between sodium cyanide and sodium thiosulfate as described by Wang and Volini (1968) with modifications. Cassava plant tissue was ground in liquid nitrogen using a motor and pestle and extracted in a buffer made up of

200 mM sodium phosphate buffer pH 7.8, 5.0 mM DTT, 5.0 mM phenylmethylsulfonyl fluoride (PMSF), 1.0 mM EDTA, 1.0 mM Na thiosulfate (to keep the enzyme in a stable rhodanese-sulfur intermediate), 5 mM potassium chloride and 2% w/v polyvynilpolypyrrolidone (PVPP for adsorbing polyphenols). The homogenate was passed through four layers of cheesecloth to remove debris and centrifuged at 21000 g for

5 minutes at 4°C. The supernatant was used in subsequent assays. Protein concentration of the supernatant was quantified by the Bradford assay, using bovine serum albumin as the standard. To start the reaction, about 100-200 µg of protein was added to 0.5 mL of

50 mM sodium cyanide and 50 mM of sodium thiosulfate in 200 mM sodium phosphate buffer (pH 7.8) to a total volume of 1.0 mL. The reaction was incubated at 30°C for 10 minutes and stopped by adding 0.5 mL 15% (v/v) formaldehyde. Absorbance at 460 nm was measured after adding 2.5 ml of ferric nitrate reagent. The reagent was prepared by adding 20 mL nitric acid (65%) to 60 mL of water, dissolving 10 g Ferric nitrate.9 H2O and making up to a final volume of 100 mL

The reaction was blanked using boiled (inactive) enzyme extract. The standard curve used to estimate the concentration of thiocyanate (Figure 2.2) was prepared using a range of known concentrations of thiocyanate in the same volume as the reaction.

2 R² = 0.9997 1.8

1.6

1.4

1.2 Thiocyanate concentration in 1 mM 0.8

0.6

0.4

0.2

0 0 0.2 0.4 0.6 0.8 1 1.2 Absorbance (460 nm)

Figure 2.2. Standard curve used for estimating the concentration of thiocyanate in the rhodanese assay. Sodium thiocyanate concentrations ranging from 0 to 2.0 mM were used in a total volume of 1.0 mL. Absorbance was measured at 460 nm after adding 2.5 mL of ferric nitrate reagent. The linear curve was used to estimate the concentration of thiocyanate from absorbance in the assay.

2.2.6 Determination of nitrilase activity

The activity of nitrilase was determined by measuring the amount of ammonia produced using the Berthelot reaction as described by Piotrowski et al (2001) with some modifications. Cassava tissue was homogenized in an extraction buffer containing 50 mM Tris-HCl (pH 8.5), 2.0 mM EDTA, 8.0 mM cysteine, 2% (w/v) polyvinylpyrrolidone

(PVP) and 0.1% (w/v) bovine serum albumin. Tuberous greenhouse roots were homogenized in the Magic Bullet MB1001 blender (Homeland Houseware LLC) for 5×2 seconds, while in vitro plant material was homogenized by grinding with liquid nitrogen in a motor and pestle. In all cases, the homogenate was filtered through four layers of cheesecloth and centrifuged for 5 min at 22000 g. Protein concentration was determined in the supernatant by the Bradford assay. Approximately 400 µg of protein were used in the subsequent enzyme assay. Enzyme extracts were pre-warmed at 37°C for two minutes before being incubated with substrate (10 mM cyanoalanine in a buffer containing 50 mM Tris-HCl, pH 8.5 and 1.0 mM DTT) for 10 minutes at 37°C. The total reaction volume was 1.0 mL. The reaction was stopped by adding 100 µL of tricarboxylic acid

(TCA) and centrifuged at 22000 g for two minutes to clarify the solution. To 500 µL of the supernatant, 1.0 mL of Nessler‟s reagent (Sigma-Aldrich, www.sigmaaldrich.com/) was added. The samples were incubated at room temperature for 10 minutes to allow color development. For blank samples, TCA (to stop the reaction) was added at time zero. Absorbance was read at 480 nm and the amount of ammonia produced was estimated using a standard curve (figure 2.3).

100 y = 466.39x + 20.541 90 R² = 0.9921 80

60 Ammonia 50 concentration (µM) 40

0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 Absorbance at 480 nm

Figure 2.3. Standard curve used to determine the amount of ammonia from ODs at 480 nm. To a range of known concentrations of ammonia (from 30 to 100 nmols) made in

500 µL of deionized reagent grade water, 1.0 mL of Nessler‟s reagent was added. The solutions were incubated at room temperature for 10 minutes to allow color development.

Absorbance was measured at 480 nm.

2.2.7 Determination of nitrate reductase activity

Nitrate reductase was assayed using the method described by the Nitrate Elimination Co.,

Inc. (NECi, www.nitrate.com) with some modifications. Cassava tissue was ground in liquid nitrogen using a motor and pestle. Crude protein was extracted from the ground tissue using an extraction buffer containing 100 mM 3-(N-morpholino)propanesulfonic acid (MOPS) pH 7.5, 1.0 mM EDTA and 10 mM L-cysteine. PVPP (1%, (w/v)) was added to the grinding mixture during extraction. Four milliliters of extraction buffer were used per gram fresh weight of plant tissue. The homogenate was passed through four layers of cheesecloth and centrifuged at 21000 g for 5 minutes at 4°C. The supernatant was used in subsequent assays. Protein concentration of the supernatant was quantified by the Bradford assay. Approximately 100-200 µg of the extracted protein was added to

800 µL of substrate solution (30 mM potassium nitrate in 100mM MOPS, pH 7.5). The reaction was started by adding 100 µL of 25 mM NADH and stopped after 10 minutes by adding 100 µL of 100 mM zinc acetate. After centrifuging at 22000 g for two minutes,

500 µL of the supernatant was add to a fresh 1.5 mL tube. To this, 500 µL (an equal volume to the volume of supernatant) was added of each of the color development reagents (1% (w/v) sulfanilamide in 1.5N HCl and 0.02% N-(napththyl)- ethylenediaminehydrochloride) were added. Samples were left at room temperature for

10-20 minutes to allow full color development. A pink color shows the presence of nitrite, the product of the nitrate reduction reaction. Absorbance was read at 540 nm.

Nitrite concentration was estimated using a standard curve (Figure 2.4) prepared by

56 diluting known concentrations of nitrite in 500 µL and adding the color development reagents.

140

120 R² = 0.9927

100

Nitrite 80 Concentration in µM 60

0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 OD at 540 nm

Figure 2.4. Standard curve used for estimating the concentration of nitrite. A range of known concentrations of potassium nitrite were made to 500 µL. Absorbance for each concentration was determined by adding 500 µL of each of the color development reagents (1% sulfanilamide in 1.5N HCl and 0.02% N-(napththyl)- ethylenediaminehydrochloride in 7.2 N HCL).

2.2.8 Transformation of Cassava with the β-cyanoalanine synthase and Nitrilase

2.2.8.1 Construct design

A modified pBI121 plasmid (named 3D) with a 1.2kb patatin promoter instead of CaMV

35S was used for both constructs (Ihemere, 2002). The β-cyanoalanine synthase (CAS) and nitrilase 4 (NIT4) (TAIR: At5g22300) genes of Arabidopsis were received from the

Arabidopsis Biological Research Center (ABRC, http://abrc.osu.edu/) in the pUNI51 vector, and were cloned into the restriction sites SmaI and SstI of 3D (Figure 2.5). To amplify CAS from pUNI51 for cloning into these sites, a 5′ primer with sequence TAC

CCG GGA AAT GGC CTC TGT TTC AAG GTT AC (named casf) and a 3′ primer with sequence GCG AGC TCC CCG AGA TTT TTG GTA AAG TGT C (named casr) were used. To amplify NIT4 from pUNI51, a 5′ primer with sequence TAC CCG GGA

TGT CCA TGC AAC AAG AAA CGT C (named nitf) and a 3′ primer with sequence

GCG AGC TCT TAG ACG GAT TCA TCT TC (named nitr). The PCR reactions were performed in a Biorad MyCycler thermal cycler (Bio-rad Laboratories, http://www.biorad.com) in a total reaction volume of 50 µL containing 2-10 µg of template DNA

(pUNI51 with CAS or NIT4), 1.0 mM MgSO4, 0.2 mM dNTP, 0.2 µM each of the forward and reverse primers, 1X Platinum pfx amplification buffer and 1 unit of platinum pfx DNA polymerase (Invitrogen, www.invitrogen.com). The reactions were started by

58 first incubating at 95°C for 5 minutes, followed by cycling at a denaturation temperature of 95°C, an annealing temperature of 52°C and an extension temperature of 68°C for thirty-two reaction cycles, an additional extension at 68°C for 7 minutes and finally holding at 4°C.

The construct was used to transform E. coli strain DH5α using the protocol from

Invitrogen. 1-10 ng of DNA was mixed with 50 µL of E. coli cells and incubated on ice for 30 minutes, followed by heat shock at 42°C for 20 seconds. After two minutes on ice,

950 µL of pre-warmed Luria Bertani (LB) medium and incubated at 37°C for 1 hour at

225 rpm. 100 µL of the transformation was spread on LB plates supplemented with 50 mg/L kanamycin. Colonies carrying construct grew on the medium. The construct was verified by digestion with SmaI and Sst1, and DNA sequencing.

Figure 2.5. Constructs design for β-cyanoalanine synthase and Nitrilase expression in cassava. The construct was made in a modified pBI121 vector with the root-specific patatin promoter instead of 35 S. The inserts were cloned into the SmaI and SstI sites.

GOI= gene of interest (β-cyanoalanine synthase or Nitrilase 4). Nos pro= Nos Promoter,

Nos ter=Nos terminator, RB= Right border, LB=Left border.

After verification, the plasmid carrying the insert was introduced into Agrobacterium tumefaciens strain LBA4404 by electroporation based on the protocol from Invitrogen

(www.invitrogen.com). To 2 µL of plasmid DNA, 20 µL of thawed competent LB4404

Agrobacterium cells were added. The cell/DNA mixture was pipetted into a 10 mm cuvette and electroporated using the Biorad MicroPulserTM with voltage set at 2.2kV.

After the electroporation, 1.0 mL of room temperature YM medium (Invitrogen, www.invitrogen.com) was added to the electroporated DNA/cell mixture and transferred to a 15 mL culture tube for incubation at 30°C and 225 rpm for 2 hours. One hundred microliters of the transformation was plated on YM plates with 50 mg/L kanamycin and

30 mg/L streptomycin at 28°C for 48 hours. Plasmids were isolated from the transgenic

Agrobacteria, and the insert was confirmed by PCR using the same primer sets used in the original cloning. Glycerol stocks of Agrobacterium cultures were prepared by adding

500 µL of 80% (v/v) glycerol to 500 µL of Agrobacterium cells. The stocks were frozen in liquid nitrogen and stored at -80°C.

2.2.8.2 Cassava transformation using embryo cotyledons

2.2.8.2.1 Induction of embryogenesis

Somatic embryogenesis was induced using the method described by Ihemere (2003). Leaf lobes from 3-4 week old in vitro cassava plants were cultured on MS medium containing

8 mg/L 2.4-dichlorophenoxyacetic acid (2.4-D) for induction of embryogenesis.

Embryogenic callus developed within four weeks of incubation at 28°C and reduced light

(by covering under two layers of cheesecloth). Embryos were picked and plated on MS with 0.5 mg/L benzyl amino purine for regenerating plants. Embryo cotyledons developed within four weeks on this medium.

2.2.8.2.2 Co-cultivation with Agrobacterium

The embryo cotyledons were cut into discs and inoculated with Agrobacterium tumefaciens strain LBA4404 carrying the construct by placing a drop of the culture on each cotyledon. Agrobacterium and the plant tissue were co-cultivated for 2 days in MS medium supplemented with 8 mg/L 2.4-D, after which the cotyledons were moved to MS containing 500 mg/L carbenicillin to remove Agrobacterium. They were kept on this

61 medium for two days and then transferred to MS with 8mg/L 2.4D, 500 mg/L carbenicillin and 30 mg/L paromomycin (as selection for successful transformation).

2.2.8.2.3 Regeneration of plant tissue

Putatively transgenic embryos were cultured on MS medium containing 1 mg/L benzyl amino purine (BAP). Plantlets were recovered within four weeks on this medium.

Regenerated plantlets were subcultured into regular MS medium to allow development into plants.

2.2.8.3 Cassava transformation using friable embryogenic callus

Cassava transformation using friable embryogenic callus was done following the method described by Taylor et al., (1996). Young leaf lobes were cultured on MS medium containing 12 mg/L picloram to generate friable embryogenic callus (FECs). The FECs were co-cultivated with Agrobacterium carrying the construct for 2 days at 28°C.

Subsequently, they were washed with MS liquid medium containing 500 mg/L carbenicillin to remove Agrobacterium. The washed FECs were transferred to solid medium (MS with 500 mg/L carbenicillin) and kept on this medium for one week before transfer to selection medium (MS with 500 mg/L carbenicillin and 30 mg/L paromomycin). During a 3-6 week period, embryos were picked from this medium into

62 regenerating medium (MS with 1.0 mg/L BAP). Plantlets developing from this medium were cultured on regular MS medium for further propagation and analysis.

2.2.8.4 RT-PCR analysis of transgenic plants

RNA was isolated from 100 mg of cassava roots using the Qiagen RNeasy Plant Mini kit

(Qiagen Inc., Valencia, CA). To quantify RNA, ultraviolet absorbance was measured at

260 nm and 280 nm (Sambrook et al. 1989). Concentrations of RNA were calculated based on absorbance at 260 nm using the Beer-Lambert law, which predicts a linear change in absorbance with concentration. An absorbance 260 nm reading of 1.0 is equivalent to about 40 µg/mL of RNA. RNA purity was judged based on the 260/280 ratio where pure RNA has a value of 2. Prior to cDNA synthesis, the RNA was treated to remove DNA contamination using the Promega DNAse treatment (Promega Corporation,

Madison, WI). About 2-10 µg of RNA was used for cDNA synthesis using the Qscript cDNA kit (Quanta Biosciences; MD).

The cDNA was used to check for the expression of the transgene by RT-PCR. For both

CAS and NIT4 constructs, the primers used amplified a 500 bp region from the end of the

CAS gene to the beginning of the NOS terminator (Figure 2.6). The primers for CAS were CS1F (CATGCTATCACAGGCAATGG) and Nos0329R (GCCAAATGTTTG

AACGATCGG), and NIT4 primers were NT1F (GCACTTGAGGGTGGATGTTT) and

Nos0329R (GCCAAATGTTTG AACGATCGG). For tubulin control, the primers TubF 63

(TATATGGCC AAGTGCGATCCTCGACA) and TubR

(TTACTCTTCATAATCCTTCTCAAGGG) were used as positive controls for the PCR reaction. The PCR reaction was based on ChoiceTM Taq DNA polymerase from Denville

Scientific Inc. (www.densci.com). The reaction mixture contained ChoiceTM buffer (5 µL of 10X), 1.0 µL of 10 mM DNTP, 1.0 µL of ChoiceTM Taq, 1.0 µL of each of the forward and reverse primers, 2-10 µg of template DNA and deionized water to a total volume of

50 µL. The reactions for both the transgene primers (CS1FF and Nos0329R) and the tubulin primers (TubF and TubR) were started by first incubating at 95°C for 5 minutes, followed by cycling at a denaturation temperature of 95°C, an annealing temperature of

52°C and an extension temperature of 72°C for thirty-two reaction cycles, followed by an addition extension at 72°C for 7 minutes before holding at 4°C. The reactions were carried out in a Biorad MyCycler thermal cycler (Bio-rad Laboratories, http://www.biorad.com).

Figure 2.6. Primers used for checking the presence of β-cyanoalanine synthase (A) and

Nitrilase 4 (B) transgenic lines in putatively transformed cassava tissues. The inserts were checked by amplifying a 500 bp region from the end of each insert to the beginning of the nos terminator. β-CAS=β-cyanoalanine synthase NIT4=Nitrilase 4; Nos ter=Nos terminator.

2.2.9 Plant growth in the greenhouse

Cassava plantlets from tissue culture were transferred to greenhouse conditions for further analysis. Greenhouse plants were grown on Fafard potting mix, which is a mixture of peat moss, perlite and vermiculite. To speed up the development of storage roots, plants were grown in the 3 Kord Traditional Square Pot (Kord Products: Canada) with a volume of 230 mL. These small pots promote the rapid development of storage roots that can be analyzed within 4 months. After 4-8 months, storage roots were harvested from these plants and analyzed.

2.2.10 Measurement of yield parameters

Plants were grown in the greenhouse in rectangular trays, each with a capacity of 21 plants. To allow for space between plants, we grew only six plants per tray. Greenhouse grown plants were harvested after 4-8 months of growth and fresh weight measurements were taken on all the tubers to get an estimate of yield. Average fresh weight of tubers per plant per line was calculated.

2.2.11 Free amino acid extraction and analysis

Free amino acid extraction was based on the method by Hacham et al. (2002).

Approximately 150 mg of tissue was ground in liquid nitrogen and homogenized by motor pestle with 600 µL of water: chloroform: methanol (3:5:12 v/v). After centrifugation at 21000 g for 2 minutes, the supernatant was collected and the residue was re-extracted with 600 µL of water: chloroform: methanol followed by centrifugation as before. Supernatants from the first and second extraction were combined in a 2 mL tube. 300 µL of chloroform and 450 µL of water were added to the combined supernatants followed by centrifugation at 21000 g for 2 minutes. The upper water: methanol phase was collected and transferred to a fresh tube, dried by speed-vac to dry

(about 3 hours) and dissolved in 200 µL of water. Detection of free amino acids was performed by the Proteomics & Mass Spectrometry Facility at the Donald Danforth Plant

Science Center using the AccQTag system.

2.2.12 Analysis of IAA

Indole acetic acid (IAA) analysis was carried out using an LC-MS/MS analysis method developed and performed by the Proteomics & Mass Spectrometry Facility at the Donald

Danforth Plant Science Center. The method is similar to (Chen, et. al 2009), but modified to include additional plant hormone species.

2.2.13 Statistical analysis

Statistical analysis was carried out using GraphPad Prism software package

(http://www.graphpad.com/prism/Prism.htm). Student t tests and one-way ANOVA with

Dunnett‟s Multiple Comparison test for comparing multiple lines with the control were used. All analyses for significant differences were performed at P≤0.05.

2.3 Results

2.3.1 Biochemical analysis of cyanide metabolism in cassava

2.3.1.1 Rhodanese is barely detectable in cassava roots

The cyanide detoxifying enzyme rhodanese (cyanide: thiosulfate sulfurtransferase) catalyzes the reaction:

To investigate the role of rhodanese in cyanide detoxification in cassava, we determined enzyme activities in tuberous roots and leaves of the cultivar MCol 2215. The activity was measured spectrophotometrically by analyzing the amount of thiocyanate produced per unit time in mature 8 months-old cassava plants grown in the greenhouse. The average activity of rhodanese detected in the leaves of cassava (8 months old, fresh greenhouse leaves) was 4.19 µM per mg protein per minute (Table 2.1). These rates were

20-fold greater than rates reported for sorghum seedlings by Miller and Conn (1980) of

0.252 µM per mg protein per minute and 200-fold greater than those reported in several plants(Sambucus chinensis, Lathyrus aphaca, Lathyrus sylvestris and Leucaena leucocephala) by Shirai and Kurihara (1991) (0.027 µM per mg protein per minute), For the case of sorghum, the differences in enzyme activity may be attributed to differences in the age of plants used (the sorghum seedlings used by Miller and Conn were 3 days old). Generally, rhodanese activity rates reported in plants were much lower than those

69 reported in animals (Shirai and Kurihara, 1991; Papenbrock and Schmidt, 2000). For example, Papenbrock and Schmidt (2000) compared rates of recombinant Arabidopsis rhodanese (6.27 µM per mg protein per min) with activities of rat liver rhodanese and rhodanese from Azotobacter, and confirmed previously published rates of rat liver and rhodanese (720 and 192 µM per mg protein per min, respectively). Generally, overall rhodanese activity in plants does not seem to correlate with cyanogenesis (Miller and

Conn, 1980). Significantly, we detected no rhodanese activity in cassava tubers, unlike leaves. These results are consistent with previous results from Nambisan and Sundaresan

(1994). Our results suggest that rhodanese is not involved metabolizing cyanide in cassava roots.

Table 2.1. Rhodanese and β-cyanoalanine synthase activity in cassava. All analyses were done in triplicates. Statistical analysis was done using T test at P≤0.05. Numbers in the same row with the same letter superscript are not significantly different. Numbers in the same row with different letter superscripts are significantly different. There was no significant difference between rhodanese and CAS activities in the leaves (P=0.078).

There was no rhodanese activity in roots while CAS root activity increased 2.7-fold. Root and leaf activities were significantly different for both enzymes.

Rhodanese activity CAS activity µM thiocyanate/mg protein/min µM H2S/mg protein/min

Leaves 4.19±0.43a 5.07±0.39a

Roots 0a 13.67±1.53b

2.3.1.2 β-cyanoalanine synthase has higher activity in roots compared to shoots

Previous research has indicated that the cyanogenic glycoside linamarin is produced in the leaves and transported to the roots (Bediako et al. 1981; Ramanujam and Indira, 1984;

Siritunga and Sayre, 2003). In the roots, linamarin is either stored or deglycosylated into acetone cyanohydrin, which spontaneously breaks down, releasing cyanide (White et al.

1998; Siritunga and Sayre, 2003; Siritunga et al., 2004). The cyanide release can be detoxified either via rhodanese or via the enzyme β-cyanoalanine synthase. No rhodanese activity was detected in cassava roots, however (Table 2.1). We hypothesized that cyanide released from the breakdown of linamarin is preferentially assimilated into primary metabolism via the enzyme β-cyanoalanine synthase. To test this hypothesis, we analyzed the activities of β-cyanoalanine synthase in cassava tuberous roots and leaves after 8 months of growth in the greenhouse. In cassava tuberous roots, the average rate of

β-cyanoalanine synthase was 13.67 µM hydrogen sulfide per mg protein per min, compared to 5.07 in the leaves, an approximately 3-fold higher activity in roots than leaves (Table 2.1). These results are contrary to published results for potato (Maruyama et al. 2001), where leaf activity (≈0.04 µM H2S per mg protein per min) was 2-fold greater than root activity (≈0.02 µM H2S per mg protein per min). Non-cyanogenic plants like potato only produce cyanide as a byproduct of ethylene biosynthesis (hence the low values). For cassava assimilation of cyanide derived from linamarin metabolism in roots would presumably require elevated β-cyanoalanine synthase activity. Furthermore, since linamarin is the dominant nitrogen containing compound in the phloem, it likely provides

72 much of the reduced nitrogen for amino acid synthesis in roots (Calatalyud and Ru,

1996). High β-cyanoalanine synthase activity in cassava roots (36 µg H2S/min/100mg protein in roots and 13µg H2S/min/100mg protein in leaves) has also been previously reported (Nambisan and Sundaresan, 1994; Elias et al. 1997). The high rates recorded in our assays are consistent with high β-cyanoalanine synthase rates in cyanogenic plants

(Miller and Conn, 1980; McMahon et al. 1995). We conclude that β-cyanoalanine synthase is the key cyanide metabolizing enzyme in cassava roots and may be responsible for incorporating cyanogens into primary metabolism. Higher root activities compared to shoots are consistent with the metabolism of linamarin in the roots (Siritunga and Sayre,

2003).

2.3.1.3 Nitrilase has increased activity in roots compared to leaves

Nitrilase is involved in the conversion of cyanoalanine, produced by β-cyanoalanine synthase activity, into aspartate, asparagine and ammonia (Piotrowski et al., 2001).

Nitrilase 4 (NIT4) has been shown to have both cyanoalanine hydrolase and asparaginase activities (Piotrowski et al., 2001). To determine the activity of nitrilase in cassava leaves and roots, in vitro plants 5 weeks old were used. The average activity of nitrilase in cassava leaves from these plants was 12.8±1 µM ammonia/mg protein per min while that in the roots was 16.4±1.1 µM ammonia/mg protein per min (Figure 2.7). Thus, root activity was about 1.3 times the activity in the roots.

Nitrilase activity Nitrilase µM/mg protein/min µM/mg

Roots Leaves

Figure 2.7. Nitrilase activity in cassava. Crude protein extract was extracted from 5 week- old cassava leaves and roots. Nitrilase activity was determined as the rate of conversion of cyanoalanine to ammonia per mg of crude protein per min. Data are averages of five trials. T-test analysis was carried out to determine significant differences between leaf and root activities. Root and leaf activities were significantly different at p ≤ 0.05. Error bars show 95% confidence interval.

2.3.1.4 Effect of decreased cyanogen content on nitrate reductase activity

Cyanide assimilation via β-cyanoalanine synthase allows entry of cyanogens into primary metabolism. Thus, cyanogenic glycosides may act as transportable forms of reduced nitrogen as demonstrated in rubber tree by Selmar et al. (1988). To determine if cyanogen content can significantly alter nitrogen metabolism, we assayed the activity of nitrate reductase, which catalyzes the first step in nitrate assimilation, the conversion of nitrate to nitrite (Campbell, 1999) in 5 week old in vitro roots. The presence of low-cyanogen transgenic plants (Cab1 lines) in the Sayre lab (Siritunga and Sayre, 2003) allowed us to compare nitrate reductase activity in low cyanogen versus wild-type lines. The wild-type background for the low cyanogen Cab1 lines was MCol 2215, thus all assays for comparison to low cyanogen lines were carried out with MCol 2215 as the wild-type.

Wild-type (MCol 2215) lines had an average activity of 1.78 µM nitrite/mg protein/ min while Cab1 lines had rates ranging from 4.5 to 5 µM nitrite/ mg protein/minute, 2.5-3X higher than in wild-type lines (Figure 2.8). Comparison of these rates with those previously published for cassava was limited by the fact that most previous assays were done per gram fresh weight, rather than per mg of protein (e.g. Cruz et. Al. 2004). The rates were at least 40 times higher than nitrate reductase activity in 21 day old in vitro rice seedlings (0.042 µM nitrite per mg protein per minute) (Hemalatha, 2002).

Figure 2.8. Analysis of nitrate reductase activity in wild-type (WT) and low cyanogen

(Cab1-1, Cab1-2 and Cab1-3) lines. The assay was conducted on 5 weeks old in vitro plants. The results are averaged from 4 trials. Statistical analysis was done by one-way

ANOVA with Dunnett‟s Multiple Comparison Test. Asterisks above bars indicate significant difference from wild-type.

These data strongly suggest that cyanogens provide reduced nitrogen for the roots and when the cyanogens are substantially reduced in steady-state amounts other enzymatic activities such as nitrate reductase, involved in nitrogen assimilation compensate. We hypothesized that engineering the assimilation of cyanide via the β-cyanoalanine synthase pathway (Figure 2.9) would result in increased availability of reduced nitrogen in the plant.

2.3.2 Overexpression of β-cyanoalanine synthase in cassava

The enzyme β-cyanoalanine synthase (CAS) catalyzes the reaction between cyanide and cysteine to produce β-cyanoalanine, which is further converted by a nitrilase to asparagine, aspartate and ammonia (Blumenthal et al., 1968; Elias et al, 1997; Hatzfeld et al. 2000; Lai et al. 2009). This detoxification pathway results in the assimilation of cyanide into primary metabolism and is available in all higher plants thus far examined

(Blumenthal et al., 1968; Miller and Conn, 1980).

Figure 2.9. Model of cyanide assimilation in plants. β-cyanoalanine synthase converts cyanide to cyanoalanine in the presence of cysteine. Cyanoalanine is converted by a nitrilase to asparagine, which is then converted to aspartate and ammonia by asparaginase. In this way, cyanide can be incorporated into the free amino acid pool of the plant.

Overexpression of β-cyanoalanine synthase (CAS) was done under the control of the patatin promoter to investigate the potential for metabolic engineering of the cyanide assimilatory pathway for enhanced protein content.

2.3.2.1 Generation of transgenic cassava lines expressing CAS

To generate transgenic plants expressing β-cyanoalanine synthase, somatic embryos of cassava (TMS 60444) were transformed with the construct carrying the CAS insert

(experimental procedures, Figure 2.4) using Agrobacterium-mediated transformation.

Plants that survived selection were regenerated into full plants for further micropropagation and screening (Figure 2.10).

Four transgenic lines were generated and confirmed by RT-PCR using primers specific to the CAS insert and nos terminator (CS1F; CATGCTATCACAGGCAATGG and

Nos0329R; GCCAAATGTTTG AACGATCGG amplifying about 500 bp region from the end of the CAS gene to the beginning of the NOS terminator). For normalization, α- tubulin was used as a control (Figure 2.12). The confirmed transgenic plants were named

PCAS1, PCAS2, PCAS3 and PCAS4. In comparison with the α-tubulin controls, expression of β-cyanoalanine synthase was very low in transgenic plants. There was no transgene expression in the wild-type (WT, Figure 2.14) as expected.

Figure 2.10. Recovery of putatively transgenic plants after transformation with the CAS gene. Embryogenic callus from cassava was transformed by Agrobacterium-mediated transformation. Putative transformants were selected on a medium containing 45 µM paromomycin.

Figure 2.11. CAS transcripts in wild-type (WT) and transgenic (PCAS1-4) cassava lines as detected by RT-PCR.. RNA was extracted from 100 mg of 5 week-old in vitro cassava roots (wild-type and CAS putative transgenic plants). RT-PCR was performed using primers for the CAS insert while tubulin primers were used for the control. Low expression was observed in transgenic lines (indicated by faint bands). CAS transgene is expressed in the transgenic lines (PCAS1-4) but not in the wild-type (WT).

2.3.2.2 Expression of CAS results in poor root development in cassava

Significantly, there was a very low recovery of transgenic plants expressing β- cyanoalanine synthase relative to transformations with other gene constructs. In addition, the recovered transgenic lines had poor root growth and were generally stunted in growth

(Figure 2.12).

Analysis of transgenic CAS plants in vitro showed poor root development compared to wild-type plants (Table 2.2). Wild-type plants had an average of 4 roots per plant at 3 weeks of growth on Murashige and Skoog medium, while CAS transgenics ranged from

1 to 3 roots per plant. Root development was poorest in PCAS2 and PCAS3. Poor root development corresponded with reduced fresh weight and poor growth (Table 2.2, Figure

2.12). There was a 2 to 4 fold decrease in fresh weight in transgenic CAS plants (with the exception of PCAS4). Interestingly, the transgenic plants exhibiting the highest CAS expression based on RT-PCR measurements had the lowest fresh weight (Figure 2.12,

Table 2.2). These data suggests that overexpression of β-cyanoalanine synthase impairs root development in cassava plants.

Figure 2.12. In vitro growth comparison of transgenic β-cyanoalanine synthase plants

(PCAS1-4) and wild type (TMS 60444) plants after 3 weeks. Transgenic CAS plants were stunted in growth with poor root development.

Table 2.2. Root development and fresh mass in in vitro transgenic CAS plants grown in

Murashige and Skoog medium for 3 weeks. Data are averages of n=20. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison Test.

Asterisks in transgenic lines indicate significant difference from wild-type at p ≤ 0.05.

Wild-type PCAS1 PCAS2 PCAS3 PCAS4

Number of roots 4±1.55 1.9±0.8* 1.1±0.7* 1±0.5* 2.7±0.5*

Fresh weight mg 103.0±27 60.00±14* 25.63±9* 26.83±10* 88.67±18

2.3.2.3 Expression of CAS increases the activity of β-cyanoalanine synthase in

cassava roots

To determine if expression of β-cyanoalanine synthase increased the activity of the enzyme in cassava roots, enzyme assays were carried out as previously described. First, we analyzed the activity in 4 week-old in vitro plants grown on regular Murashige and

Skoog medium supplemented with 2% sucrose. Except for PCAS4, β-cyanoalanine synthase activity in transgenic CAS roots (Table 2.3) was significantly higher (1.6-2- fold) than the activity in wild-type roots. The highest increase was recorded in PCAS2, which also had the poorest growth rate in vitro (Figure 2.11).

Two of the transgenic lines (PCAS1 and PCAS2) were transferred to the greenhouse for further analysis. The activity of β-cyanoalanine synthase was determined after 4 months of growth under greenhouse conditions (Table 2.3). These rates were on average two- fold higher than those recorded for in vitro plants, although the trends remained the same for the lines analyzed. PCAS1 had a 1.4-fold increase in activity of β-cyanoalanine synthase while PCAS2 had twice as much activity. Thus, root-specific expression of β- cyanoalanine synthase increases the activity of the enzyme in cassava roots.

Table 2.3. Expression of β-cyanoalanine synthase increases the activity of the enzyme in cassava roots. Greenhouse plants used in these assays were 4 months old while in vitro plants were 4 weeks old. Rates are in µM H2S/mg protein/min. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison Test. Numbers in the same row with the same letter superscript are not significantly different (at p≤0.05).

Numbers in the same row with different letter superscripts are significantly different. All enzyme assays were performed in triplicates.

WT PCAS1 PCAS2 PCAS3 PCAS4

In vitro roots 2.1±0.34a 3.42±0.33b 4.64±0.17c 3.83±0.24b 2.12±0.33a

Greenhouse 4.5±1.1d 6.51±0.44e 8.93±1.14f - - roots

2.3.2.4 High activity of CAS in transgenic plants correlates with stunted growth

and low yield

Most of the embryos transformed with the CAS gene died and the recovery of transgenic plants was low. Two early transgenic lines PCAS1 and PCAS2 were monitored under in vitro and greenhouse conditions. PCAS2 showed more stunted growth both in vitro and in the greenhouse after 4 months. In addition, PCAS2 showed a much greater decrease in tuber fresh weight per plant (36% of wild-type) compared to PCAS1 (92% of wild-type).

However, there was no significant difference between the shoot fresh weight of PCAS1 and that of PCAS2 (Table 2.3). These data suggest that high transgenic expression of

CAS is detrimental to the plant, indicating that only weakly expressing CAS transgenic plants survived.

Table 2.4. Fresh weight analysis of wild-type and transgenic cassava plants (n=20). The data are ±SDs. Statistical analysis was done by one-way ANOVA with Dunnett‟s

Multiple Comparison Test. Numbers in the same row with the same letter superscript are not significantly different (at p≤0.05). Numbers in the same row with different letter superscripts are significantly different.

WT PCAS1 PCAS2

Tuber fresh 64±12.4a 58.9±12.3a 22.8±8.6b weight per plant (g )

Shoot fresh 47.1±5.8c 34.9±5.5d 32.7±7.9d weight (g)

2.3.2.5 Expression of CAS increase amino acid pool sizes in cassava roots

Since the enzyme β-cyanoalanine synthase is involved in assimilation of cyanide released from ethylene biosynthesis or the breakdown of cyanogenic glycosides, we hypothesized that transgenic expression of CAS would increase amino acid pool sizes in cassava roots.

Hydrolyzed amino acid analysis was carried out at the Proteomics & Mass Spectrometry

Facility at the Donald Danforth Plant Science Center, using 4mg of dried root tissue.

There was a significant increase (at P≤0.05) in total amino acids between wild-type and transgenic plants (Table 2.5). Individual amino acids also showed significant increases, particularly in PCAS2, which had the highest activity of β-cyanoalanine synthase. The β- cyanoalanine synthase pathway produces aspartate and asparagine. In hydrolyzed amino acid analysis, asparagine is hydrolyzed to aspartate, which increased significantly in the transgenic lines (Table 2.5).

Table 2.5. Hydrolyzed amino acid analysis in transgenic and wild-type cassava roots in pmole/mg dry weight. Superscript „a‟ indicates significant difference from wild-type at

P≤0.05. Superscript b indicates significant difference between the two transgenic lines at

P≤0.05. CyA=cysteic acid (hydrolysis product of cysteine). Asp = Asp+Asn (due to hydrolysis of Asn to Asp).

Amino acid WT PCAS1 PCAS2 CyA 1.30±0.05 1.49±0.19 1.40±0.08 His 0.56±0.17 0.7±0.22 0.7±0.2 Ser 1.63±0.4 2.08±0.42 2.19±0.28a Arg 1.30±0.21 2.68±0.54a 1.83±0.25ab Gly 4.01±0.18 4.87±1.01 5.56±0.45a Asp 6.56±0.78 9.12±0.92a 8.43±0.86a MetS 1.64±0.25 1.74±0.3 1.86±0.11 Glu 7.27±0.35 9.76±1.07a 9.92±0.88a Thr 1.62±0.32 1.98±0.51 2.25±0.15a Ala 3.93±0.28 4.49±1 4.96±0.34a Pro 2.05±0.17 2.41±0.56 2.67±0.3 Lys 3.01±0.16 3.62±0.69 3.87±0.19a Val 2.78±0.19 3.32±0.74 3.64±0.28a Ile 2.01±0.11 2.35±0.48 2.58±0.17a Leu 3.0±0.16 3.51±0.73 3.89±0.26a Phe 1.62±0.09 1.78±0.32 1.86±0.14a TOTAL 44.3 55.9a 57.6a

2.3.2.6 Total protein analysis in transgenic plants expressing β-cyanoalanine

synthase

For total protein analysis, lyophilized root samples from 4 moths-old plants were ground using motor and pestle and total protein was extracted in a buffer containing 50mM Tris-

HCl (pH 8.5), 5mM dithiothreitol (DTT) and 1mM EDTA. The protein was quantified using by the Bradford method with bovine serum albumin as the standard. The total protein in the wild-type was 16.73 mg protein per mg dry weight. Total protein increased

9.3 % in PCAS2 plants (18.28 mg protein per mg dry weight) relative to wild type (Table

2.6).

Table 2.6. Total protein comparison in wild-type and transgenic cassava roots. Numbers in the same row with the same letter superscript are not significantly different at P≤0.05.

Numbers in the same row with different letter superscripts are significantly different.

Measurements were performed in triplicates.

Cassava WT PCAS1 PCAS2

Line

Total 16.73±0.56a 17.36±0.24a 18.28±0.17b protein mg/mg dry weight

2.3.3 Expression of Nitrilase 4 in cassava

During cyanide assimilation via the CAS pathway, the cyanoalanine produced by the action of β-cyanoalanine synthase is converted into aspartate, asparagine and ammonia by nitrilase. The Arabidopsis NIT4 orthologue has been shown to have high substrate specificity for cyanoalanine (Piotrowski et al. 2001). We hypothesized that expression of the Arabidopsis NIT4 gene in cassava would increase the turnover of cyanoalanine from cyanide metabolism, resulting in increased assimilation into amino acids.

2.3.3.1 Generation of transgenic plants expressing NIT4

Somatic embryos of cassava (TMS 60444) were transformed with the construct carrying the NIT4 insert using Agrobacterium-mediated transformation. Plants that survived selection were allowed to develop roots and develop into full plants for further micropropagation and screening. Five transgenic lines were confirmed by RT-PCR

(Figure 2.13) using primers specific to the NIT4 insert and nos terminator (CS1F;

GCACTTGAGGGTGGATGTTT and Nos0329R; GCCAAATGTTTG AACGATCGG) amplifying about 500bp from the end of the CAS gene to the beginning of the NOS terminator. For normalization, α-tubulin was used as a control. The confirmed transgenic plants were named PNIT1-6.

Figure 2.13. NIT4 transcripts by RT-PCR. RNA was extracted from 100 mg of 5 week- old in vitro cassava roots. RT-PCR was performed using primers for the NIT4 insert while tubulin primers were used for the control. There was no expression of transgene in wild-type lines (WT).

2.3.3.2 Expression of NIT4 increases cyanoalanine hydrolase activity in cassava

roots

Nitrilase 4 converts β-cyanoalanine to aspartate, asparagine and ammonia (Figure 2.8).

To determine whether overexpression of NIT4 in cassava roots increased nitrilase activity, enzyme assays were carried out as previously described (see experimental procedures). Three transgenic lines were chosen for this analysis. Briefly, 400 µg of crude protein extract from four months-old tuberous cassava roots from the greenhouse was added to 10 mM final concentration of cyanoalanine in a buffer containing 50 mM

Tris-HCl (pH 8.5) and 1.0 mM DTT. The reaction was run for 10 minutes. Ammonia

94 produced was quantified using the Berthelot reaction as described by Piotrowski et al.

(2001).

Nitrilase activity in wild-type roots was 13 µM ammonia/mg protein/min. Transgenic lines PNIT2 and PNIT4 had a 4-fold increase in activity (41 and 30 µM ammonia/mg protein/min.

Figure 2.14. Expression of Nitrilase increases cyanoalanine hydrolase activity in cassava roots. Rates of conversion of cyanoalanine to ammonia were determined for n=4.

Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison

Test. Asterisks above bars indicate significant difference from the wild-type (WT) at p ≤

0.05. Error bars show 95% confidence interval.

2.3.3.3 Expression of NIT4 causes branching in cassava plants

Transgenic lines overexpressing NIT4 (herein called NIT4 plants) were grown in the greenhouse for analysis. Transgenic NIT4 plants displayed an increased branching phenotype compared to wild-type plants (Figure 2.14, Table 2.7). In addition, in the early stages of growth (up to 8 weeks) in the greenhouse, NIT4 plants tended to have more fibrous root development compared to wild-type plants. Also in the first 4 months of growth in the greenhouse, transgenic NIT4 plants had smaller leaves compared to wild type, which was compensated for by doubling in number (after 12 weeks, the average number of leaves in wild type plants was 21 while that in nitrilase transgenic plants was

47).

Some of these phenotype traits are potentially consistent with changes in auxin levels in plants. Auxins are involved in regulation of root and shoot growth. For example, auxin promotes cell division in root pericycle cells, which leads to lateral root formation, but inhibits cell division in lateral meristems of the shoot, resulting in the inhibition of lateral bud growth, or apical dominance (Rogg et al. 2001). The branching phenotype in our transgenic plants is similar to what would be observed on decapitation of apical meristem, which removes the inhibition of lateral bud growth (apical dominance) and results in branching. Decapitation reduces endogenous levels of indoleacetic acid (IAA), the most common bioactive form of auxin (Ferguson and

Beveridge, 2009). While a significant amount of root auxin is derived from the shoot, it 97 is now known that roots are also significant sites of auxin biosynthesis (Ross et al. 2006).

Since nitrilases are also known to be involved in auxin biosynthesis, and based on the phenotypes described above, we hypothesized that NIT4 overexpression affected auxin metabolism in cassava roots. To test this, we measured IAA concentrations in cassava roots of wild-type and NIT4 transgenic plants.

Figure 2.15. Branching in cassava plants expressing the Nitrilase 4 gene. 60444=wild- type, PNIT1, 3= NIT4 transgenic lines.

Table 2.7. Average number of branches per plant in 12 week old cassava plants in the greenhouse. Values are ±standard deviation where n=5. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison Test. Asterisks indicate significant difference from wild-type at p≤0.05.

Cassava line Average number of branches

Wild-type 1±0.5

PNIT1 4±1.1*

PNIT2 5±0.9*

PNIT3 5±0.9*

PNIT4 5±0.9*

Our assays for IAA, performed after 4 months of growth in the greenhouse, gave results in the range 7-17 ng/g fresh weight in both leaves and roots of cassava. Similar ranges have been reported in literature for Arabidopsis 9 day old seedlings (13 ng/g fresh weight) and tomato (6-12 ng/g fresh weight) (Tam et al. 2000; Negi et al. 2010). After 4 months, transgenic cassava plants expressing NIT4 had up to 50% less root IAA compared to wild-type plants (Table 2.8). Wild-type (TMS 60444) roots, with ≈1 6ng/g fresh weight, had twice amount of free IAA in transgenic plants.

100

Table 2.8. IAA analysis in wild-type and transgenic cassava roots and leaves. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison Test.

Asterisks in the columns indicate significant difference from the wild type at p≤0.05.

Root IAA ng/ g fresh Leaf IAA ng/g fresh weight

weight

Wild Type 15.93±2 16.76±0.8

PNIT1 9.53±1.7* 11.45±2.5*

PNIT2 7.33±1.4* 8.21±1.6*

PNIT3 10.58±2* 11.67±3.5

PNIT4 11.94±0.7 13.78±2.9

101

In the leaves, transgenic plants also had reduced IAA concentration, with PNIT2 having

50% less than wild-type levels (Table 2.8).

The reduction in IAA concentration may explain the branching observed. The nitrilase family of genes is involved in a number of reactions in plants, including auxin biosynthesis and cyanide metabolism (Oreilly and Turner, 2003). Arabidopsis NIT1,

NIT2 and NIT3 convert indole-3-acetonitrile (IAN) to the plant hormone indole-3-acetic acid (IAA) in vivo (Bartling et al., 1992; Bartling et al., 1994; Schmidt et al., 1996).

Arabidopsis nitrilase 4 (NIT4) is known to be involved in cyanide metabolism, accepting

β-cyanoalanine as a substrate but not IAN (Piotrowski et al., 2001). However, Zea mays

NIT2, a homolog of Arabidopsis NIT4, recognizes IAN as substrate unless it forms heteromers with ZmNIT1 in which case it has β-cyanoalanine hydrolase activity

(Kriechbaumer et al. 2007). There is some evidence that in some species, complex formation is a requirement for β-cyanoalanine hydrolase activity in nitrilases, which otherwise recognize IAN (Kriechbaumer et al. 2007; Jenrich et al., 2007). It is not likely that transgenic NIT4 had increased IAA biosynthetic activity, since IAA actually decreased in the transgenic plants. A possible explanation for these results is that overexpression of NIT4 led to increased heteromer or homomers formation, resulting in a switch in substrate recognition by endogenous nitrilases from IAN to β-cyanoalanine.

The switch may have impaired IAA biosynthesis, resulting in reduced levels of IAA observed.

102

2.3.3.4 Effect of NIT4 overexpression on the yield of cassava roots

Vegetative development and reduced root development in NIT4 transgenic lines led to low tuber yield in the initial stages. However, after 4 months, partitioning shifted to the tubers and tuber development increased. After eight months, there was no significant difference in fresh weight in wild-type and transgenic lines suggesting no correlation between IAA and fresh weight of tuberous roots (Table 2.9).

Table 2.9. Effect of NIT4 overexpression on tuber fresh weight in cassava. The data are ± standard deviation. Statistical analysis was done by one-way ANOVA with Dunnett‟s

Multiple Comparison Test. There were no significant differences at P≤0.05 between wild-type and transgenic plants.

WT PNIT1 PNIT2 PNIT6

Tuber fresh weight g per 50.9±9.2 57.5±1.6 30.2±10.3 41.2±6 plant

103

2.3.3.5 Free amino acid analysis in transgenic NIT4 plants

To determine if expression of NIT4 in cassava roots increased amino acid pool sizes, free amino acid analysis was carried out. The range of total free amino acid content in the analysis was between 5 and 14.5 pmole/mg fresh weight. There were no definitive effects of NIT4 overexpression on amino acid pool size based on CAS hydrolysis. In some transgenic lines there were significant decreases while in others there were significant increases (Table 2.10). These results do not support an effect of nitrilase overexpression on free amino acid pool sizes. The effect on auxin metabolism might be masking any effect due to assimilation of cyanoalanine into free amino acids.

104

Table 2.10. Total amino acid and protein analysis in wild-type and NIT4 transgenic lines.

The data are averages of 3 trials. The data are ± standard deviation. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple Comparison Test. Asterisks indicate significant difference with wild-type at P≤0.05.

Line Amino acids (pmole per mg fresh Total protein mg/g fresh

weight) weight

WT 9.33±0.22 4.22±0.15

PNIT1 10.25±0.32* 3.13±0.05*

PNIT2 5.49±0.63* 0.38±0.05*

PNIT3 8.72±0.08* 2.25±0.13*

PNIT4 11.3±0.44* 2.09±0.03*

105

2.3.3.6 Total protein analysis in transgenic NIT4 lines

NIT4 transgenic lines generally showed reduction in total protein content (Table 2.10).

PNIT2 had the lowest protein concentration. Total protein appears to correlate to IAA concentration suggesting that changes in total protein were cause by changes in auxin metabolism rather than cyanide metabolism.

Taken together, the results from the analysis of nitrilase transgenic plants suggest that overexpression of nitrilase in cassava roots altered auxin metabolism, masking any contributions to reduced nitrogen via the metabolism of cyanoalanine synthase.

106

2.4 Discussion

Cyanogenic plants allocate a substantial amount of nitrogen to cyanogenic glycoside accumulation. For example, Eucalyptus cladocalyx allocates up to 20% of leaf nitrogen to the cyanogenic glycoside prunasin (Gleadow et al. 1998). Cyanide assimilation via the β- cyanoalanine synthase pathway allows cyanide and cyanogenic glycosides to act as alternate forms of reduced nitrogen (Nartey, 1969; Selmar et al. 1988; Siritunga and

Sayre, 2004; Ebbs et al. 2010). We detected high activity of root β-cyanoalanine synthase and nitrilase activity in cassava roots. Since these enzymes are involved in cyanide assimilation, the data supports the hypothesis that cyanide assimilation via β- cyanoalanine synthase takes place in vivo. Elias et. al. (1997) also reported higher β- cyanoalanine synthase activity in tuberous roots compared to leaves. The cyanide detoxifying enzyme rhodanese, prevalent in mammals, was not active in cassava roots.

Since rhodanese produces thiocyanate while β-cyanoalanine synthase leads to production of amino acids and ammonia, it is likely that a pathway that allows re-entry of cyanide into primary metabolism is the preferred pathway for cyanide metabolism in cassava roots. While there was no rhodanese activity in the roots, there was no significant difference between rhodanese and β-cyanoalanine synthase activity in leaves. Whether cyanide detoxification is the physiological role of the enzyme in the leaves remained to be determined. The metabolic fate of thiocyanate in plants has not been defined.

A reduction in linamarin synthesis in leaves would also be expected to impact nitrogen assimilation in plants. The high nitrate reductase activity in low cyanogen plants (Figure

107

2.6) suggests that substantial reduction (99%) of cyanogens in cassava roots reduces nitrogen available for amino acid synthesis in roots. The plants apparently compensate for reduced linamarin levels by increasing nitrate reduction activity in roots. Indeed, growth of low cyanogen plants is poor without ammonia in the media (Siritunga and

Sayre, 2004). Over 60% of the reduced nitrogen in stem phloem exudates of cassava is in the form of linamarin (Calatalyud and Ru, 1996). Thus, cyanogenesis in crop plants presents a challenge; while toxic to humans, it has an important role in primary nitrogen assimilation. Redirection of cyanogenic glycoside metabolism towards incorporation into amino acids is therefore a more viable option to blocking linamarin synthesis. This requires suﬃcient linamarin transport to support amino acid synthesis without linamarin accumulation, as well as the partition of all or most of the linamarin transported to the roots towards amino acid synthesis (Siritunga et al. 2004).

Expression of β-cyanoalanine synthase in cassava roots increased CAS activity and root total amino acid levels, although it was associated with poor root development, growth and reduced fresh weight. Recently, Garcia et al. (2010) have shown that mitochondrial

β-cyanoalanine synthase in Arabidopsis is essential for maintaining low-levels of cyanide essential for root hair development. Mutants of Arabidopsis β-cyanoalanine synthase, which accumulate cyanide, were defective in root-hair development. This phenotype could be rescued by addition of hydroxocobalamin, a cyanide antidote, indicating that the effect of β-cyanoalanine synthase mutants on root hair development was due to cyanide accumulation (Garcia et al. 2010). Hydroxocobalamin is an endogenous vitamin B12 precursor with a higher affinity for cyanide than cytochrome oxidase; it binds cyanide to

108 its cobalt moiety in equimolar amounts forming non-toxic cyanocobalamin (Shepherd and Velez, 2008).

The effect of cyanide on root development has been previously studied (e.g. Smithers and

Sutcliffe, 1967), but these have been restricted to the inhibitory effect of cyanide on root growth through inhibition of respiration. The results in this chapter and those of Garcia et al. (2010) suggest that β-cyanoalanine synthase is involved in regulation of root development, possibly by regulating the amount of free cyanide available in the roots.

Cyanide assimilation through the enzyme β-cyanoalanine synthase is also closely related to cysteine synthesis (Figure 2.16). The cysteine produced from the cysteine synthesis can be used in cyanide detoxification, while the hydrogen sulfide from cyanide detoxification can be used in cysteine synthesis. In addition, cysteine synthase, which catalyzes formation of cysteine from O-acetyl-L-serine (OAS) and disulfide, also possesses CAS activity in plants while CAS has detectable cysteine synthase activity

(Hasegawa et al. 1995; Maruyama et al. 1998; Hatzfeld et al. 2000). Thus, β- cyanoalanine synthase or its products may play a significant role in regulating sulfur metabolism in plants. Hydrogen sulfide is a central molecule in sulfur metabolism and is toxic to plants at high concentrations as it inhibits respiration. The overproduction of hydrogen sulfide in CAS transgenic plants may act as a repressor of sulfate uptake since sulfate uptake is repressed in a negative feedback loop when reduced sulfur is available to plants (Davidian and Kopriva, 2010). However, free amino acid analysis did not show any significant differences in cysteine between wild-type and transgenic plants.

109

Figure 2.16. Relationship between cyanide detoxification and cysteine biosynthesis in plants. β-cyanoalanine synthase uses cysteine to detoxify cyanide. The sulfide released from the action of β-cyanoalanine synthase can be fixed by cysteine synthase to produce cysteine, itself a substrate of β-cyanoalanine synthase. After Hell and Wirtz, 2008.

110

Expression of Arabidopsis NIT4 resulted in reduction of IAA in roots and leaves. At least three nitrilase homologues, NIT1, NIT2 and NIT3 are known to be involved in auxin biosynthesis. They convert indole-3-acetonitrile (IAN) to the plant hormone indole-3- acetic acid (IAA) in vivo (Bartling et al., 1992; Bartling et al., 1994; Schmidt et al.,

1996). NIT4 however has been reported to not only to have high substrate specificity for cyanoalanine, but also not to recognize IAN as a substrate (Piotrowski et al., 2001;

O‟Reilly and Turner, 2003). Despite this, NIT4 is also thought to be involved in IAA biosynthesis (Piotrowski et al., 2000). In maize, a dual role in auxin homeostasis and cyanoalanine hydrolysis by the NIT1/NIT2 heteromers has been reported (Kriechbaumer et al. 2007). On its own, ZmNIT2 hydrolyses IAN to IAA, and Zmnit2 knockout mutants accumulate lower quantities of IAA conjugates in the kernels and roots of young seedlings, indicating that an IAN-dependent pathway contributes substantially to auxin biosynthesis in these tissues (Kriechbaumer et al. 2007). ZmNIT1/ZmNIT2 heteromers are also involved in cyanide detoxiﬁcation via β-cyanoalanine turnover (Kriechbaumer et al. 2007). Interestingly, ZmNIT2 bears closer homology to Arabidopsis NIT4 than to the other three Arabidopsis nitrilases (Park et al. 2003; Kriechbaumer et al. 2007).

Arabidopsis NIT4 has a 66-68% amino acid identity with the other three nitrilases (NIT1-

3), which share 84-90% amino acid identity (Janowitz et al. 2009) and about 74% identity with Zea mays NIT2 (Figure 2.17).

111

Figure 2.17. Analysis of amino acid sequence similarity between Arabidopsis NIT4

(AtNIT4) and Zea mays NIT2 (ZmNIT2). ZmNIT2 bears closer homology to Arabidopsis

NIT4, with a 74% amino acid identity.

112

In other grasses, such as sorghum and rice, no enzymatic activities were detected for homomers of members of the AtNIT4 group (Kriechbaumer et al. 2007).

However, nitrilase heteromers, composed of ZmNIT1 and ZmNIT2 homologues, hydrolysed β-cyanoalanine (Jenrich et al., 2007). Thus, in some species, formation of heteromers appears to be a requirement for β-cyanoalanine hydration by nitrilase. The data observed for maize nitrilases, and the surprising auxin-related phenotype in our transgenic plants, open up the possibility that shifting substrate recognition from IAN to

β-cyanoalanine is part of a regulatory mechanism linking IAA biosynthesis and the degradation of cyanide (Kriechbaumer et al. 2007). The reduction in IAA levels in transgenic NIT4 cassava plants may be due to a shift in substrate recognition from IAN to cyanoalanine, resulting in reduced IAA biosynthesis in the roots. This would happen if transgenic expression of Arabidopsis NIT4 resulted in formation of heteromers with cassava homologs, which function like the Zea mays ZmNIT1/ZmNIT2 heteromers. The lack of a significant increase in free amino acid content may suggest that hydration of cyanoalanine is not the rate-limiting step in the assimilation of cyanide to amino acids via the β-cyanoalanine synthase pathway.

While the two strategies explored here for enhancing cyanide assimilation had shortcomings, one of them (expression of CAS) resulted in increased total amino acid and protein content correlating with increased activity of β-cyanoalanine synthase, suggesting conversion of cyanogens to free amino acids. Whether this pathway is the central mechanism for incorporating cyanogens into primary metabolism remains to be seen.

113

Transgenic approaches based on this pathway seem to be limited in their potential to redirect root linamarin towards protein production by poor growth and survival of transgenic plants (CAS) and alteration of auxin metabolism (NIT4). Siritunga et al.

(2004) hypothesized that acetone cyanohydrin, which is produced by root linamarase activity, may be directly assimilated into amino acid(s). This is especially desirable since roots lack HNL and presumably would have slow rates of spontaneous conversion of acetone cyanohydrin to cyanide (White et al. 1998; Siritunga et al. 2004). In sorghum, a nitrilase 4 complex was recently found, responsible for the endogenous turnover of dhurrin proceeding via 4-hydroxyphenylacetonitrile, avoiding release of hydrogen cyanide (Jenrich et al. 2007). It is therefore possible that while non-cyanogenic plants utilize the β-cyanoalanine synthase pathway for cyanide detoxification/assimilation, cyanogenic plants may have evolved a mechanism for converting cyanogens to amino acids without intermediate toxic cyanide.

In addition, the failure of CAS overexpression to significantly increase protein may be related to the lack of storage proteins in cassava (Shewry, 2003). Storage proteins act as nitrogen sinks facilitating accumulation under conditions of excess nutrient supply

(Shewry, 2003). Recently, Abhary et at. (2011) have demonstrated that transgenic expression of the storage protein zeolin in cassava roots resulted in a four-fold increase in total protein. Similar increases have also been recorded from the use of sporazein (Leyva-

Guerrero, 2011).

Our results show that cyanogens have an effect on cyanide metabolism. Successful redirection of cyanogens into free amino acids will require additional studies in cyanide

114 metabolism as well as coupling the strategies with expression of storage proteins. The transgenic lines generated here, expressing CAS and NIT4 will be important tools in further dissecting cyanide metabolism and its relationship with auxin metabolism and regulation of cysteine synthase in cassava, as well as the possible role of β-cyanoalanine synthase in regulating root development.

115

Chapter 3: The mechanism and control of postharvest physiological

deterioration in cassava

3.1 Introduction

The potential of cassava as a market and food security crop is limited by its short shelf life of only two to three days. Harvested cassava tuberous roots undergo rapid postharvest physiological deterioration (PPD), which reduces the quality of storage roots for market and consumption (Booth, 1976). Cassava processing facilities, if available, must therefore be at or near the site of production in order to reduce postharvest losses. In most smallholder settings, this is not a practical consideration.

PPD in cassava is initiated by mechanical damage, which generally occurs during harvesting operations rendering tuberous roots unpalatable (Buschmann et al., 2000).

This deterioration is an active process distinct from the secondary deterioration resulting from microbial infection (Booth, 1976). It progresses from the site of damage and is characterized by vascular streaking resulting from occlusions in the vascular parenchyma

(Wenham, 1995). Environmental factors affecting the rate of deterioration include temperature, humidity and oxygen. Careful manipulation of these conditions can help in managing PPD. For example, storage at 10°C and 80% humidity, waxing and careful

116 avoidance of physical damage can all delay PPD significantly (Wenham, 1995; Rickard,

1985; Plumbey and Rickard, 1991). While storage conditions and practices can go a long way in controlling PPD however, the development of high shelf-life varieties of cassava remains one of the most important goals of cassava breeding and biotechnology.

Several studies have been carried out to understand the biochemistry of PPD (e.g.

Buschmann et al., 2000; Reilly et al., 2001; Reilly et al., 2004). The emerging picture from these studies places reactive oxygen species (ROS) at the center of the process. In plants, ROS are continuously produced as byproducts of aerobic respiration (Apel and

Hirt, 2004). Under normal conditions, the plant has several mechanisms to scavenge the

ROS, preventing or ameliorating their toxicity. Under conditions of stress however, the equilibrium between production and scavenging of ROS is disturbed. Environmental stresses increase ROS production, disrupting the redox balance and causing oxidative stress (Apel and Hirt, 2004; Desikan et al. 2004; Miller et al. 2010). The production and accumulation of ROS is one of the earliest detectable events during abiotic stress

(Brosche et al. 2010). Abiotic factors such as high salt, flooding, temperature and light have been shown to accelerate production of reactive oxygen species in plants (Mittler,

2002; Rao, 2005; Bailey-Serres and Chang 2005; Miller et al. 2008; Van Breusegem et al. 2008). The sudden build-up of ROS is known as an oxidative burst (Apostol et al.,

1989).

In cassava roots, an oxidative burst occurs within 15 minutes of harvest (Reilly et al.,

2004). Also among the early events is increased activity of enzymes that modulate ROS, such as catalase, peroxidase and superoxide dismutase (Reilly et al., 2001; Buschmann et

117 al., 2000). Further evidence in support of a role of oxidative stress in PPD comes from the observation that cassava cultivars that have high levels of β-carotene (which quenches

ROS) are less susceptible to PPD (Sanchez et al., 2005). Earlier studies have reported decline in phospholipid content during PPD, indicating membrane degradation, a known symptom of oxidative damage (Wenham, 1995).

Early events that trigger the oxidative burst have not yet been fully investigated, however. In this study, we examine the role of cyanide, released due to mechanical damage during harvesting of cassava, in initiating the oxidative burst observed leading to the onset of PPD. We propose that the released cyanide inhibits mitochondrial cytochrome oxidase in the electron transport chain, resulting in the production and accumulation of reactive oxygen species. Further, we show that overexpression of alternative oxidase (a cyanide-resistant terminal oxidase in plant mitochondria) in cassava storage roots reduces accumulation of ROS and delays PPD by at least two weeks. An increase in shelf-life of at least two weeks allows cassava farmers to transport and process the crop without incurring heavy postharvest losses. In addition, since overexpression of alternative oxidase (AOX) reduces accumulation of ROS, we tested the hypothesis that transgenic plants overexpressing AOX have enhanced tolerance to abiotic stress.

118

3.2 Experimental procedures

3.2.1 Tissue culture propagation of plant material

Cassava plants were propagated in vitro in Murashige and Skoog (MS) basal medium

(Murashige and Skoog, 1962) supplemented with Gamborg vitamins (Gamborg et al.,

1968) and 2% (w/v) sucrose. In vitro plants were propagated in growth chambers at 28°C with 16 hours of light and 8 hours of darkness. Two wild-type lines were used (MCol

2215 and TMS 60444) as well as three previously generated low cyanide transgenic lines,

Cab1-1, Cab1-2 and Cab1-3 in cultivar MCol 2215(Siritunga and Sayre, 2003). In vitro cultures were maintained by subculturing nodal stems every four to eight weeks depending on the experiment.

3.2.2 Detection of Reactive Oxygen Species by H2DCF-DA

Accumulation of reactive oxygen species (ROS) was measured using the method described by Maxwell et al. (1999). The method utilizes 2‟, 7‟-dichlorofluorescein diacetate (H2DCF-DA; Molecular Probes), which emits green fluorescence on reaction with reactive oxygen species. Stock solutions (15 µM) of H2DCF-DA were made in dimethyl sulfoxide (DMSO) and used immediately for the assays. Sections (1-2 mm) of in vitro cassava roots were incubated in H2DCF-DA for 30 minutes, after which they were washed with distilled water and immediately analyzed on a Zeiss LSM 510 laser confocal microscope (Carl Zeiss Inc.; North America) with excitation and emission

119 wavelengths of 488 nm and 520 nm, respectively. Fluorescence intensity was quantified using ImageJ image processing software (NIH; http://rsbweb.nih.gov/ij/).

3.2.3 Detection of hydrogen peroxide in cassava roots

Hydrogen peroxide production was measured by an endogenous peroxidase staining procedure described by Rea at al., (2004). This procedure uses 3, 3-diaminobenzidine

(DAB) which forms a reddish-brown precipitate on exposure to hydrogen peroxide. The

DAB solution was prepared by dissolving 10 mg pellets of 3, 3-diaminobenzidine in water to a concentration of 1.0 mg/mL. The pH of the solution was adjusted to 3.8.

Sectioned cassava in vitro roots were incubated in 500 µL of 1.0 mg/mL DAB in closed microfuge tubes for 6-18 hours under light after which they were analyzed for hydrogen peroxide production. Images were captured under the Olympus DP20 light microscope

(Olympus; PA). Quantitation of the intensity of coloration was done on captured images using ImageJ image processing software (NIH; http://rsbweb.nih.gov/ij/).

3.2.4 Inhibition of NADPH oxidase and mitochondrial electron transport chain

Experiments to inhibit NADPH oxidase in cassava roots were conducted using diphenyl iodonium chloride (DPI) as described by Orozco-Cárdenas et al. (2001). For this treatment, sectioned cassava roots were pre-treated with 100 µM DPI in water for 30

120 minutes before analysis of reactive oxygen species by H2DCFDA fluorescence. Controls were treated with water for the same period.

In experiments to complement cyanide in low cyanide transgenic cassava lines, Cab1-1,

Cab1-2 and Cab1-3 transgenic lines were pre-treated with 5.0 mM potassium cyanide

(dissolved in water) for ten minutes prior to H2DCFDA treatment. Control plants were pre-treated with water for the same period.

3.2.5 Design of alternative oxidase construct for cassava transformation

A cyanide-insensitive mitochondrial alternative oxidase of Arabidopsis thaliana

(AtAOX1A; At3g22370, GenBank Accession #: M96417) was codon-optimized for expression in cassava by a PCR-based method (Sremmer et al. 1995). The entire sequence including the 918 bp of AtAOX1A coding sequence and cloning sites were divided into 23 fragments with 42 bases each except the last one (19 bases). Twenty two forward primers were designed based on the first 22 fragments with codons optimized for cassava. Twenty reverse primers were designed so that they were overlapped with a half of two consecutive forward primers. A primary PCR was done with the mixture of the forward and reverse primers at the same molar ratio. The product was diluted prior to use in a secondary PCR. The secondary PCR was done with the diluted product of the primary PCR as a template and a primer set of AOF1 (5‟-

CGCACCCGGGATATGGACACTAGAGCACCAACCATTGGAGGT-3‟) and AOR1

(5‟-TGCCGAGCTCGAATCAATGATACCCAATTGGAGCTGGAGC-3‟). AOF1

121 primer contains SmaI site for cloning and the start codon, ATG, and AOR1 has SstI site and the stop codon as indicated with the underlined bases. The final product was cloned into pUC19 using SmaI and SstI sites. Sequencing of the insert revealed a single base mutation, and the mutation was fixed using QuikChange® Site-Directed Mutagenesis Kit

(Stratagene). The primers used for the site-directed mutagenesis reaction were

AtAox1SDMF (5‟-

CGTGATGTTGTGATGGTTGTTCGTGCTGACGAGGCTCATCACC-3‟) and

AtAox1SDMR (5‟-

GGTGATGAGCCTCGTCAGCACGAACAACCATCACAACATCACG-3‟). The corrected codon-optimized AtAOX1A was removed from pUC19 using SmaI and SstI, and placed under control of root-specific patatin promoter in a pBI121-based binary vector to generate 3D-AtAox1A CO (Siritunga, 2002; Ihemere, 2003, Figure 3.1). After subcloning into the binary vector, the insert AtAOX1A as well as the cloning sites in the vector was fully sequenced in both directions for verification.

122

HindIII SmaI SstI EcoRI

Patatin promoter AtAox1A CO Nos terminator pBI121

Figure 3.1. Plasmid map of 3D-AtAox1A CO. The codon-optimized Arabidopsis alternative oxidase, AtAox1A was cloned into pBI121 based-3D vector in which the

CaMV 35S promoter was replaced by the root-specific patatin promoter followed by the

NOS terminator (Siritunga, 2002; Ihemere, 2003).

The plasmid was introduced into Agrobacterium tumefaciens LBA4404 by electroporation based on the protocol from Invitrogen (www.invitrogen.com). To 2 µL of plasmid DNA, 20 µL of thawed competent Agrobacterium LB4404 were added. The cell/DNA mixture was pipetted into a 10 mm cuvette and electroporated using the Biorad

MicroPulserTM with voltage set at 2.2 kV. After the electroporation, 1 mL of room temperature YM medium (Invitrogen, www.invitrogen.com) was added to the electroporated DNA/cell mixture and transferred to a 15 mL culture tube for incubation at

30°C and 225 rpm. A hundred microliters of the transformation was plated on YM plates with 50 mg/L kanamycin and 30 mg/L streptomycin at 28°C for 48 hours. Plasmids were isolated from the transgenic Agrobacteria, and the insert was confirmed by PCR.

123

3.2.6 Regeneration and transformation of Cassava

3.2.7 Cassava transformation using embryo cotyledons

3.2.8 Induction of embryogenesis

Somatic embryogenesis was induced using the method described by Ihemere (2003). Leaf lobes from 3-4 week old in vitro cassava plants, variety TMS 60444, were cultured on

MS medium containing 8mg/L 2.4-dichlorophenoxyacetic acid (2.4-D) for induction of embryogenesis. Embryogenic callus developed within four weeks of incubation at 28°C and reduced light. Embryos were picked and plated on MS with 0.5 mg/L benzyl aminopurine for regenerating plants. Embryo cotyledons developed within four weeks on this medium.

3.2.9 Co-cultivation with Agrobacterium

124

3.2.10 Regeneration of plant tissue

Putatively transgenic embryos were cultured on MS medium containing 1.0 mg/L benzyl amino purine (BAP). Plantlets were recovered within four weeks on this medium.

Regenerated plantlets were subcultured into regular MS medium to allow development into plants.

3.2.11 Cassava transformation using friable embryogenic callus

Subsequently, they were washed with MS liquid medium containing 500 mg/L carbenicillin to remove Agrobacterium. The washed FECs were transferred to solid medium (MS with 500mg/L carbenicillin) and kept on this medium for one week before transfer to selection medium (MS with 500mg/L carbenicillin and 30mg/L paromomycin). During a 3-6 week period, embryos were picked from this medium into regenerating medium (MS with 1 mg/L BAP). Plantlets developing from this medium were cultured on regular MS medium for further propagation and analysis.

125

3.2.12 RT-PCR analysis of transgenic plants

RNA was isolated from 100 mg of cassava roots using the Qiagen RNeasy Plant Mini kit

(Qiagen Inc., Valencia, CA). Concentrations of RNA were measured on a spectrophotometer at 260 nm. Prior to cDNA synthesis, the RNA was treated to remove

DNA contamination using the Promega DNAse treatment (Promega Corporation,

Madison, WI). About 2-10 µg of RNA were used for cDNA synthesis using the qscrpt cDNA kit (Quanta Biosciences; MD).

The cDNA was used to check for the expression of the transgene. The primers used were

X0329F (GGATTAAGGCTCTTCTTGAGGAAGCA) and Nos0329R

(GCCAAATGTTTG AACGATCGG), and they amplified a 500 bp region from the end of the AOX1A gene to the beginning of the NOS terminator (Figure 3.2). Tubulin primers

TubF (TATATGGCC AAGTGCGATCCTCGACA) and TubR

(TTACTCTTCATAATCCTTCTCAAGGG) were used as positive controls for the PCR reaction. The PCR reaction was based on ChoiceTM Taq DNA polymerase from Denville

Scientific Inc. (www.densci.com). The reaction mixture contained ChoiceTM buffer (5 µL of 10X ), 1 µL of 10mM DNTP, 1 µL of ChoiceTM Taq, 0.2 µM of each of the forward and reverse primers, 2-10 µg of template DNA and deionized water to a total volume of

50 µL. The reactions for both the transgene primers (X0329F and Nos0329R) and the tubulin primers (TubF and TubR) were run with a denaturation temperature of 95°C, an annealing temperature of 52°C and an extension temperature of 72°C in the Biorad

126

MyCycler thermal cycler (Bio-rad Laboratories, http://www.bio-rad.com). Thirty-two reaction cycles were used.

Figure 3.2. Region of the construct amplified in RT-PCR analysis of the transgenes. RNA

(100 µg) was extracted from 4-week old in vitro cassava roots growing on regular

Murashige and Skoog medium supplemented with 2% sucrose. Expression of the AOX transgene was verified by amplifying a 300 bp region indicated between the arrows.

127

3.2.13 Plant growth in the greenhouse

Cassava plantlets from tissue culture were transferred to greenhouse conditions for further analysis. Greenhouse plants were grown on the Fafard potting mix, a mixture of peat moss, perlite and vermiculite. To speed up the development of tuberous roots, plants were grown in the 3 Kord Traditional Square Pot (Kord Products: Canada). The pots have a volume of 230 mL and the small size promotes the rapid development of tuberous roots that can be analyzed within 4 months. After 4-6 months, tuberous roots were harvested from these plants and analyzed.

3.2.14 Determination of the alternative oxidase capacity in cassava roots

Mitochondria were isolated from cassava roots using the method described by Millar et al. (2007) with some modifications. All procedures were done at 4°C. Twenty grams of root tissue were washed and cooled to 4°C before being homogenized in a pre-chilled buffer containing 300 mM sucrose, 50 mM Tris-HCl (pH 7.5), 2 mM EDTA, 8 mM cysteine, 2% Polyvinylpyrrolidone (PVP) and 0.1% bovine serum albumin using the

Magic Bullet MB1001 blender (Homeland Houseware LLC) for 5×2 seconds. The homogenizing buffer was added at a ratio of 5 ml per g fresh weight of tissue. The suspension was filtered through 4 layers of cheesecloth and the filtrate was collected in a pre-chilled beaker on ice. The filtered homogenate was centrifuged at 3000 g at 4°C for 5

128 min to remove starch and cell debris. The supernatant was then centrifuged at 15000 g for

15 minutes and the pellet was resuspended gently using a clean soft paint brush in 5 mL of wash buffer containing 10 mM Tris-HCl pH 7.5, 300 mM sucrose, 2 mM EDTA and

0.1% bovine serum albumin. The resuspended solution was adjusted to 40 mL by adding more wash buffer and centrifuged at 1000 g for 5 min. The supernatant was centrifuged at

15000 g for 15 min and the pellet resuspended in 5 ml of wash buffer and kept on ice.

Mitochondrial protein was determined by Bradford method. Since the extraction buffer included BSA, protein determination in the samples was estimated from similar root extractions in a buffer without BSA.

To determine the alternative capacity, oxygen consumption was measured polarographically in the Hansatech oxygen electrode (model Oxygraph) as described by

Bhate and Ramasarma (2010) with some modifications. All assays were conducted in1.0 mL of standard reaction medium (10 mM Tris-HCl pH 7.5, 10 mM MgCl2, 10 mM KCl2,

10 mM KH2PO4) using approximately 0.2 mg of protein. The respiratory capacity of

AOX (alternative capacity) was measured by adding 20 mM of succinate and 1.0 mM

ADP (all concentrations final). Pyruvate (5 mM) and DTT (10 mM) were added to ensure that AOX was activated. Selective inhibition of the cytochrome pathway was achieved by addition of 1.0 mM potassium cyanide while selective inhibition of alternative oxidase was achieved by addition of 2 mM Salicylhydroxamic acid (SHAM). Alternative oxidase capacity was defined as the rate of oxygen consumption that was insensitive to 1.0 mM potassium cyanide and sensitive 2 mM SHAM.

129

3.2.15 Evaluation of postharvest physiological deterioration in cassava roots

Postharvest physiological deterioration was evaluated using the method described by

Morante et al. (2010) with some modifications. Cassava plants were harvested carefully to avoid physical damage. They were stored on plastic laboratory weighing boats at room temperature. Evaluation of postharvest physiological deterioration was done every 7 days up to 35 days. Harvested roots were cut into equal cross sections and scored for vascular discoloration using ImageJ image processing and analysis software

(http://rsb.info.nih.gov/ij/).

3.2.16 Measurement of PPD using UV fluorescence

UV fluorescence has been considered a visual marker of PPD. This is because one of the early events in the sequence of processes occurring after harvest is scopoletin-dependent

UV fluorescence (Wenham, 1995). Sectioned cassava tuberous roots were viewed under

UV light and the appearance of fluorescence was compared with the visual appearance of the root.

3.2.17 Measurement of yield parameters

For monitoring yield parameters, three plants from each line were grown in the 3 Kord

Traditional Square Pots for 4 months in the greenhouse. Greenhouse grown plants were harvested and fresh weight measurements were taken on all the tuberous roots to get an estimate of yield. Average tuber fresh weight per plant per line was calculated. In

130 addition, stem length was also measured, and an average stem length per line was calculated.

3.2.18 Stress treatments

3.2.18.1 Waterlogging stress

For waterlogging stress, cassava plants were submerged in water in sub-irrigation trays

(Figure 3.3) with no draining for 14 days. The water in the tray was replenished daily to prevent build-up of pathogens. After 14 days, the plants were harvested and analyzed as described below.

Figure 3.3. Illustration of submergence experiment setup in non-draining trays.

Submergence was done at 4 and 8 week stage of growth in the greenhouse.

131

3.2.18.2 Salt stress

The maximum salt concentration used in these experiments was 8.0 g/L (138 mM), determined previously as the concentration at which varietal differences in survival, growth, leaf water content and accumulation of Na+ and Cl- were observed (Carretero et al. 2007). To avoid osmotic shock, salt stress was introduced gradually over five days by adding 50 mL of an 8.0 g/L (138 mM) solution of sodium chloride every day for five days in draining pots. After 7 days, the plants were harvested and analyzed as described below.

3.2.19 Statistical analysis

Standard deviation, F-tests for differences between sample variances, T-tests for significant differences and one-way analysis of variance with Dunnett‟s multiple comparison tests were carried out using Microsoft Excel 2010 and GraphPad Prism software packages. All analyses for significant differences were determined at P≤0.05.

132

3.3 Results

3.3.1 The oxidative burst in damaged cassava roots is cyanogen-induced

Mechanical injury in cassava storage roots is a trigger of both cyanogenesis (McMahon et al. 1995) and production of reactive oxygen species (Reilly et al. 2004). To investigate a causal link between cyanogenesis and the oxidative burst, production of reactive oxygen species was measured in sectioned in vitro roots of transgenic low cyanogen and wild- type plants using two methods. In the first, the roots were incubated in 1.0 mg/mL of 3, 3 diaminobenzidine for six hours, and the brown coloration resulting was used as a measure of hydrogen peroxide. Transgenic low cyanide plants with less than 1% of wild-type linamarin levels (Cab1-1, Cab1-2 and Cab1-3; Siritunga and Sayre, 2003) produced reduced hydrogen peroxide (by between 2 to 8-fold) compared to wild-type plants

(Figure 3.4A; Table 3.1). In the second method, roots were exposed to the fluorescent dye

2‟, 7‟-dichlorofluorescein diacetate (H2 DCF-DA; Molecular Probes) for thirty minutes, and then analyzed on a laser confocal microscope (Figure 3.4B). Transgenic plants with reduced cyanogenesis also showed reduced reactive oxygen species-induced fluorescence compared to wild-type plants by between 2 to 11-fold (Figure 3.4B).

133

Figure 3.4. ROS production in low and high cyanogen plants. ROS accumulation is reduced in low cyanogen (Cab1) cassava lines. Analysis was carried out (A) Using 3, 3 diaminobenzidine (DAB) and viewed under the Olympus DP20 light microscope and (B)

Using H2DCFDA (DCF) and viewed under a Zeiss LSM 510 laser confocal microscope.

134

Table 3.1. Flourescence in wild-type and Cab1 plants as scored by ImageJ software.

Units are arbitrary. (H2DCFDA=2‟, 7‟-dichlorofluorescein diacetate, DAB=3, 3

Diaminobenzidine). All transgenic low cyanogen cassava lines (Cab1-1, Cab1-2 and

Cab1-3) had significantly lower levels of ROS accumulation at p≤0.05. Statistical analysis was carried out by one-way ANOVA with Dunnett‟s Multiple Comparison Test.

Asterisks show significant difference from wild-type (MCol 2215).

MCol2215 Cab1-1 Cab1-2 Cab1-3

H2DCFDA 18.85±1.1 3.87±0.3* 2.24±0.6* 9.6±0.6*

DAB 68.66±3.1 6.1±1.0* 32.08±2.7* 38.12±2.7*

135

These results suggest that reactive oxygen species accumulation in damaged cassava roots is due to cyanide release. To ascertain whether the differences in reactive oxygen species production were specifically due to differences in cyanogen content, we carried out biochemical complementation experiments in which the low cyanogen plants were pre-treated with 5 mM potassium cyanide (the cyanogenic potential of cassava roots is 1-

5mM). As shown in Figure 3.5, low cyanogen plants showed increased ROS production when they were exposed to cyanide. Wild-type ROS accumulation increased two-fold with cyanide, while ROS accumulation in transgenic low cyanogen (Cab1) lines increased between 2 and 16-fold in the presence of cyanide. Thus, exogenous cyanide rescues the low oxidative burst phenotype in low cyanogen lines.

These data show that the oxidative burst occurring in mechanically damaged cassava roots is induced by cyanogenesis, suggesting that a possible solution to cassava postharvest physiological deterioration is to reduce the cyanide-induced accumulation of reactive oxygen species.

136

20 Fluorescence

10 units per area unit units

WT Cab1-1 Cab1-2 Cab1-3 WT + KCN Cab1-1 + KCN Cab1-2+ KCN Cab1-3+ KCN

Figure 3.5. Biochemical complementation of low cyanide plants with 5 µM potassium cyanide (KCN) results in increased ROS production in 4 weeks-old in vitro low cyanogen transgenic plants. In vitro roots were stained with H2DCFDA and analyzed on a laser confocal microscope. Quantification of fluorescence was done using Image J image processing software. The data are averages of four experiments. Error bars show 95% confidence interval.

137

3.3.2 The cyanide-induced oxidative burst in wounded cassava roots is of

mitochondrial origin

We hypothesized that cyanide release, which occurs on mechanical damage in cassava roots (Siritunga and Sayre, 2003; McMahon et al., 1995) was causing the accumulation of reactive oxygen species via cyanide induced inhibition of mitochondrial electron transfer and over-reduction of the upstream electron transfer complexes leading to ROS production (Moller, 2001). To check if other sources of ROS in plants, such as the plasma membrane NADPH oxidase, had a significant contribution to the oxidative burst in wounded cassava roots, wild-type cassava roots were pre-treated with 100 µM diphenyl iodonium chloride (DPI), a non-specific inhibitor of the plasma membrane NADPH oxidase. As shown in Figure 3.6, DPI reduced the accumulation of reactive oxygen species by only 20%, suggesting that a substantive proportion of the observed oxidative burst was from sources other than plasma membrane NADPH oxidase. We hypothesized that this source was the mitochondria, due to the potential of cyanide to inhibit cytochrome oxidase in the mitochondrial electron transport chain, leading to the production of ROS.

138

Figure 3.6. Inhibition of the plasma membrane NADPH oxidase. 100 mM DPI, an inhibitor of the plasma membrane NADPH oxidase, does not substantially reduce ROS production in 4 weeks-old in vitro cassava, suggesting that the ROS may be of mitochondrial origin. Fluorescence intensity was scored from images of 3 experiments using image J (B). The data was analyzed by t-tests using GraphPad Prism software package (version 5). There was no significant difference between treated and untreated roots at p≤0.05. Error bars show confidence interval at p≤0.05.

139

Figure 3.6 Continued.

Fluorescence units per area unit units

Wild-type

Wild-type + DPI

140

3.3.3 Generation of AOX transgenic lines

Taken together, these results support the hypothesis that cyanide release triggers the oxidative burst in cassava roots. To block this accumulation of reactive oxygen species, we expressed the cyanide-insensitive alternative oxidase in cassava roots.

The following hypotheses were proposed

i. Expression of alternative oxidase would provide an escape valve for electrons in

cyanide poisoned mitochondria reducing the over-reduction of complex I and III

leading to reduced ROS production

ii. Reducing ROS accumulation would delay PPD in cassava.

To reduce cyanogen-induced accumulation of reactive oxygen species, cassava lines were genetically transformed using Agrobacterium-mediated transformation, with codon- optimized Arabidopsis alternative oxidase (AOX1A) driven by a patatin promoter (see materials and methods). A total of 19 independent lines were generated and seven

(PAOX1-7) were selected for further analysis. Expression of alternative oxidase in transgenic lines was confirmed by RT-PCR analysis (Figure 3.7). Wild-type lines (WT) were negative for the transgene.

141

Figure 3.7. Alternative oxidase expression in roots of transgenic lines as determined by

RT-PCR. RNA was extracted from four-week old in vitro lines. Primers specific to the end of AOX1A and the beginning of the nos terminator were used. Expression was normalized by α-tubulin primers were used. The expected 500 bp band for the AOX transgene was seen in PAOX1-7 and not in the wild-type.

142

3.3.4 Expression of Arabidopsis alternative oxidase prevents ROS accumulation in

cassava roots

The selected AOX lines were analyzed for ROS accumulation using the fluorescent dye

2‟, 7‟-dichlorofluorescein diacetate (see experimental procedures). In the transgenic lines,

ROS accumulation was reduced to barely detectable levels in PAOX1-4 and by 4-14 times in PAOX5, PAOX6 and PAOX7 (Figure 3.8). Similarly, the hydrogen peroxide- specific 3, 3 –diaminobenzidine (DAB, Figure 3.8A) stain gave brown occlusions in wild-type lines (WT) but not in transgenic AOX lines (Table 3.2). These results were also confirmed by confocal microscopy analysis of plants (Figure 3.8B) which showed reduced fluorescence in transgenic plants overexpressing AOX by at least 18 times

(Table 3.2). To show that the reduction in ROS also occurred in storage roots, DAB analysis was carried out for two selected transgenic lines (PAOX1 and PAOX2). The brown coloration characteristic of hydrogen peroxide production was more visible in the wild-type than transgenic AOX lines Figure 3.8C).

143

Figure 3.8. ROS accumulation in roots of transgenic AOX lines. In A, roots were exposed to 3, 3 Diaminobenzidine (DAB) while in B they were exposed to 2‟, 7‟- dichlorofluorescein diacetate. The DAB analysis was also carried out in harvested greenhouse roots (C) and shows reduced hydrogen peroxide (indicated by brown coloration) in transgenic lines compared to the wild-type (TMS 60444). Continued.

144

Figure 3.8 continued.

145

Table 3.2. ROS-induced fluorescence in wild-type and transgenic AOX plants. The values (± standard deviation) are results for the fluorescence (Figure 3.8B) quantified using ImageJ image processing software. Statistical analysis was done by one-way anova with Dunnett‟s multiple comparisons test. All transgenic lines were significantly different from the wild-type at p≤0.05.

Flourescence units/unit area

WT 18.69±2.1

PAOX1 0.19±0.4

PAOX2 0.06±0.02

PAOX3 0.06±0.01

PAOX4 0.47±0.1

PAOX5 1.61±0.7

PAOX6 1.29±0.3

PAOX7 4.6±1.4

146

3.3.5 Transgenic AOX plants have increased alternative oxidase capacity

To determine whether expression of AOX in cassava roots increased the alternative oxidase capacity, mitochondria were isolated from tuberous roots (harvested and stored at

-20°C). Approximately 0.2 mg crude mitochondrial protein was used to measure respiration rate in an oxygen electrode. To determine the alternative oxidase capacity, the rate of oxygen consumption was measured in the presence of 1.0 mM potassium cyanide

(which inhibits the cytochrome pathway). The alternative oxidase pathway was defined as the rate of oxygen consumption resistant to cyanide and sensitive to 2 mM SHAM.

Figure 3.9 shows the results of the analysis. The alternative capacity for wild-type was

22.6 nmol O2/mg protein/min, while in the transgenic lines it was between 37.5 nmol

O2/mg protein/min and 51.2 nmol O2/mg protein/min (approximately two-fold increase).

These rates are within the range of previous cyanide-insensitive respiratory rates for cassava tuberous roots published by Passam (1976) ranging from 15-57 nmol of O2 per mg protein per min. The rates in transgenic lines are not as high expected based on ROS accumulation data. This might be because tuberous roots used had been stored frozen for some time and may have lost some activity in the process. Additional mitochondrial purification steps, such as density gradient centrifugation with Percoll may increase the activities recorded.

Cassava roots are, on average 33.5% dry matter (Ceballos et al. 2010). Thus, on a dry weight basis, cassava has significantly lower rates of AOX compared to other crops such as soyabean (66 nmol/mg protein/min, Kearns et al. 1992) and tobacco under nutrient

147 stress (60 nmol/mg protein/min, Parsons et al. 1999). This may be due to the fact that the cassava storage root is metabolically less active compared to other plant roots.

40 Consumption

2 20

O nmoles/mg protein/min nmoles/mg 0

WT PAOX1PAOX2PAOX3PAOX4PAOX5PAOX6PAOX7

Figure 3.9. Alternative oxidase activity in roots of wild-type (WT) and transgenic plants overexpressing AOX. The data (in nmol of O2/mg protein/min) are averages of three experiments. Data analysis was by one-way ANOVA with Dunnett‟s Multiple

Comparisons test. Error bars show 95% confidence interval. All transgenic lines were significantly different from the wild-type at p≤0.05.

148

3.3.6 Expression of Arabidopsis alternative oxidase delays post-harvest

physiological deterioration in cassava storage roots

After ascertaining that AOX plants had reduced accumulation of ROS, the plants were grown in the greenhouse for 4-6 months, during which time they developed small storage roots. The storage roots were assayed for PPD (see experimental procedures). Transgenic plants expressing alternative oxidase showed delayed PPD by at least 14 days (Figure

3.10). There were no signs of vascular discoloration in transgenic AOX lines after 14 days while the wild-type had vascular streaking signs (Figure 3.10A). At 21 days, PPD was mixed with secondary rotting (Figure 3.10B), but most transgenic lines still showed minimal deterioration. Most transgenic lines showed no signs of PPD beyond two weeks after harvest, with PPD beginning to show only after 21 days in some lines. One of the lines (PAOX2) showed no signs of deterioration after 28 days but exhibited signs of secondary rot after 35 days (Figure 3.11). These results show that reducing reactive oxygen species accumulation delays postharvest physiological deterioration in cassava.

149

Figure 3.10. Delayed PPD in transgenic plants expressing alternative oxidase 14 days after harvest (A) and 21 days (B). Wild-type (TMS 60444) lines had symptoms of vascular discoloration after 14 days at room temperature while AOX transgenic lines had none (A). The level of deterioration (discoloration) was scored after 21 days using

ImageJ (C). Roots with a PPD score below 40 were considered suitable for consumption and marketing. The roots were obtained from 4 months-old greenhouse-grown cassava plants..

150

Figure 3.10.

151

Table 3.3. Delayed PPD in transgenic AOX lines after 21 days. PPD scores were obtained using image J image processing software based on the intensity of vascular discoloration. The values are ±standard deviation. Statistical analysis was done by one- way ANOVA with Dunnett‟s Multiple Comparison Test. All transgenic lines were significantly different from the wild type at p≤0.05.

Cassava line PPD score

WT 76.2±8.7

PAOX1 27.3±6.3

PAOX2 29.6±2.9

PAOX3 34±7

PAOX4 21.7±3.1

PAOX5 59.1±21.7

PAOX6 7±4.4

152

100

50 PPD PPD score

WT 28 days WT 35 days PAOX2 28 days PAOX2 35 days

Figure 3.11. Prolonged shelf-life in PAOX2. PAOX2 showed no PPD 4 weeks after harvest. After 5 weeks, secondary rot but not vascular streaking was observed. The roots were obtained from 4 months-old greenhouse-grown cassava plants.

153

3.3.7 Analysis of PPD using UV imaging

Scopoletin-induced UV fluorescence is one of the early indicators of PPD (Wenham,

1995). Within 1-2 days after harvest, scopoletin concentration increases 150-200 fold

(Blagbrough et. al. 2010). After 6 days, however, scopoletin concentration is reduced and becomes unreliable as a predictor of PPD (Salcedo et al. 2010; Blagbrough et. al.

2010). We compared UV fluorescence with the visual characteristics of PPD 14 days after harvest. In particular, we wanted to find out the fluorescence status of transgenic roots showing delayed PPD. While analysis of UV did show distinct differences between plants based on their level of PPD (Figure 3.12), scopoletin fluorescence itself could not be used as a measure of PPD. Similar questions regarding the efficacy of UV fluorescence in evaluating PPD have recently discussed by Salcedo et al. (2010) and

Blagbrough et. al. (2010). Roots that had not yet showed signs of vascular discoloration showed high UV fluorescence while roots that had started undergoing discoloration had reduced fluorescence. Scopoletin fluorescence disappeared once discoloration had begun.

It has been previously proposed that the blue-black vascular streaking characteristic of

PPD is due to peroxidase-mediated oxidation of scopoletin (Wheatley and Schwabe,

1985). The fact that scopoletin fluorescence was reduced only when there was vascular streaking supports this hypothesis.

154

Figure 3.12. Relationship between scopoletin fluorescence and PPD. Wild-type roots, which show vascular discoloration (left) have reduced UV fluorescence (Right).

However, reduction in fluorescence does not necessarily indicate that vascular discoloration has begun as show by PAOX2, which has both reduced fluorescence and no visual signs of PPD.

155

3.3.8 Yield parameters in AOX lines grown under greenhouse conditions

Agronomic parameters were measured to determine whether AOX expression had any effect on the yield of cassava plants. Stem length in plants grown in greenhouse pots for 4 months was measured. Stem length in cassava is an important agronomic parameter since, along with the number of internodes, it determines the number of propagules that can be obtained from the plant. There was no significant difference between wild-type plants and transgenic lines in terms of stem length except in PAOX5 and 6 (Table 3.4).

However, all transgenic lines except PAOX1 had as much as a 3-fold increase in tuber fresh weight suggesting that AOX overexpression may provide cassava plants with a substantial yield advantage in the economically important part of the plant.

156

Table 3.4. Stem length and fresh weight in transgenic AOX plants. Stem length in meters and root tuber fresh weight in grams were measured to determine the effect of AOX overexpression on yield parameters in cassava. Statistical analysis was done by one-way

ANOVA with Dunnett‟s Multiple Comparison Test. Asterisks indicate significant difference from the wild-type (WT).

Stem Length in m Root tuber fresh weight g per plant WT 0.89±0.03 17.3±7

PAOX1 0.84±0.07 22±6

PAOX2 0.83±0.06 37.8±7*

PAOX3 0.80±0.08 33.2±10*

PAOX4 0.91±0.07 43.5±10*

PAOX5 0.76±0.06* 35±7*

PAOX6 0.67±0.08* 51.2±6*

PAOX7 0.74±0.09* 31.7±4.9*

157

3.3.9 Overexpression of AOX enhances tolerance to waterlogging and salinity

stress

Cassava is particularly susceptible to waterlogging and succumbs to excess water in the soil (Lebot, 2009). Poor drainage may cause storage roots to rot (Leihner, 2002). Since cassava growing areas are often prone to seasonal flooding (Henry and Hershey, 2002), tolerance to transient periods of waterlogging stress is essential to minimize yield losses under these conditions.

Oxygen deprivation and subsequent re-oxygenation, which occur during transient flooding, cause an increase in oxidative damage (Blokhina et al. 2003; Garnczarska and

Bednarski, 2004). To test whether AOX overexpression enhances tolerance to waterlogging, 8 week-old cassava plants were exposed to 14 days of submergence treatment. Plants grown in the Number 3 Kord Traditional Square Pot (Kord Products:

Canada) start producing storage roots between 5 and 8 weeks after transfer from tissue culture to the greenhouse. There were no differences between wild-type and transgenic plants in the time of onset of tuber formation. Thus, in this experiment, waterlogging stress coincided with tuberization. Where storage roots formed, prolonged exposure to waterlogging resulted in partial tuber rot in both wild-type and transgenic lines.

In transgenic lines, 2-5 fully developed storage roots were recovered after the 14-day submergence treatment (Figure 3.13; Table 3.5). The wild-type lines had reduced tuber growth, with on average, one tuber per plant. In wild-type plants, therefore, the effect of waterlogging was to suppress the development of storage roots. These preliminary

158 results suggest that overexpression of AOX may protect cassava plants against short-term waterlogging, such as transient heavy rains.

159

Figure 3.13. Tuber development in 8-week old cassava plants after 14 days of submergence treatment. Wild type (TMS 60444) roots were mostly fibrous. While roots developed well in transgenic lines, there were signs of rot due to prolonged submergence.

160

Table 3.5. Number of storage roots per plants in 8-week old plants exposed to submergence treatment. 10 plants were analyzed for each line. The data are ±standard deviation. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple

Comparison Test. All transgenic lines were significantly different from the wild-type at p

≤ 0.05.

Line Average number of storage roots formed per plant

TMS 60444 1±0.5

PAOX1 3±0.83

PAOX3 5±0.71

PAOX7 2±0.43

PAOX4 2±0.43

PAOX5 3±0.71

PAOX6 3±0.43

Salt stress has also been associated with increased ROS production in addition to ion homeostasis and salt-induced injury (Apel and Hirt 2004; Smith et al. 2009). Plants exposed to salt stress display complex molecular responses including the production of stress proteins and compatible osmolytes, and increases in enzymes that modulate reactive oxygen species, such as superoxide dismutase, ascorbate peroxidase and

161 glutathione reductase (Zhu, 2001; Smith, 2009). An increase in the AOX capacity has been demonstrated during salt stress (Jolivet et al. 1990; Hilal et al.1998; Ferreira et al.

2008; Smith et al. 2009), indicating the involvement of the alternative oxidase pathway in salt stress response. To determine tolerance of transgenic plants overexpressing AOX to salt stress, 4 week old cassava plants were exposed to salt stress for 5 days by adding 138 mM of sodium chloride gradually in draining pots. Salt stress generally caused wilting and drying of leaves within 7 days. In transgenic plants overexpressing AOX, wilting progressed slower than in the wild-type (Figure 3.14).

In 4-week old plants, root fresh weight showed no significant difference between wild- type and transgenic lines, except in PAOX4 and PAOX7, which had fresh weights significantly higher than wild-type plants (Table 3.6).

When plants were exposed to salt stress (50 mL of 138 mM NaCl daily for 5 days followed by 7 days of normal watering) at 10 weeks after transplanting to greenhouse, differences in root fresh weight between wild-type and transgenic plants were significant

(Table 3.6). These results suggest that transgenic AOX cassava plants may provide tolerance to salt stress at later stages in their development.

162

Figure 3.14. Wilting in 4-week old transgenic lines expressing AOX is delayed compared to wild-type (TMS 60444) plants after 5 days of salt treatment. TMS 60444 had near- complete wilting of leaves while PAOX lines only partially wilted.

163

Table 3.6. Effect of salt stress in wild-type and transgenic cassava plants. Fresh weight and height measurements were taken for 4 week and 10 week old plants grown in the greenhouse. Statistical analysis was done by one-way ANOVA with Dunnett‟s Multiple

Comparison Test. Asterisks denote significant difference at P≤0.05 from wild-type.

Fresh weight in g for Height of 10-week Root fresh 4-week old plants old plants after weight of 10- after salt stress salt stress week old plants g WT 2.23±0.31 24.55±1.47 13.74±2.3

PAOX1 1.65±0.35 30.9±1.5* 15.83±1.6

PAOX2 2.08±0.56 33.02±1.27* 17.63±1.6*

PAOX3 2.05±0.46 30.48±0* 20.8±0.3*

PAOX4 3.15±0.07* 33.15±1* 21.7±0.3*

PAOX5 2.27±0.51 30.48±2* 23.6±0.4*

PAOX7 3.03±0.21* - -

164

In this work, we report increased performance of AOX transgenic plants both under normal greenhouse conditions and under stress. Stress tolerance has been recently correlated with increased alternative oxidase capacity. For example, Smith et al. (2009) found 30-40% improved growth rates in Arabidopsis plants overexpressing AOX under salinity stress. Fiorani et al. (2005) saw a strong correlation between AOX levels and shoot growth of Arabidopsis plants under low temperatures. Plants lacking AOX showed reduced leaf area, whereas plants over-expressing AOX showed increased leaf area in comparison to the wild-type (Fiorani et al. 2005). There were no differences at normal growth temperature. When tobacco suspension cells lacking AOX were grown under macronutrient deﬁciency, they accumulated signiﬁcantly more biomass than wild-type cells, which induced large amounts of AOX protein under these conditions (Sieger et al.

2005).

Overexpression of AOX may be an important strategy for delaying PPD and improving abiotic stress tolerance in cassava. There appears to be a correlation between alternative oxidase capacity, reduction of ROS and abiotic stress tolerance.

165

3.4 Discussion

Postharvest physiological deterioration is a major problem in cassava farming communities. Estimates of economic losses due to PPD range from 5-25% (Wenham,

1995). Several control strategies have been employed to control PPD in cassava. Farmers can harvest piecemeal, thereby minimizing storage constraints. However, keeping the crop in the soil for too long can affect quality. Additionally, in a semi-commercial setting, the land may need to be released for other uses. A more effective control strategy is oxygen exclusion, such as waxing the roots. This is generally not practical in smallholder settings due to its high costs. A convenient control strategy for all farmers would be cultivars that have a longer shelf-life. Sanchez et al. (2005) showed that cassava cultivars with yellow roots (higher carotene content) have a delayed onset of PPD by 1 to

2 days. In addition, Morante et al. (2010) surveyed different sources of germplasm and found delayed PPD (by up to 40 days) in three genotypes high in carotenoid content.

However, cassava farming is also characterized by strong farmer preferences, such that naturally tolerant cultivars may not meet farmer preferences. Genetic transformation offers the possibility of transferring the trait to any farmer-preferred variety.

Postharvest physiological deterioration in cassava has been shown to be associated with an oxidative burst (Reilly et al., 2003). We show that this oxidative burst is due to cyanide released when cassava is mechanically damaged. Cassava produces potentially toxic levels of cyanogenic glycosides which break down to release cyanide (Miller and

Conn, 1980; McMahon et al. 1995; Siritunga and Sayre, 2003; Siritunga et al. 2004).

166

During cyanogenesis, the sugar moiety is cleaved from linamarin by the enzyme linamarase. The resulting cyanohydrin is unstable and degrades spontaneously at pH >

5.0 or temperatures > 35°C, or enzymatically by hydroxynitrile lyase (HNL), releasing hydrogen cyanide and acetone (Cutler and Conn, 1981; White et al., 1994; Hughes et al., 1994; White and Sayre, 1995; White and Sayre, 1998). However, cyanide release generally does not happen in intact cells because linamarin is localized in the vacuole while the deglycosylase linamarase is localized in the cell wall and in laticifers (Mkpong et al., 1990; Hughes et al., 1994; McMahon et al. 1995). Harvesting operations generally cause tissue disruption, which allows linamarin and linamarase to come into contact, initiating cyanogenesis. Cyanide is a potent cellular toxin which inhibits the mitochondrial electron transport chain, by attaching to the iron within cytochrome c oxidase, potentially leading to the accumulation of reactive oxygen species (Yip and

Yang, 1988; Boveris and Cadenas, 1982). The availability of transgenic low cyanogen plants (Siritunga and Sayre, 2003), allowed us to investigate a causal link between cyanogenesis and the oxidative burst associated with PPD. Our results show that this oxidative burst is initiated by cyanide release occurring at mechanical wounding.

Reduced accumulation of ROS in low cyanogen plants was complemented by addition of potassium cyanide in concentrations closely matching the cyanogenic potential of cassava.

It has been shown that inhibition of electron transport results in increased ROS formation

(Boveris and Cadenas, 1982). Thus, the cyanide released during cyanogenesis can cause accumulation of reactive oxygen species via inhibition of cytochrome oxidase in the

167 respiratory electron transport chain. The Ki for cyanide inhibition of cytochrome c oxidase is 10-20 µM, well below the cyanogenic potential of roots (1-5 mM) (Yip and

Yang, 1988; Hell and Wirtz, 2008).

In plants, reactive oxygen species can be produced from various sources including the plasma membrane NADPH oxidase and mitochondria (Moller, 2001; Sagi and Flurh,

2001). Experiments with diphenyl iodonium chloride, an inhibitor of the plasma membrane NADPH oxidase showed no substantial reduction in ROS production. This, together with the biochemical complementation experiments with potassium cyanide, support the hypothesis that ROS accumulation in damaged cassava roots is due to the cyanide inhibition of mitochondrial electron transport chain (ETC). Mitochondria produce ROS at complexes I and III of the ETC (Bhattacharjee, 2011). It is estimated that up to 5% of oxygen consumed by isolated mitochondria results in the formation of

ROS (Millar and Leaver 2000).

Unlike in animal mitochondria, the reduction of oxygen in plant mitochondria can occur by two different mechanisms (Vanlerberghe and McIntosh, 1997; Apel and Hirt, 2004).

In addition to cytochrome oxidase, plants possess an alternative oxidase (AOX), which catalyzes the tetravalent reduction of oxygen to water and branches from the main respiratory chain at ubiquinone (Vanlerberghe and McIntosh, 1997; Maxwell et al., 1999;

Apel and Hirt, 2004). The alternative pathway is cyanide-resistant, can function when the cytochrome pathway is impaired, and has been shown to play a role in lowering ROS formation in plant mitochondria by maintaining upstream electron-transporting components in a more oxidized state (Purvis, 1997; Popov et al., 1997; Maxwell et al.,

168

1999; Finnegan et al., 2004). Alternative oxidase has a considerably lower affinity for oxygen (km >1 µM) compared to cytochrome oxidase (km < 1µM) (Medenstev et al.

2001). We hypothesized that over-expression of AOX could reduce accumulation of cyanide-induced ROS. Alternative oxidase provides an alternative route for electrons passing through the respiratory electron transport chain, which is not associated with a proton motive force, and thus reduces ATP generation. It is encoded by a small family of nuclear genes; AOX1, known for response to stress, and AOX2 which is found in dicots

(Juszczuk and Rychter, 2003). We expressed Arabidopsis AOX1A in cassava roots to reduce cyanide-induced ROS accumulation. AOX function in this respect has been demonstrated in transgenic cultured tobacco cells with altered levels of AOX (Maxwell et al., 1999). Antisense suppression of AOX in tobacco (Nicotiana tabacum) resulted in cells with a significantly higher levels of ROS compared to wild-type cells, whereas the overexpression of AOX resulted in cells with 4-fold reduction in ROS abundance

(Maxwell et al., 1999). In our results, AOX expression using a patatin promoter resulted in at least an 18-fold reduction in ROS abundance in cyanide inhibited mitochondria.

Reduction of ROS has also been reported when AOX is activated by pyruvate (Moller et al. 2010). These data contradict the recent publication by Bhate and Ramarma (2010), which states that H2O2, rather than H2O is the product of AOX. Overexpression of AOX would increase, rather than decrease ROS if the product was H2O2.

Since postharvest physiological deterioration has been shown to be triggered by ROS accumulation, reducing accumulation of ROS in mechanically damaged cassava roots was hypothesized to delay postharvest physiological deterioration. Evaluation of root

169 discoloration using Image J image processing software in transgenic plants expressing alternative oxidase showed a delay in the onset of PPD by at least two weeks. This window gives cassava producers enough time for transport and processing operations required after harvesting the crop. This strategy is therefore anticipated to improve the transition of cassava production from subsistence to commercial. One transgenic line

(PAOX2) did not show any signs of PPD 4 weeks after harvest, suggesting potential for extended storage.

These results suggest a model of PPD in cassava based on reactive oxygen species production (Figure 3.15). In this model, cyanide released on tissue damage causes an oxidative burst, which triggers postharvest physiological deterioration. Expression of

AOX can reduce accumulation of reactive oxygen species, resulting in delayed PPD. It is feasible, from this model, that PPD may also be controlled by increased activities (via transgenic expression) of enzymes that scavenge reactive oxygen species, such as catalase, superoxide dismutase and peroxidase.

170

Figure 3.15. The mechanism and control of postharvest physiological deterioration in cassava. Cyanide released from mechanically damaged roots triggers postharvest physiological deterioration (PPD) by increasing production of reactive oxygen species

(ROS) (blue arrows). Expression of alternative oxidase (AOX) can reduce ROS accumulation since it is cyanide-resistant. This reduction of ROS accumulation results in delayed PPD (dashed red arrow). It is feasible that ROS accumulation can also be prevented by overexpressing enzymes that scavenge ROS, such as catalase, peroxidase and superoxide dismutase (SOD) (dashed black arrow).

171

The oxidative stress model for PPD has also been supported by the discovery of high beta carotene varieties of cassava that have a longer shelf life than low beta-carotene lines

(Sanchez et al. 2006; Morante et al. 2010). Beta carotene is known to have antioxidant properties and is able to quench reactive oxygen species (Smirnoff, 2005).

It was interesting to note that AOX lines with no signs of vascular streaking showed high

UV fluorescence associated with scopoletin release 14 days after harvest (Figure 3.12).

This fluorescence was virtually non-existent in lines at an advanced stage of discoloration. This suggests that scopoletin is involved in the development of vascular discoloration, probably through interaction with hydrogen peroxide (Figure 3.16). It has been proposed that the blue/black vascular streaking observed as PPD may be due to peroxidase-mediated oxidation of scopoletin (Wheatley and Schwabe, 1985). Scopoletin interaction with horseradish peroxidase forms the basis of the so-called scopoletin method of measuring hydrogen peroxide (Sirois and Miller, 1972; Miller et al., 1975).

Scopoletin modifies horseradish peroxidase rapidly to give a stable, spectrophotometrically distinguishable form of the enzyme (Sirois and Miller, 1972).

Inhibition of peroxidase reduces the quenching of peroxide. This would explain why scopoletin ceases to be a reliable indicator of PPD after 6 days.

172

Figure 3.16. Possible interaction of scopoletin and ROS (hydrogen peroxide) in the events leading to vascular discoloration. Scopoletin may react with ROS generated by cyanide inhibition of cytochrome oxidase, producing the blue-black discoloration characteristic of PPD.

173

Transgenic expression of alternative oxidase raises concerns about possible effects on yield because the alternative pathway is associated with a reduced transmembrane potential and potential reductions in ATP synthesis. In addition, delayed PPD in cassava plants has previously been associated with reduced dry matter content (Wenham, 1995).

Our results show an increase in yield of cassava storage roots under normal greenhouse conditions in transgenic AOX lines as well as under stress. The positive yield responses of AOX transgenic lines may be related to the role of AOX in uncoupling carbon metabolism from ATP generation (Sieger et al. 2005; Vanlerberghe et al. 2009; Smith et al. 2009). Normally, electron transport to O2 is coupled to ATP synthesis by oxidative phosphorylation (Sieger et al. 2005). Since the AOX pathway is not associated with energy conservation, it may provide metabolic flexibility, broadening the conditions under which respiration can occur effectively, such as under stress (Vanlerberghe et al.

2009). Thus, the partitioning of electrons to AOX may represent a dynamic system responding to the availability of carbon and reducing power in the mitochondrion

(Vanlerberghe and McIntosh, 1997, Vanlerberghe et al. 2009). These roles may be related to how alternative oxidase is regulated. AOX activity is stimulated by α-keto acids such as pyruvate, which allow it to be active at low concentrations of quinone (Umbach et al.

1994; Day and Wiskich, 1995). AOX must be reduced in order to be activated by pyruvate since the oxidized dimer is unresponsive to pyruvate (Day and Wiskich, 1995).

Thus, AOX activity depends on allosteric regulation by α-keto acids such as pyruvate as well as by its redox state (Sluse and Jarmuszkiewicz, 1998). The redox status may be 174 important in linking AOX activity to the general redox status of the cell and may be related to its role in preventing overreduction of ETC components. Since pyruvate levels can integrate several aspects of respiratory status (for example, it is produced from glycolysis and used in the citric acid cycle) AOX regulation by pyruvate may be well suited to couple respiratory carbon metabolism and electron transport (Vanlerberghe and

McIntosh, 1997). The partitioning of electrons to AOX, therefore, is a dynamic system which responds to the availability of carbon and reducing power in the mitochondrion

(Vanlerberghe and McIntosh, 1997, Vanlerberghe et al. 2009). AOX expression could provide a means by which respiration is adjusted to the needs of other cellular activities

(Vanlerberghe et al. 2009). Hansen et al. (2002) has shown that the plant growth rate is proportional to the rate of respiration and the efficiency by which this process is coupled to phosphorylation. By modulating energy yield (ATP production and NADH consumption), AOX may also play a role in optimizing the rate of highly energy- consuming processes such as growth, matching growth rate with the availability of key resources such as mineral nutrients (Vanlerberghe et al. 2009).

The transgenic AOX plants generated here may improve commercialization of cassava by reducing postharvest losses and improve cassava production under marginal conditions.

Some of the lines are currently growing in field trials for further analysis.

175

Chapter 4: Conclusions and Perspectives

4.1 Introduction

Cassava is the sixth most important crop in the world after wheat, rice, maize, potato and barley, and is an important staple for more than 800 million people in the world. Once considered a neglected crop, cassava has recently received attention from scientists and donor organizations as an important crop for improving global health and food security.

The Bill and Melinda Gates Foundation-sponsored BioCassava Plus project, tackling various aspects of cassava, has recently successfully completed an initial phase. In addition, the cassava genome has recently been sequenced

(http://www.phytozome.net/cassava.php), providing a wealth of information for researchers to use in crop improvement programs.

Transgenic crop improvement allows the incorporation of useful traits into farmer- preferred cultivars. For cassava, these traits include a reduction in cyanogenesis, increase in protein and other nutrients, resistance to viruses, tolerance to biotic and abiotic stresses and increasing of shelf-life. Especially, increasing cassava shelf-life is a challenge hindering full commercialization of cassava. This dissertation has covered cassava cyanogenesis; it‟s relation to nitrogen metabolism, postharvest physiological

176 deterioration and abiotic stress. Our data support a model for postharvest physiological deterioration based on cyanide-induced oxidative stress. Based on this, we used a control strategy for PPD based on reducing cyanide-induced ROS by over-expression of the cyanide-insensitive alternative oxidase (AOX). These transgenic plants, in addition to longer shelf-life, consistently out-yield wild-type lines and are tolerant to abiotic stress in preliminary studies.

4.2 The role of cyanogens in nitrogen metabolism

Cyanogenic glycosides, such as linamarin in cassava, have long been proposed as transportable forms of reduced nitrogen (McMahon et al. 1995; Selmar et al. 1988;

Siritunga et al. 2004). The β-cyanoalanine synthase pathway, present in both cyanogenic and non-cyanogenic plants, has been proposed as the pathway by which the nitrogen in cyanide re-enters primary metabolism. Support for this has come from several sources. β- cyanoalanine synthase activity is higher in cyanogenic species compared to non- cyanogenic species (Miller and Conn, 1980). In addition, nutrient enrichment studies have shown that cyanide is converted into ammonia in response to decreased nitrogen supply (Ebbs et al. 2010).

Previous research in the Sayre lab has led to the generation of transgenic low cyanogen lines (Siritunga and Sayre, 2003). We compared the activity of nitrate reductase, a key enzyme in nitrogen metabolism, in wild-type lines and low cyanogen lines. The results showed that nitrate reductase activity is 3X higher in low cyanogen plants than in wild- 177 type lines, suggesting that cyanogenesis provides a significant source of reduced nitrogen which has to be compensated for by nitrate reduction in low cyanogen plants. In addition, our assays of β-cyanoalanine synthase and rhodanese confirmed that β-cyanoalanine synthase is the key cyanide metabolizing enzyme in cassava roots, with three times more activity in the roots compared to shoots, while rhodanese had no detectable activity in cassava roots. Nitrilase, which is involved in metabolism of cyanoalanine produced by β- cyanoalanine synthase activity, also had higher activity in roots. We concluded from these assays that the β-cyanoalanine synthase pathway is active in cassava roots, and is likely to be the route for assimilation of cyanogens into free amino acids (Figure 4.1).

178

Figure 4.1. Enzymes involved in cyanide metabolism in cassava roots and the relationship between cyanogenesis and nitrogen metabolism. β-cyanoalanine synthase catalyzes the reaction between cyanide and cysteine to produce cyanoalanine. Rhodanese detoxifies cyanide to thiocyanate. Nitrilase 4 catalyzes the conversion of cyanoalanine to free amino acids and ammonia. Relative enzyme rates are compared to activities in leaves.

179

To test the possibility of enhancing cyanide assimilation via β-cyanoalanine synthase, we generated transgenic plants expressing β-cyanoalanine synthase and nitrilase 4 under the control of the patatin promoter. Expression of β-cyanoalanine synthase in cassava roots increased root total amino acid content by up to 30%, an increase which was associated with increased activity of β-cyanoalanine synthase, although it was also associated with poor growth and reduced fresh weight. The negative effects on growth may be related to the relationship between cyanide assimilation through the enzyme β-cyanoalanine synthase and cysteine synthesis, or may suggest a role for β-cyanoalanine synthase and cyanide in regulating root development. Cysteine synthase, which catalyzes formation of cysteine from O-acetyl-L-serine (OAS) and disulfide, also possesses CAS activity

(Hasegawa et al. 1995; Maruyama et al. 1998; Hatzfeld et al. 2000). The formation of cysteine from OAS is the ﬁrst point at which sulfur is incorporated into the carbon skeleton of amino acids in the sulfur assimilatory pathway of higher plants (Warrilow et. al. 2002). As a point of convergence between nitrogen and sulfur assimilatory pathways, cysteine synthase represents an important control point (Warrilow et. al. 2002). Run- away cysteine production in the presence of excess S-supply is prevented as sulfide is an inhibitor as well as a substrate for cysteine synthase (Warrilow and Hawkesford,

2000). In addition, accumulation of sulfide may have a role in repression of gene expression such as seen for the sulfate transporter (Smith et al., 1997). Future studies to test the possible effect of sulfide accumulation on sulfate uptake will need to be carried out, and may be undertaken through sulfur uptake experiments in wild-type and transgenic plants.

180

The second strategy for redirecting cyanogens to free amino acids was the expression of

Arabidopsis nitrilase 4 (NIT4) under the control of the patatin promoter. Expression of

Arabidopsis NIT4 resulted in up to 50% reduction in IAA in roots and leaves. This was surprising, since only three Arabidopsis nitrilase homologues (NIT1, NIT2 and NIT3) are known to be involved in auxin biosynthesis, converting indole-3-acetonitrile (IAN) to indole-3-acetic acid (IAA) in vivo (Bartling et al., 1992; Bartling et al., 1994; Schmidt et al., 1996). NIT4 has been reported not only to have high substrate specificity for cyanoalanine, but also not to recognize IAN as a substrate (Piotrowski et al., 2001;

O‟Reilly and Turner, 2003). In addition, even if NIT4 had IAA biosynthesis rather than cyanoalanine hydrase activity, IAA concentrations would have increased rather than decreased in transgenic plants. In maize, a dual role in auxin homeostasis and cyanoalanine hydrolysis by the NIT1/NIT2 heteromers has been reported (Kriechbaumer et al. 2007). On its own, Zea mays NIT2 (ZmNIT2) hydrolyses IAN to IAA, and Zmnit2 knockout mutants accumulate lower quantities of IAA conjugates in the kernels and roots of young seedlings, indicating that an IAN-dependent pathway contributes substantially to auxin biosynthesis in these tissues (Kriechbaumer et al. 2007). Assembly of multimeric complexes is a known feature of nitrilases, where complexes with 2-18 subunits are formed (O‟Reilly and Turner, 2003). ZmNIT1/ZmNIT2 heteromers are involved in cyanide detoxiﬁcation via β-cyanoalanine turnover (Kriechbaumer et al.

2007). Interestingly, ZmNIT2 bears closer homology to Arabidopsis NIT4 than to the other three Arabidopsis nitrilases (Park et al. 2003; Kriechbaumer et al. 2007). Nitrilase heteromers, composed of ZmNIT1 and ZmNIT2 homologues, recognized β-cyanoalanine

181 as a substrate instead of IAN (Jenrich et al., 2007). Expression of NIT4 may have resulted in heteromer formation, resulting in a shift in substrate preference from IAN to cyanoalanine, leading to a reduction in IAA biosynthesis in cassava roots. Shifting substrate recognition from IAN to β-cyanoalanine may be part of a regulatory mechanism linking IAA biosynthesis and the degradation of cyanide (Kriechbaumer et al. 2007).

With the cassava genome complete, characterization of cassava nitrilases may shed some light on the relationship between cyanide metabolism and auxin biosynthesis in cassava.

In conclusion, engineering the β-cyanoalanine synthase pathway by overexpressing CAS and NIT4 is not an effective means of redirecting cyanogens to reduced nitrogen in cassava possibly due to the unanticipated effects on sulfur metabolism and IAA biosynthesis. Direct assimilation of acetone cyanohydrin without release of cyanide, as proposed by Siritunga et al. (2004) and later demonstrated in sorghum by Jenrich et al.

(2007) is worth pursuing in cassava. Conversion of cyanogens to reduced nitrogen without the production of toxic cyanide as an intermediate may be a component of cyanogenic plant cyanide metabolism (Figure 4.2). Meanwhile, CAS and NIT4 plants will be useful as tools in studying cyanide metabolism and its possible relationship with auxin biosynthesis and sulfur metabolism.

With the cassava genome completed, there is the potential now to study the functions of these genes in cassava. For example, a BLAST in the cassava genome using Arabidopsis

NIT4 yields two nitrilase-like genes, one of which has 70% nucleotide identity with Zea mays NIT2.

182

For β-cyanoalanine synthase, a corresponding transcript in cassava

(cassava4.1_011826m.g) was found. As expected, both cassava genes are closely related to sequences in the castor bean, Ricinus communis, a member of the Eurphobiaceae family.

Studying expression levels of these genes in cassava lines of different cyanogenic content, as well as under conditions of cyanide enrichment, may provide more information on their potential role in cyanide metabolism.

183

Figure 4.2. Direct conversion of acetone cyanohydrin into ammonia without intermediate cyanide release may be the mechanism of cyanogen assimilation in cyanogenic plants.

Based on Siritunga et al. 2004; Jenrich et al. 2007.

184

4.3 The potential role for β-cyanoalanine synthase and cyanide in regulating root

development

Expression of β-cyanoalanine synthase in cassava roots resulted in poor root development, reduced growth, and reduced fresh weight. The number of roots in in vitro plants was reduced up to 4 fold in transgenic plants overexpressing β-cyanoalanine synthase while the fresh weight of tuberous roots decreased by up to 65%. Recently,

Garcia et al. (2010) have shown that mitochondrial β-cyanoalanine synthase in

Arabidopsis is essential for maintaining low-levels of cyanide essential for root hair development. β-cyanoalanine synthase mutants accumulated cyanide and were defective in root hair formation (Garcia et al. 2010). β-cyanoalanine synthase may therefore be involved in regulation of root development, possibly by regulating the amount of free cyanide available in the roots. Cyanide enrichment studies with β-cyanoalanine synthase transgenic lines will need to be carried out to test this hypothesis.

4.4 A cyanogen-induced oxidative burst initiates postharvest physiological

deterioration (PPD) in cassava

Cassava has the shortest shelf-life of any tuber crop (Ghosh et al. 1988). Harvested cassava storage roots undergo rapid postharvest physiological deterioration (PPD) within the first 72 hours, which reduces the quality of storage roots for market and consumption

(Booth, 1976; Alves, 2002). PPD is initiated by mechanical damage and is characterized by vascular streaking resulting from occlusions in the vascular parenchyma (Wenham,

1995; Alves, 2002). Studies on PPD in cassava have revealed that it is an active process 185 characterized by oxidative damage (Buschmann et al., 2000; Reilly et al., 2001; Reilly et al., 2004). By analyzing ROS production in low and high cyanogen roots, we have shown that the oxidative burst that occurs in damaged cassava roots correlates with the level of cyanogens in the plants. Biochemical complementation experiments with potassium cyanide on low cyanogen plants confirmed that the oxidative burst is induced by cyanogenesis, which is also initiated in cassava roots by mechanical damage. Based on these results, we propose that cyanogenesis is directly involved in PPD via cytochrome oxidase inhibition and consequent oxidative burst.

4.5 Over-expression of Alternative Oxidase (AOX) extends cassava shelf-life

The finding that ROS production was central to PPD led to the hypothesis that blocking

ROS accumulation would delay PPD, increasing the shelf life of cassava. To block ROS accumulation, we expressed alternative oxidase (AOX) in cassava roots. AOX, which branches from the main respiratory chain at ubiquinone, catalyzes the tetravalent reduction of oxygen to water and has been shown to minimize accumulation of ROS by maintaining upstream electron-transporting components in a more oxidized state

(Vanlerberghe and McIntosh, 1997; Maxwell et al., 1999; Apel and Hirt, 2004).

Expression of AOX in cassava under the control of the patatin promoter effectively blocked ROS accumulation in cassava roots (reducing ROS-induced fluorescence by 10-

14X) and delayed PPD by at least two weeks. For further field trials and analysis, these lines have been shipped to Puerto Rico.

186

In addition to ROS, another key component in PPD is scopoletin (Sirois and Miller, 1972;

Miller et al., 1975; Wheatley and Schwabe, 1985). Scopoletin has been considered as a marker for PPD, but a correlation has not been firmly established (Salcedo et al. 2010).

We found AOX lines with no visible PPD but with high UV fluorescence indicative of scopoletin. This adds support to the hypothesis that vascular streaking may be due to peroxidase-mediated oxidation of scopoletin (Wheatley and Schwabe, 1985) and explains why scopoletin ceases to be a reliable marker of PPD has progressed. Further research into role of scopoletin and hydrogen peroxide in PPD will reveal the picture.

4.6 Over-expression of Alternative Oxidase controls abiotic stress

Reactive oxygen species have emerged as ubiquitous stress markers in various studies of biotic and abiotic stress (Blokhina and Fagerstedt, 2010). Redox homeostasis depends on a balance between ROS generation and scavenging. This balance can be perturbed by abiotic stresses, resulting in oxidative stress (Apel and Hirt, 2004; Desikan et al. 2004).

Alternative oxidase (AOX) has been proposed to play a central role in plant stress response by blocking reactive oxygen species accumulation (Aken et. al. 2009). Three lines of evidence accumulated over the past two decades support this proposal. First, the absence of AOX alters stress defenses so that plants cannot respond appropriately to stress, thereby suffering greater damage (Aken et al. 2009). Second, AOX is up-regulated by a number of stress conditions including chilling injury, drought as well as treatments that disrupt mitochondrial functions, such as rotenone (Complex I inhibitor), antimycin A

(Complex III inhibitor) and cyanide (cytochrome c oxidase inhibitor) (Vanlerberghe et 187 al. 1994; Aken et al. 2009). Third, AOX acts as a buffer against programmed cell death by blocking accumulation of reactive oxygen species when cytochrome c oxidase is inhibited such as during stress or mechanical wounding (Purvis, 1997; Popov et al., 1997;

Maxwell et al., 1999; Aken, 2009). We hypothesized that suppression of ROS accumulation by overexpression of AOX may also provide tolerance to abiotic stress in cassava. The presence of transgenic plants expressing alternative oxidase in the roots allowed us to test this hypothesis.

Cassava is susceptible to waterlogging and poorly drained soils may cause tuber rot

(Leihner, 2002; Lebot, 2009). Also associated with oxygen deprivation during flooding and subsequent re-oxygenation is an increase in oxidative damage (Blokhina et al. 2003;

Garnczarska and Bednarski, 2004). Since cassava growing areas are often prone to seasonal flooding (Henry and Hershey, 2002), tolerance to transient periods of waterlogging stress is essential to minimize yield losses under these conditions.

Transgenic AOX plants had more storage roots per plant and up to twice as much tuber fresh weight after a submergence treatment.

Salt stress has also been associated with increased ROS production in addition to ion homeostasis and salt-induced injury (Apel and Hirt 2004; Smith et al. 2009). An increase in the AOX capacity has been demonstrated during salt stress (Jolivet et al. 1990; Hilal et al.1998; Ferreira et al. 2008; Smith et al. 2009), indicating the involvement of the alternative pathway in salt stress response.

The results of these experiments indicate a promising role for alternative oxidase in transgenic approaches for stress tolerance in cassava. Testing the transgenic plants under

188 field trial conditions will be an important step in their evaluation. In addition, future experiments will need to compare the root-specific expression of AOX reported here with constitutive expression, which might provide tolerance to a broader range of stresses where the stresses organ is the leaf.

189

References

Abhary M, Siritunga D, Stevens G, Taylor NJ, Fauquet CM (2011). Transgenic biofortification of the starchy staple cassava (Manihot esculenta) generates a novel sink for protein. Plos one 6(1): 1-9.

Adamolekun B (2010). Etiology of Konzo, epidemic spastic paraparesis associated with cyanogenic glycosides in cassava: Role of thiamine deficiency? Journal of the neurological sciences; 296(1-2):30-33.

Aken OV, Giraud E, Clifton R (2009). Whelan J. Alternative oxidase: a target and regulator of stress responses. Physiologia Plantarum: 354-361.

Albury MS, Elliott C, Moore AL (2009). Towards a structural elucidation of the alternative oxidase in plants. Physiologia plantarum. 137(4):316-27.

Alves AAC (2002). Cassava Botany and Physiology. In: Hillocks, R.J., Thresh, J.M. and Bellotti, A.C. (eds.) Cassava Biology, Production and Utilization. CAB International, Wallingford. pp 67-89.

Anderssen M, Bush P, Svendsen I and Moller B (2000). Cytochromes P450 from cassava catalyzing the first steps in the biosynthesis of the cyanogenic glycosides linamarin and lotaustralin. J Biol Chem. 275:1966-1975.

190

Apel K, Hirt H (2004). Reactive Oxygen Species : Metabolism , Oxidative Stress, and Signal Transduction. Annu. Rev. Plant Biol. 2004. 55:373–99.

Apostol I, Heinstein PF, Low PS (1989). Rapid stimulation of an oxidative burst during elicidation of cultured plant cells. Role in defense and signal transduction. Plant Physiol. 90:106–16

Bailey Serres J, Chang R (2005). Sensing and signalling in response to oxygen deprivation in plants and other organisms. Ann Bot 96:507 -518.

Bailey-Serres J, Chang R (2005). Sensing and signalling in response to oxygen deprivation in plants and other organisms. Ann. Bot. 96:507–18

Balagopalan C, Padmaja G, Nanada S & Morthy S (1985). Cassava in food, feed and industry. CRC press, Boca Raton, FL.

Balagopalan, C (2002). Cassava utilization in food, feed and industry. In: Hillocks, R.J., Thresh, J.M. and Bellotti, A.C. (eds.) Cassava Biology, Production and Utilization. CAB International, Wallingford.pp 301-318.

Bartel, B. and Fink, G.R. (1994). Differential regulation of an auxin-producing nitrilase gene family in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA, 91, 6649-6653.

Bartling D, Seedorf M, MithoÈ fer A and Weiler EW (1992). Cloning and expression of an Arabidopsis nitrilase which can convert indole-3-acetonitrile to the plant hormone, indole-3- acetic acid. Eur. J. Biochem. 205, 417-424.

Bartling D, Seedorf M, Schmidt RC and Weiler EW (1994). Molecular characterization of two cloned nitrilases from Arabidopsis thaliana: key enzymes in biosynthesis of the plant hormone indole-3-acetic acid. Proc. Natl Acad. Sci. USA, 91, 6021-6025.

Bartoli CG, Gomez F, Gergoff G, Guiam´et JJ, Puntarulo S (2005). Up-regulation of the mitochondrial alternative oxidase pathway enhances photosynthetic electron transport under drought conditions. J Exp Bot 56: 1269–1276. 191

Bediako M, Tapper B & Pritchard G (1981). Metabolism synthetic site and translocation of cyanogenic glucoside in cassava. In: Proceedings of the first triennial root crops symposium of the International society for tropical root crops. eds. Terry, E. R., IDRC Canada, pp 143-148

Beeching, JR, Dodge AD, Moore KG, Phillips HM and JE Wenham (1994). Physiological deterioration in cassava: possibilities for control. Trop Sci 34: 335–343.

Bennett RN and Wallsgrove RM (1994). Secondary metabolites in plant defence mechanisms. New Phytol. 127, 617–633.

Bhate, Radha, and T Ramasarma (2010). Reinstate hydrogen peroxide as the product of alternative oxidase of plant mitochondria. Indian Journal of Biochemistry & Biophysics 47: 306- 310.

Bhattacharjee S (2011). Sites of Generation and Physicochemical Basis of Formation of Reactive Oxygen Species in Plant Cell. In: Reactive Oxygen Species and Antioxidants in Higher Plants. Pp1-30. S.D Gupta Ed. Science Publishers, New Hampshire.

Bhattacharjee S (2005). Reactive oxygen species and oxidative burst: Roles in stress, senescence and signal transduction in plants. Current Science; 89 (7):1113-1121.

Blagbrough IS, Bayoumi SAL, Rowan MG, Beeching JR (2010). Cassava: An appraisal of its phytochemistry and its biotechnological prospects. Phytochemistry. 2010.

Blokhina O and Fagerstedt KV (2010). Oxygen Deprivation, Metabolic Adaptations and Oxidative Stress S. Mancuso and S. Shabala (eds.), Waterlogging Signalling and Tolerance in Plants, Springer Verlag Berlin Heidelberg.

Blokhina O, Virolainen E, Fagerstedt KV (2003). Antioxidants, Oxidative Damage and Oxygen Deprivation Stress: a Review. Annals of Botany; 91(2):179-194.

Blumenthal S, Hendrickson H, Abrol Y, Conn E (1968). Cyanide metabolism in higher plants: III. The biosynthesis of b-cyanoalanine. J Biol Chem 243: 5302–5307.

192

Booth RH (1976). Storage of fresh cassava (Manihot esculenta)1:post-harvest deterioration and its control. Exper. Agric., 12: 103- 111.

Brosche M, Overmyer K, Wrzackek M, Kangasjarvi J (2010). Stress Signalling III: Reactive oxygen species (ROS). In Pareek A, Sopory SK, Bohnert HJ, Govindjee (eds). Abiotic Stress Adaptation in Plants: Physiological, Molecular and Genomic Foundation, pp. 91-102.

Boveris A and Cadenas E (1982) in Superoxide Dismutase, ed. Oberley, L. W. (CRC Press, Boca Raton, FL), pp. 15–30.

Bowles DJ (1990). Defence-related proteins in higher plants. Ann. R. Biochem., 59: 873907.

Bradbury JH (1988). The chemical composition of tropical root crops. Asian Food J. 4:3-13

Buschmann H, Reilly K, Rodriguez MX, Tohme J, Beeching JR (2000). Hydrogen Peroxide and Flavan-3-ols in Storage Roots of Cassava ( Manihot esculenta Crantz ) during Postharvest Deterioration. J. Agric. Food Chem. 48: 5522−5529

Calatayud PA and Le Ru B (1996). Study of the nutritional relationships between cassava and mealybug and its host plant. Bull Soc. Zool. Fr. Evol. Zool. 121: 391-398.

Campbell WH (1999). Nitrate Reductase Structure , Function And Regulation : Bridging the Gap between Biochemistry and Physiology. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:277–303

Carretero CL, Cantos M, García JL, Troncoso A (2007). In vitro–ex vitro salt (NaCl) tolerance of cassava (Manihot esculenta Crantz) plants. In Vitro Cellular & Developmental Biology - Plant; 43(4):364-369.

Chaudhuri M, Hill GC (1996). Cloning, sequencing and functional activity of the Trypanosoma brucei brucei alternative oxidase. Mol Biochem Parasitol 83: 125–129

Chen Q, Zhang B, Hicks LM, Zhang Q, Jez JM (2009). A liquid chromatography-tandem mass spectrometry-based assay for indole-2-acetic acid-amido synthetases. Anal. Biochem.390, 149- 154.

193

Chew, M.Y. (1973) Rhodanese in higher plants. Phytochemistry, 12, 2365–7.

Chong J, Baltz R, Fritig B, Saindrenan P (1999). An early salicylic acid-, pathogen- and elicitor- inducible tobacco glucosyltransferase: role in compartmentalization of phenolics and H2O2 metabolism. FEBS letters; 458(2):204-8.

Clifton R, Lister R, Parker KL, Sappl PG, Elhafez D, Millar AH, Day DA, Whelan J (2005). Stress-induced co-expression of alternative respiratory chain components in Arabidopsis thaliana. Plant Mol Biol 58: 193–212.

Considine MJ, Holtzapffel RC, Day DA, Whelan J, and Millar AH (2002). Molecular distinction between alternative oxidase from monocots and dicots. Plant Physiol. 129: 949–953.

Cruz JL, Mosquim PR, Pelacani CR, Araújo WL, Damatta FM (2004). Effects of nitrate nutrition on nitrogen metabolism in cassava. Biologia Plantarum. 48(1):67-72.

Cutler A and Conn E (1981). The biosynthesis of cyanogenic glycosides in Linum usitatissumum (Linen flax) in vitro. Arch Biochem Biophys 212: 468–474.

Davidian, Jean-Claude, and Stanislav Kopriva (2010). Regulation of Sulfate Uptake and Assimilation--the Same or Not the Same? Molecular plant 3, no. 2: 314-25.

Day DA, Wiskich JT (1995). Regulation of Alternative Oxidase Activity in Higher Plants. Journal Of Bioenergetics. 27(4):379-385.

Dai D, Hu Z, Pu G, Li H, Wang C (2006). Energy efficiency and potentials of cassava fuel ethanol in Guangxi region of China. Energy Conversion and Management 47:1686-1699.

Darwin CR (1839). The Voyage of the Beagle. London: Henry Colburn. pp. 619.

Decoteau DR (2005). Principles of Plant Science Environmental Factors and Technology in Growing Plants. Prentice Hall.

194

Desikan R, Hancock JT, Neill SJ (2004). Oxidative stress signalling. In: Topics in Current Genetics, Vol. 4 H. Hirt, K. Shinozaki (Eds.) Plant Responses To Abiotic Stress Springer-Verlag Berlin Heidelberg.

Dohmoto M, Sano J, Tsunoda H and Yamaguchi K (1999). Structural analysis of the TNIT4 genes encoding nitrilase-like protein from tobacco. DNA Res. 6, 313-317.

Dohmoto M, Tsunoda H, Isaji G., Chiba R and Yamaguchi K (2000). Genes encoding nitrilase- like proteins from tobacco. DNA Res. 7, 283-289.

Dong JG, Fernandez-Maculet JC, and Yang SF (1992). Purification and characterization of 1- aminocyclopropane-1-carboxylic acid oxidase from apple fruit. Proc. Natl. Acad.Sci. USA 89: 9789-9793.

Drapcho C M, Nhuan N P and Walker T H (2008). Biofuels engineering process and technology. McGraw Hill. pp385.

Drew MC (1997). Oxygen deficiency and root metabolism: Injury and Acclimation Under Hypoxia and Anoxia. Annual review of plant physiology and plant molecular biology 48:223- 250.

Dufour DL (1989). Effectiveness of cassava detoxiﬁcation techniques used by indigenous peoples in northwest Amazonia. Interciencia 14:88-91

Dufour DL (2007). „Bitter‟ cassava: toxicity and detoxiﬁcation. In: Ortiz, R. and Nassar, N.M.A. (eds) Proceedings of the First International Meeting on Cassava Breeding, Biotechnology and Ecology. Universidade de Brazilia, Brazil, pp. 171–184.

Elias M, Sudhakaran PR And Nambisan B (1997). Purification And Characterization Of β- Cyanoalanine Synthase From Cassava tissues. Phytochemistry, 46 (3): 469 472.

Ellsbury MM, Pederson GA, Fairbrother TE (1992). Resistance to foliar feeding hipergine weevils (Coleoptera, Curculionidae) in cyanogenic white clover. Journal of Economic Entomology 85, 2467–2472. 195

Essers AJA, Bosveld M, van der Grift RM and Voragen AGJ (1993). Studies on the quantification of specific cyanogens in cassava products and introduction of a new chromogen. Journal of the Science of Food and Agriculture 63, 287–296.

Essers AJA, Ebong C, van der Grift RM, Nout MJR, Otim-Nape W, Osling R (1995). Reducing cassava toxicity by heap-fermentation in Uganda. Intl J Food Sci Nutr 46:125-136.

Ferguson BJ, Beveridge C (2009). Roles for auxin, cytokinin, and strigolactone in regulating shoot branching. Plant physiology. 2009;149(4):1929-44.

Finnegan PM, Soole KL, Umbach AL (2004). Alternative mitochondrial electron transport proteins in higher plants. In DA Day, AH Millar, J Whelan, eds, Plant Mitochondria: From Gene to Function, Vol 17, Advances in Photosynthesis and Respiration. Kluwer, Dordrecht, The Netherlands, pp 163–230.

Fiorani F, Umbach AL, Siedow JN (2005) The alternative oxidase of plant mitochondria is involved in the acclimation of shoot growth at low temperature: a study of Arabidopsis AOX1A transgenic plants. Plant Physiol 139: 1795–1805.

Fregene M and J Puonti-Kaerlas (2002). Cassava Biotechnology. In: Hillocks RJ, Thresh JM and AC Belloti (eds). Cassava: Biotechnology, Production and Utilization. pp179-207.

Garcia I, Castellano JM, Vioque B, Solano R, Gotor C, Romero LC (2010). Mitochondrial β- Cyanoalanine Synthase Is Essential for Root Hair Formation in Arabidopsis thaliana. The Plant Cell Online. 2010:1-13.

Gamborg OL, Miller RA and Ojima K (1968). Nutrient requirements of suspension cultures of soybean root tissue. Expt. Cell Res. 50: 151-158.

Garnczarska, M., and W. Bednarski (2004). Effect of a short-term hypoxia treatment followed by re-aeration on free radicals level and antioxidant enzymes in lupine roots. Plant Physiol. Biochem. 42: 233–240.

196

Ghosh SP, Ramanujam T, Jos JS, Moorthy SN and Nair RG (1988). Tuber Crops. Oxford & IBH, New Dehli, pp. 3–146.

Giraud E, Ho LHM, Clifton R, Carroll A, Estavillo G, Tan Y-F,Howell KA, Ivanova A, Pogson BJ, Millar AH, Whelan J (2008). The absence of alternative oxidase1a in Arabidopsis results in acute sensitivity to combined light and drought stress. Plant Physiology 147: 595-610

Glander KE, Wright PC, Seigler DS, Randrianasolo V, Randrianasolo B (1989). Consumption of cyanogenic bamboo by a newly discovered species of bamboo lemur. Am J Primatol, 19:119- 124.

Gleadow RM, Woodrow IA (2002). Mini-Review Constraints on Effectiveness Of Cyanogenic Glycosides In Herbivore Defense. Journal of Chemical Ecology 28:1301-1313.

Gleadow, RM, Foley WJ and IE Woodrow (1998). Enhanced CO2 alters the relationship between photosynthesis and defense in cyanogenic Eucalyptus cladocalyx F. J. Muell. Plant Cell Environ. 21:12–22.

Hacham Y, Avraham T, Amir R (2002). The N-Terminal Region of Arabidopsis Cystathionine γ-Synthase Plays an Important Regulatory Role in Methionine Metabolism Plant Physiology 128:454-462.

Halkier, B.A. and Gershenzon, J. (2006). Biology and biochemistry of glucosinolates. Annu. Rev. Plant Biol. 57, 303–333.

Han Y, Gomez-Vasquez R, Reilly K, Li H, Tohme J, Cooper RM and Beeching JR (2001). Hydroxyproline-rich glycoproteins expressed during stress responses in cassava. Euphytica 120: 59–70.

Hasegawa R, Maruyama A, Nakaya M, Tsuda S and Esashi Y (1995). The presence of two types of β-cyanoalanine synthase in germinating seeds and their responses to ethylene. Physiol. Plant. 93: 713-718.

197

Hatzfeld Y, Maruyama A, Schmidt A, Noji M, Ishizawa K, and Saito K (2000). β-Cyanoalanine Synthase Is a Mitochondrial Cysteine Synthase-Like Protein in Spinach and Arabidopsis. Plant Physiology 123:1163–1171.

Hatzfeld Y, Saito K (2000). Evidence for the existence of rhodanese (thiosulfate: cyanide sulfurtransferase) in plants: preliminary characterization of two rhodanese cDNAs from Arabidopsis thaliana. FEBS Letters 470:147-150.

Hell R, Wirtz M (2008). Metabolism of cysteine in plants and phototrophic bacteria. In:Rüdiger Hell R, Dahl C, knaff DB, Leustek T. (eds.). Advances in photosynthesis and Respiration 27: Sulfur Metabolism in Phototrophic Organisms, 59–91.

Hemalatha H (2002). Regulation of Nitrate Reductase Activity In Rice (Oryza Sativa L.) By Growth Regulators. Journal of Central European Agriculture, 3 (3); 231-238.

Henry G, Hershey C, Paulo R, Nogueira CP, Road LG (2002). Cassava in South America and the Caribbean. In: Hillocks RJ, Thresh JM and AC Belloti (eds). Cassava: Biotechnology, Production and Utilization. pp 17-40.

Hirose S, Data ES & Quevedo MA (1984). Changes in respiration and ethylene production in cassava roots in relation to post-harvest deterioration In: Uritani & E.D. Reyes. eds. Tropical root crops: post-harvest physiology and processing, pp. 83-89. Tokyo, Japan Scientific Societies Press.

Hiser C, McIntosh L (1990). Alternative oxidase of potato is an integral membrane-protein synthesized de novo during aging of tuber slices. Plant Physiol 93: 312–318.

Hodges DM, DeLong JM, Forney C, Prange RK (1999). Improving the thiobarbituric acid- reactive-substances assay for estimating lipid peroxidation in plant tissues containing anthocyanin and other interfering compounds. Planta 207, 604–611.

Howden AJ, Harrison CJ, Preston GM (2009). A conserved mechanism for nitrile metabolism in bacteria and plants. The Plant journal. 57(2):243-53.

198

Hughes J, Carvahlo F & Hughes M (1994). Purification, characterization and cloning of β- hydroxynitrile lyase from cassava (Manihot esculenta Crantz). Arch of Biochem and Biophys. 311:496-502.

Ihemere U, Siritunga D and Sayre RT (2008). Cassava. In: Compendium of transgenic crop plants 7; Transgenic sugar, tuber and fiber crops. pp 177-198. Kole, C and Hall, TC (Eds). Wiley-Blackwell.

Ihemere, U (2003). Somatic embryogenesis and transformation of cassava for enhanced starch production. PhD Thesis, The Ohio State University.

Goudey JS, Tittle FL and MS Spencer (1989). A Role for Ethylene in the Metabolism of Cyanide by Higher Plants. Plant Physiol. 89: 1306-1310.

Janowitz T, Trompetter I, Piotrowski M. Evolution of nitrilases in glucosinolate-containing plants (2009). Phytochemistry. 2009;70(15-16):1680-6.

Jenrich R, Trompetter I, Bak S, Olsen CE, Møller BL, and Piotrowski M (2007). Evolution of heteromeric nitrilase complexes in Poaceae with new functions in nitrile metabolism. PNAS 104 (47): 18848–18853.

Jolivet Y, Pireaux JC, Dizengremel P (1990) Changes in properties of barley leaf mitochondria isolated from NaCl-treated plants. Plant Physiol 94: 641–646

Jones DA (1998). Why are so many food plants cyanogenic? Phytochemistry 47:155–162.

Juszczuk IM, Rychter AM (2003). Alternative oxidase in higher plants. Acta Biochimica Polonica 50(4):1257-1271.

Kakes, P. and Hakvoort, H. (1992) Is there rhodanese activity in plants? Phytochemistry, 31, 1501–5.

Kartha KK (1974). Regeneration of cassava plants from apical meristems. Plant Science Letters 2, 107–113.

199

Kearns A, Whelan J, Young S, Elthon TE, and Day DA (1992). Tissue-specific expression of the alternative oxidase in soybean and siratro. Plant Physiology 99, no. 2: 712-7.

Kriechbaumer V, Park WJ, Piotrowski M, Robert B. Meeley, Gierl A, Glawischnig E (2007). Maize nitrilases have a dual role in auxin homeostasis and b-cyanoalanine hydrolysis. Journal of Experimental Botany 58(15):4225-4233.

Krischke M, Hoeberichts FA, Ksas B, Gresser G, Havaux M, Van Breusegem F, Mueller MJ (2008). Singlet Oxygen Is the Major Reactive Oxygen Species Involved in Photooxidative Damage to Plants. Plant Physiology: 148: 960-968.

Lai KW, Yau CP, Tse YC, Jiang L, Yip WK (2009). Heterologous expression analyses of rice OsCAS in Arabidopsis and in yeast provide evidence for its roles in cyanide detoxification rather than in cysteine synthesis in vivo. Journal of Experimental Botany 60 (3)993–1008.

Lambers H (1982). Cyanide-resistant respiration–A non-phosphorylating electron-transport pathway acting as an energy overﬂow. Physiol Plant 55: 478–485

Lancaster PN, Ingram JS, Lin HY and Coursey DG (1982). Traditional cassava based foods, survey of processing techniques. Economic Botany 36, 12–25.

Lancaster PA and Coursey DG (1984). Traditional postharvest technology of perishable tropical staples. FAO Agricultural Services Bulletin 59, FAO, Rome.

Lebot V (2009). Tropical Root and Tuber Crops; Cassava, sweet potato, yams and aroids. CABI. pp434.

Leihner D (2002). Agronomy and Cropping Systems. . In: Hillocks RJ, Thresh JM and AC Belloti (eds). Cassava: Biotechnology, Production and Utilization. pp91-113.

Legras JL, Chuzel G, Arnaud A and Galzy P (1990). Natural nitriles and their metabolism. World J. Microbiol. Biotechnol. 6, 83–108.

200

Lennon AM, Neuenschwander UH, Ribas-Carbo M, Giles L, Rylas JA, Siedow JN (1997). The effects of salicylic acid and tobacco mosaic virus infection on the alternative oxidase of tobacco. Plant Physiol 115: 783–791.

Leyva-Guerrero E (2011). Enhancement of the free amino acid and protein content in cassava storage root roots and evaluation of root-specific promoters in cassava. Ph. D. Dissertation. The Ohio State University.

Liang W (2003). Drought stress increases both cyanogenesis and β-cyanoalanine synthase activity in tobacco. Plant Science. 165:1109-1115.

Lokko Y, Okogbenin E, Mba C, Dixon A, Raji A, Fregene M (2007). Cassava. In: Genome Mapping and Molecular Breeding in Plants Volume 3. Pulses, Sugar and Tuber Crops. 320pp. Kole C ed.

Makame M, Akoroda MO & Hahn SK (1987). Effects of reciprocal stem grafts on cyanide translocation in cassava. J Agric Sci Cambridge. 109:605-608

Maruyama A, Ishizawa K, Takagi T, Esashi Y (1998). Cytosolic β -Cyanoalanine Synthase Activity Attributed to Cysteine Synthases in Cocklebur Seeds . Purification and Characterization of Cytosolic Cysteine Synthases. Plant Cell;39(7):671-680.

Maxwell DP, Wang Y, McIntosh L (1999). The alternative oxidase lowers mitochondrial reactive oxygen production in plant cells. Proc Natl Acad Sci USA 96: 8271-8276

McMahon J and Sayre R (1994). Regulation of cyanogenic potential in cassava (Manihot esculenta Crantz). In: W.M. Roca and A.M. Thro (Eds.) Proceedings of the Second International Scientiﬁc Meeting of the Cassava Biotechnology Network, pp. 423–438, Bogor, Indonesia.

McMahon J, White W and Sayre R (1995). Cyanogenesis in cassava (Manihot esculenta). J. Exp. Bot. 46: 731–741.

201

Medentsev, A G, A Yu Arinbasarova, and V K Akimenko (2001). Intracellular cAMP Content and the Induction of Alternative Oxidase in the Yeast Yarrowia lipolytica. Microbiology 70, no. 1: 22-25.

Millar A. H., Liddell A and C. J. Leaver (2007). Isolation and subfractionation of mitochondria from plants. Methods in Cell Biology, vol. 80: 65-90.

Millar, A.H. and C.J. Leaver (2000). The cytotoxic lipid peroxidation product, 4-hydroxyl-2- nonenal speciﬁcally inhibits dehydrogenase in matrix of plant mitochondria. FEBS Lett. 481: 117–121.

Miller G, Shulaev V, Mittler R (2008). Reactive oxygen signaling and abiotic stress. Physiol Plant 133:481-489.

Miller G, Suzuki N, Ciftci-Yilmaz S, Mittler R (2010). Reactive oxygen species homeostasis and signalling during drought and salinity stresses. Plant, cell & environment 33(4):453-67.

Miller JM, Conn EE (1980). Metabolism of Hydrogen Cyanide by Higher Plants1. Plant Physiology:1199-1202.

Miller RW, Sirois J, Morita H (1975). The Reaction of Coumarins with Horseradish Peroxidase1. Plant Physiology 55: 35-41.

Mittler R (2002). Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci. 7, 405- 410.

Miwa S, Muller FL, Beckman KB (2008). The Basics of Oxidative Biochemistry. In: Oxidative Stress in Aging; From Model Systems to Humans. Miwa S, Muller FL, Beckman KB eds. Humana Press pp 11-38.

Miyakawa S, Cleaves HJ and SL Miller (2002). The Cold Origin Of Life: Hydrogen Cyanide And Formamide. Origins of Life and Evolution of the Biosphere. 32:195-208.

202

Mkpong OE, Yan H, Chism G & Sayre RT (1990). Purification, characterization, and localization of linamarase in cassava. Plant Physiol. 93:176-181.

Mlingi NLV, Bainbridge Z (1994). Reduction of cyanogen levels during sun-drying of cassava in Tanzania. Acta Horticulturae 375:233-239.

Møller BL (2010). Functional diversifications of cyanogenic glucosides. Current opinion in plant biology 13(3): 338-47.

Møller, Ian M, Allan G Rasmusson, James Jim N Siedow, and Greg C Vanlerberghe (2010).

The product of the alternative oxidase is still H2O. Archives of biochemistry and biophysics 495, no. 1: 93-4; author reply 95-6.

Møller IM (2001). Plant Mitochondria and Oxidative Stress: Electron Transport, NADPH Turnover, and Metabolism of Reactive Oxygen Species. Annu. Rev. Plant Physiol. Plant Mol. Biol. 2001. 52:561–91

Morante N, Sanchez T, Ceballos H, Calle F, Perez JC, Egesi C, Cuambe CF, Escobar D, Ortiz D, Chavez AL and Fregene M (2010). Tolerance to Postharvest Physiological Deterioration in Cassava Roots. Crop Science 50(4): 1333-1338.

Moore AL, Albury MS, Crichton PG, Affourtit C (2002). Function of the alternative oxidase: is it still a scavenger? Trends Plant Sci 7: 478–481.

Muller A and Weiler EW (2000). IAA-synthase, an enzyme complex from Arabidopsis thaliana catalyzing the formation of indole-3-acetic acid from (S)-tryptophan. Biol. Chem. 381, 679-686.

Murashige T & Skoog F (1962). A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol Plant. 15:473-497.

Murray RK, Granner DK, Mayes PA, Rodwell VW (2003). Harper's illustrated biochemistry. McGraw Hill. 693pp

203

Nahrstedt A (1985). Cyanogenesis and the role of cyanogenic compounds in insects. Plant Syst. Evol.150:35–47.

Nambisan B and Sundaresan S (1994). Distribution of linamarin and its metabolizing enzymes in cassava tissues. J Sci Food Agric. 66:503-507.

Nartey, F (1969). Studies on cassava Manihot utillisima, biosynthesis of asparagines-14 C from 14 C-labelled hydrogen cyanide and its relations with cyanogenesis. Physiol. Plant. 22: 1085– 1096.

Nassar NMA (2007). Cassava Genetic Resources: Wild Species and indigenous cultivars and their utilization for breeding of the crop. In: Ortiz, R. and Nassar, N.M.A. (eds) Proceedings of the First International Meeting on Cassava Breeding, Biotechnology and Ecology. Universidade de Brazilia, Brazil, pp. 5-30.

Negi S, Sukumar P, Liu X, Cohen JD, Muday GK (2010). Genetic dissection of the role of ethylene in regulating auxin-dependent lateral and adventitious root formation in tomato. The Plant journal;61(1):3-15.

Ndunguru GT, Thomson M, Waida TDR, Rwiza E, Jeremiah S and Westby A (1999). Relationship between quality and economic value of dried cassava products in Mwanza, Tanzania. In: Akoroda, M.O. and Terri, J. (eds) Food Security and Crop Diversification in SADC Countries. The Role of Cassava and Sweetpotato. SARRNET, Malawi, pp. 408–414.

Nweke FI, Spencer DSC, Lynam JK (2002). The Cassava Transformation. Michigan State University Press, East Lansing, MI.

Nyirenda DB, Chiwona-Karltun L, Chitundu M, Haggblade S, Brimer L (2010). Chemical safety of cassava products in regions adopting cassava production and processing – Experience from Southern Africa. Food Chem Toxicol. In Press, Corrected Proof.

O‟Reilly C and Turner PD (2003). The nitrilase family of CN hydrolysing enzymes– a comparative study. J. Appl. Microbiol. 95, 1161–1174.

204

Omole TA (1977). Cassava in the nutrition of layers. In: Nestel, B. and Graham, M. (eds.) Cassava as Animal Feed. IDRC, Ottawa, pp. 51–55.

Orozco-cárdenas ML, Narváez-vásquez J, Ryan CA (2001) Hydrogen Peroxide Acts as a Second Messenger for the Induction of Defense Genes in Tomato Plants in Response to Wounding , Systemin , and Methyl Jasmonate. Plant Cell 13:179-191.

Osuntokun B (1981). Cassava diet, chronic cyanide intoxiﬁcation and neuropathy in Nigerian Africans. World Rev Nutr Diet 36:141–173.

Otoo JA (1996). Rapid multiplication of cassava. IITA Research Guide 51.

Oyewole OB and Odunfa SA (1988). Microbiological studies on cassava fermentation for „lafun‟ production. Food Microbiology 5, 125–133.

Pace HC and Brenner C (2001). The nitrilase superfamily: classiﬁcation, structure and function. Genome Biol. 2 (1)1–9.

Papenbrock J, Schmidt A (2000). Characterization of a sulfurtransferase from Arabidopsis thaliana. Eur. J. Biochem. 267, 145-154.

Park WJ, Kriechbaumer V, Muller A, Piotrowski M, Meeley RB, Gierl A, Glawischnig E (2003). The Nitrilase ZmNIT2 Converts Indole-3-Acetonitrile to Indole-3-Acetic Acid. Plant Physiology.133:794-802.

Passam H C (1976). Cyanide-insensitive respiration in root tubers of cassava (Manihot esculenta Crantz. Plant Science Letters 7: 211-218.

Peiser GD, Wang TT, Hoffman NE, Yang SF, and CT Walsh (1984). Formation of cyanide from carbon 1 of 1-aminocyclopropane-1-carboxylic acid during its conversion to ethylene. Proc. Natl. Acad. Sci. USA 81: 3059–3063.

205

Piotrowski M, SchoÈ nfelderc S and Weiler EW (2001). The Arabidopsis thaliana isogene NIT4 and its orthologs in tobacco encode β-cyano-L-alanine hydratase/nitrilase. J. Biol. Chem. 276, 2616-2621.

Pirrung MC, Brauman JI (1987). Involvement of cyanide in the regulation of ethylene biosynthesis. Plant Physiol Biochem 25: 55–61

Plumbley R and Rickard J E (1991). Post-Harvest Deterioration of Cassava, in: Tropical Science, Vol. 31 pp 295-303.

Polidorosa AN, Mylonab PV, and Arnholdt-Schmitt B (2009). AOX gene structure, transcript variation and expression in plants. Physiol. plant. 137(4):342-353.

Popov VN, Simonina RA, Skulachev VP, Starkov AA (1997). Inhibition of the alternative oxidase stimulates H2O2 production in plant mitochondria. FEBS Lett 415: 87–90.

Poulton JE (1990). Cyanogenesis in plants. Plant Physiol. 94: 401–405.

Purvis AC (1997). Role of the alternative oxidase in limiting superoxide production in plant mitochondria. Physiol Plant 100: 165–170.

Purvis AC, Shewfelt RL (1993). Does AOX ameliorate chilling injury. Physiol Plant. 88:712- 718.

Ramanujam T & Indira P (1984). Effect of girdling on the distribution of total carbohydrates and hydrocyanic acid in cassava. Indian J Plant Phys. 27:355-360.

Rassmuson AG, Fernieb AR, van Dongenb JT (2009). Alternative oxidase: a defence against metabolic ﬂuctuations? Physiol. plant. 137(4):371-382.

Rea G, de Pinto MC, Tavazza R, Biondi S, Gobbi V, Ferrante P, Gara LD, Federico R, Angelini R and Tavladoraki P (2004). Ectopic Expression of Maize Polyamine Oxidase and Pea Copper Amine Oxidase in the Cell Wall of Tobacco Plants. Plant Physiology 134: 1414–1426.

206

Reilly K, Gomez-Vasquez R, Buschmann H, Tohme J and Beeching JR (2004). Oxidative stress responses during cassava post-harvest physiological deterioration. Plant Molecular Biology: 669- 685.

Rhoades JD and Loveday J (1990). Salinity in irrigated agriculture. In American Society of Civil Engineers, Irrigation of Agricultural Crops (Monograph 30) (Steward, B.A. and Nielsen, D.R., eds), pp. 1089–1142, American Society of Agronomists.

Rickard JE (1985). Physiological deterioration in cassava roots. J. Sci. Food Agri., 36: 167-176.

Rickard JE and Coursey DG (1981). Cassava storage, part 1: storage of fresh cassava roots. Tropical Science 23, 1–32.

Ross JJ, Symons GM, Aba L, Reid JB, Lusching C (2006). Hormone distribution and transport. In: Annual Reviews 24; Plant Hormone Signaling. Pp 257-292. Hedden P, Thomas SG, eds.

Rosling H (1974). Cassava cyanide and epidemic spastic paraparesis: a study in Mozambique on dietary cyanide exposure. ACTA Universitatis Upsalienis, Ph.D. thesis, Uppsala University, Sweden.

Rosling H (1994). Measuring the effects in human of dietary cyanide exposure from cassava. Acta Hort. 375:271-283.

Rusike J, Mahungu NM, Jumbo, Sandifolo VS and G Malindi (2010). Estimating impact of cassava research for development approach on productivity, uptake and food security in Malawi. Food Policy 35 (2): 98-111.

Sagi M, Fluhr R (2001). Superoxide Production by Plant Homologues of the gp91 phox NADPH Oxidase. Modulation of Activity by Calcium and by Tobacco Mosaic Virus Infection. Plant Physiology 126:1281-1290.

Sairam RK, Tyagi A (2004). Physiology and molecular biology of salinity stress tolerance in plants. Current Science 86 (3):407-421.

207

Salcedo A, Valle A, Sanchez B, Ocasio V, Ortiz A, Marquez P and Siritunga D (2010). Comparative evaluation of physiological post-harvest root deterioration of 25 cassava (Manihot esculenta) accessions: visual vs. hydroxycoumarins fluorescent accumulation analysis. African Journal of Agricultural Research 5, no. 22: 3138-3144.

Sambrook J, Fritsch EF, Maniatis T (1989). Molecular cloning: a laboratory manual, 2nd edn. New York. Cold Spring Harbor Laboratory Press.

Sanchez T, Chávez AL, Ceballos H, Rodriguez-Amaya DB, Nestel P, Ishitani M (2005). Reduction or delay of post-harvest physiological deterioration in cassava roots with higher carotenoid content. J. Sci. Food Agric. 86, 634–639.

Schaal BA, Olsen KM and Luiz JCB, Carvalho LJ (2006). Evolution, Domestication, and Agrobiodiversity in the Tropical Crop Cassava. Pp226-284. In: Darwin‟s harvest; New Approaches to origins, evolution and conservation of crops. Motely, TJ., Zerega, N and Cross, H Eds. Columbia University Press, NY

Schmidt RC, MuÈ ller A, Hain R, Bartling D and Weiler EW (1996). Transgenic tobacco plants expressing the Arabidopsis thaliana nitrilase II enzyme. Plant J. 9, 83-691.

Schultz F (1949). Cyanate. Experientia 5, 133–171.

Selmar D (2010). Biosynthesis of cyanogenic glycosides, glucosinolates and non-protein amino acids. Annual Plant Reviews 40, 92-181.

Selmar D, Lieberei R, Biehl B (1988). Mobilization and utilization of cyanogenic glycosides. Plant Physiology 86: 711–716.

Selmar D (1993). Transport of cyanogenic glycosides: uptake of linustatin by Hevea cotyledons. Planta 191: 191–199.

Selma D (1994). Translocation of cyanogenic glucosides in cassava. Acta Horticulturae 375: 61-67.

208

Shepherd G, Velez LI (2008). Role of hydroxocobalamin in acute cyanide poisoning. Ann Pharmacotherapy 42 (5): 661–9.

Shewry PR (2003). Tuber Storage Proteins. Annals of Botany 91: 755-769.

Shibamoto T, Bjeldanes LF (1993). Introduction to Food Toxicology. Academic Press, San Diego, California. pp: 213.

Shirai R and Kurihara T (1991). Distribution of rhodanese in plants. Bot. Mag. Tokyo 104:341- 346.

Sluse, F E, and W Jarmuszkiewicz (1998). Alternative oxidase in the branched mitochondrial respiratory network : an overview on structure , function , regulation , and role. Brazilian Journal Of Medical And Biological Research 31: 733-747.

Sieger SM, Kristensen BK, Robson CA, Amirsadeghi S, Eng EWY, Abdel-Mesih A, Møller IM, Vanlerberghe GC (2005). The role of alternative oxidase in modulating carbon use efﬁciency and growth during macronutrient stress in tobacco cells. J Exp Bot 56: 1499–1515

Siripornadulsil S, Traina S, Verma PS, Sayre RT (2002). Molecular Mechanisms of Proline- Mediated Tolerance to Toxic Heavy Metals in Transgenic Microalgae. Plant Cell. 14: 2837- 2847.

Siritunga D, Arias-Garzon D, White W, Sayre RT (2004). Overexpression of hydroxynitrile lyase in transgenic cassava roots accelerates cyanogenesis and food detoxiﬁcation. Plant Biotechnology Journal 2: 37–43.

Siritunga D and Sayre R (2003). Generation of cyanogen-free transgenic cassava. Planta. 217:367-373.

Siritunga D (2002). Generation of Acyanogenic Cassava (Manihot Esculenta Crants): Transgenic Approaches. PhD thesis. The Ohio State University.

209

Sirois JC and RW Miller (1972). The Mechanism of the Scopoletin-induced Inhibition of the Peroxidase Catalyzed Degradation of Indole-3-acetate. Plant Physiology 49:1012-1018.

Smith CA, Melino VJ, Sweetman C, Soole KL (2009). Manipulation of alternative oxidase can influence salt tolerance in Arabidopsis thaliana. Physiologia Plantarum 137: 459–472.

Smith FW, Hawkesford MJ, Ealing PM, Clarkson DT, Berg PJV, Belcher AR, Warrilow GS (1997). Regulation of expression of a cDNA from barley roots encoding a high affinity sulphate transporter. The Plant journal:12(4):875-84.

Smithers, A G, and J F Sutcliffe (1967). Effects of Cyanide on Respiration , Growth , and Potassium Absorption by Carrot Root Tissue Cultures. 18 no. 57: 758-768.

Solomonson LP (1981). Cyanide as a metabolic inhibitor. In Cyanide in Biology. Edited by Vennesland, B., Conn, E.E., Knowles, C.J., Westley, J. and Wissing, F. pp. 11-28. Academic Press, New York.

Stemmer WPC, Crameri A, Ha KD, Brennan TM, Heyneker HL (1995). Single-step assembly of a gene and entire plasmid from large numbers of oligodeoxyribonucleotides. Gene 164: 49-53.

Tam YY, Epstein E, Normanly J (2000). Characterization of Auxin Conjugates in Arabidopsis . Low Steady-State Levels of Indole-3-Acetyl-Aspartate. Plant Physiology.123:589-595.

Taylor NJ, Edwards M, Kiernan RJ, Davey CMD, Blakestey D, Henshaw GG (1996). Development of friable embryogenic callus and embryogenic suspension systems in cassava (Manihot esculenta Crantz). Nature Biotechnol. 14(6):726-730.

Tor-Agbidye J, Palmer VS, Lasarev MR, Craig MA, Blythe LL, Sabri MI, Spencer PS (1999). Bioactivation of cyanide to cyanate in sulfur amino acid deficiency: relevance to neurologic disease in humans subsisting on cassava. Toxicol Sci 999;50:228–35.

210

Tor-Agbidye J, Palmer VS, Sabri MI, Craig MA, Blythe LL, Spencer PS (1998). Dietary deﬁciency of cystine and methionine in rats alters thiol homeostasis required for cyanide detoxiﬁcation, J. Toxicol. Environ. Hlth. 55, 583–595.

Torres MA, Dangl JL (2005). Functions of the respiratory burst oxidase in biotic interactions, abiotic stress and development. Current opinion in plant biology; 8(4):397-403.

Triantaphylidès C, Krischke M, Hoeberichts FA, Ksas B, Gresser G, Havaux M, Breusegem FV, Mueller MJ (2008). Singlet oxygen is the major reactive oxygen species involved in photooxidative damage to plants. Plant physiology. 148(2):960-8.

Tsunoda H, Yamaguchi K (1995). The cDNA sequence of an auxin-producing nitrilase homolog in tobacco. Plant Physiol. 109, 339.

Tylleskar T (1994). The association between cassava and the paralytic disease Konzo. Acta Hortic; 375:331–9.

Umbach AL, Wiskichb JT, Siedow JN (1994). Regulation of alternative oxidase kinetics by pyruvate and intermolecular disulfide bond redox status in soybean seedling mitochondria. FEBS Letters; 348:181-184.

Van Breusegem F, Bailey Serres J, Mittler R (2008). Unraveling the tapestry of networks involving reactive oxygen species in plants. Plant Physiol 147:978 -984

Vanlerberghe GC, Robson CA, Yip JYH (2002). Induction of Mitochondrial Alternative Oxidase in Response to a Cell Signal Pathway Down-Regulating the Cytochrome Pathway Prevents Programmed Cell Death. Plant Physiology;129:1829-1842.

Vanlerberghe GC, Cvetkovska M, Wang J (2009). Is the maintenance of homeostatic mitochondrial signaling during stress a physiological role for alternative oxidase? Physiol. Plant. 137: 392–406.

Vanlerberghe GC, McIntosh L (1992). Lower growth temperature increases alternative pathway capacity and alternative oxidase protein in tobacco. Plant Physiology 100: 115–119 211

Vanlerberghe GC, McIntosh L (1996). Signals regulating the expression of the nuclear gene encoding alternative oxidase of plant-mitochondria. Plant Physiology 111: 589–595

Vanlerberghe GC, Mcintosh L (1997). Alternative Oxidase: From Gene to Function. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1997. 48:703–34.

Vanlerberghe GC, Vanlerberghe AE, McIntosh L (1994) Molecular genetic alteration of plant respiration (silencing and overexpression of alternative oxidase in transgenic tobacco). Plant Physiology 106: 1503–1510

Vasconcelos AT, Twiddy DR, Westby A, Reilly PJA (1990). Detoxiﬁcation of cassava during gari preparation. Intl J Food Sci Tech 25:198-203.

Vorwerk S, Biernacki S, Hillebrand H, Janzik I, MuÈ ller A, Weiler EW and Piotrowski M (2001). Enzymatic characterization of the recombinant Arabidopsis thaliana nitrilase subfamily encoded by the NIT2/NIT1/NIT3-gene cluster. Planta 212: 508-516.

Wang S, Volini M (1968). Studies on the Active Site of Rhodanese. Journal of Biological Chemistry 243(20):5465-5470.

Warrilow AGS, Hawkesford MJ. Modulation of cyanoalanine synthase and O-acetylserine (thiol) lyases A and B activity by beta-substituted alanyl and anion inhibitors (2002). Journal of experimental botany;53(368):439-45.

Warrilow a G, Hawkesford MJ (2000). Cysteine synthase (O-acetylserine (thiol) lyase) substrate specificities classify the mitochondrial isoform as a cyanoalanine synthase. Journal of experimental botany.51(347):985-93.

Wenham JE (1995). Post-harvest deterioration of cassava: a biotechnology perspective. Plant Production and Protection Paper FAO No 130. FAO Plant Production and Protection, Rome Italy 99 p.

212

Westby A (2002). Cassava Utilization, Storage and Small-scale Processing. In: Hillocks RJ, Thresh JM, Belloti AC (eds.) Cassava: biology, production and utilization. CAB International Publishing, Wallingford. pp. 281-300.

Westley J (1981). Thiosulfate: cyanide sulfur transferase (rhodanese). Meth Enzymol; 77:85–91.

Wheatley CC and Chuzel G (1993). Cassava: the nature of the tuber and use as a raw material. In: Macrae, R., Robinson, R.K. and Saddler, M.J. (eds.) Encyclopedia of Food Science, Food Technology and Nutrition. Academic Press, San Diego, pp. 734–743.

Wheatley C, Lozano C, and Gomez G (1985). Post-harvest deterioration of cassava roots, Cassava: Research, Production and Utilization. UNDP, CIAT, 655,671.

White W, McMahon J, Sayre RT (1994). Regulation of cyanogenesis in cassava. Acta Hortic 375: 69–78.

White W, Sayre RT (1995). The characterization of hydroxynitrile lyase for the production of safe food products from cassava (Manihot esculenta, Crantz) In DL Gustine, HE Flores, eds, Phytochemicals and Health, Current Topics in Plant Physiology, Vol 15. American Society of Plant Physiologists, Rockville, MD, pp 303–304.

White WLB, Arias-Garzon DI, McMahon JM and Sayre RT (1998). The role of hydroxynitrile lyase in root cyanide production. Plant Physiology 116: 1219-1225.

Williamson KS, Hensley K, Floyd RA (2003). Fluorometric and Colorimetric Assessment of Thiobarbituric Acid-Reactive Lipid Aldehydes in Biological Matrices. Henseley K, Floyd RA eds. Methods in oxidative stress pp 57-65. Humana Press.Totowa, NJ

Wurtele EVES, Nikolau BJ, Conn EE (1984). Tissue Distribution of β-Cyanoalanine Synthase. Plant Physiology. 75; 979-982.

Yang SF (1989). Metabolism of 1-aminocyclopropane-1-carboxylic acid in relation to ethylene biosynthesis. In "Plant Nitrogen Metabolism" (JE Poulton, JT Romeo, EE Conn eds), Rec. Adv. Phytochem., Vol. 23, Plenum Press, New York, pp. 263-287. 213

Yip WK, Yang S (1998). Ethylene biosynthesis in relation to cyanide metabolism. Botanical Bulletin of Academia Sinica 39:1-7.

Zhu J (2001). Plant salt tolerance. Trends in Plant Science. 6(2):66-71.

214