Investigating Water Responsive Actuation Using the Resurrection Plant Selaginella Lepidophylla As a Model System

Investigating Water Responsive Actuation using the Resurrection Plant Selaginella lepidophylla as a Model System

Véronique Brulé

Department of Biology

McGill University, Montréal

Summer 2018

A thesis submitted to McGill University in partial fulfillment of the requirements of the

degree of Doctor of Philosophy

ABSTRACT

Nature is a wealth of inspiration for biomimetic and actuating devices. These devices are, and have been, useful for advancing society, such as giving humans the capability of flight, or providing household products such as velcro. Among the many biomimetic models studied, plants are interesting because of the scope of functions and structures produced from combinations of the same basic cell wall building blocks.

Hierarchical investigation of structure and composition at various length-scales has revealed unique micro and nano-scale properties leading to complex functions in plants.

A better understanding of such micro and nano-scale properties will lead to the design of more complex actuating devices, including those capable of multiple functions, or those with improved functional lifespan.

In this thesis, the resurrection plant Selaginella lepidophylla is explored as a new model for studying actuation. S. lepidophylla reversibly deforms at the organ, tissue, and cell wall level as a physiological response to water loss or gain, and can repeatedly deform over multiple cycles of wetting and drying. Thus, it is an excellent model for studying properties leading to reversible, hierarchical (i.e., multi length-scale) actuation.

S. lepidophylla has two stem types, inner (developing) and outer (mature), that display different modes of deformation; inner stems curl into a spiral shape while outer stems curl into an arc shape.

In depth investigation of S. lepidophylla revealed that morphological and compositional gradients result in hierarchical stiffness gradients leading to differential tissue swelling/shrinking and, ultimately, directional stem (un)curling. Morphological

2 gradients are observed at the tissue level between adaxial and abaxial stem sides, and include changes in tissue density, cell shape, size, and cell angle relative to the stem axis.

Compositional gradients are observed along the length of the stem from tip to base at both tissue and cell wall levels and include changes in lignification and xylan distribution. Comparison between inner and outer S. lepidophylla stem types showed that morphological gradients are most likely involved in directional stem bending while compositional gradients appear to contribute to the degree of stem curling.

RÉSUMÉ

La biologie est source d’inspiration pour la production d’appareils biomimétiques et actionnement qui font avancer la technologie. Parmi les modèles biologiques étudiés, les plantes sont des sujets intéressants parce qu’elles présentent plusieurs types de fonctions et structures dérivées des mêmes matériaux de base (i.e., les polymères des parois cellulaires). Des enquêtes hiérarchiques ont démontré que des traits de morphologie et composition au niveau micro et nano sont responsables de fonctions complexes dans les plantes. En étudiant des modèles hiérarchiques pour mieux comprendre les traits qui provoquent le mouvement, on serait en mesure de les reproduire dans des matériaux synthétiques avec des fonctions plus complexes, ou avec une durée de fonction prolongée.

Cette thèse explore les espèces Selaginella, une ‘plante de la résurrection’, comme nouveau modèle d’actionnement. Les plantes de la résurrection changent de forme réversiblement à plusieurs échelles de longueur (organe, cellule, paroi cellulaire) en réponse aux fluctuations de contenu en eau. Le changement de forme se répète sur plusieurs cycles de déshydratation et réhydratation. Donc, ces espèces sont des modèles idéaux pour étudier les traits qui provoquent l’actionnement hiérarchique et répétable. De plus, les tiges de S. lepidophylla démontrent des modalités de mouvement différentes. En déshydratant, les tiges intérieures se recourbent sur elles-mêmes en spirale, et les tiges extérieures se courbe en forme d’arc.

Une enquête compréhensive de S. lepidophylla a démontré que des changements gradés en morphologie et composition résultent en des changements gradés et

4 hiérarchiques de rigidité responsables de l’engorgement et l’assèchement différentiels des tissus, ce qui explique le mouvement directionnel des tiges. Les changements gradés en morphologie s’observent au niveau du tissu entre les régions adaxiale et abaxiale de la tige incluant des changements dans la densité du tissu ainsi que l’angle cellulaire relatif à l’axe de la tige. Les changements gradés en composition sont observables au long de la tige, au niveau du tissu et la paroi cellulaire, et incluent des changements en lignification et en distribution de xylane. En comparant des tiges vivantes et mortes de S. lepidophylla, il a été démontré que les changements en morphologie sont responsables pour du mouvement directionnel de la tige, et que les changements en composition déterminent le point où la tige peut courber.

ACKNOWLEDGMENTS

I would like to begin by thanking Dr. Tamara Western for her all her guidance and support during my research project. Tamara, I feel privileged to have learned so much from you. Not only did you show me how to perform research, but you fostered my love for teaching and helped me to grow as an instructor. Thanks for all our conversations, from the serious ones right down to cat stories and chats about writing.

Thank you to my collaborators in engineering: Dr. Damiano Pasini, Dr. Ahmad

Rafsanjani, and Dr. Meisam Asgari. Mechanics was a steep learning curve, but you showed me how interesting a field it is and how to navigate it best from a biologist’s point of view. I appreciate all your insight and guidance, and am grateful to have worked alongside you. I would also like to thank my committee members, Dr. Tom Bureau and

Dr. Alejandro Rey, for your excellent feedback on my project.

Thanks to my awesome labmates: Mike Ogden, Jonathan Palozzi, Lydiane

Gaborieau, and Bronwen Froward. Coming to the lab was guaranteed to be fun with all of you around! And to Joe, Frank Anne-Marie, and Kathy – thanks for all the great conversations and laughs! You made Stewart Bio a memorable place.

MB, this journey would have been a lot more difficult without your daily smiles, words of encouragement, and all your homemade bread! Manimal and Taya, you are

THE best furry thesis co-writers. Your love and snuggles kept me sane. Mom B and Pop

B, I am forever grateful for all your love, support and your belief in me. You gave me the strength and confidence to see this through.

Finally, thank you to everyone else whose feedback or kind words helped me on this PhD journey! J

AUTHOR CONTRIBUTION AND CHAPTER OVERVIEW

This thesis contains four chapters, three of which are original research. Chapter 1 is a literature review that provides an overview of the various topics explored in this research project. Chapters 2-4 contain original contribution to our understanding of the hierarchical compositional and morphological features, as well as the mechanical forces arising from these features, that drive stem deformation in the resurrection plant

Selaginella lepidophylla. Appendix 1 includes a meta-analysis of stiffness in Arabidopsis thaliana as it relates to plant function, and also highlights other plant models used to study stiffness in relation to plant deformation and actuation. Appendices 2-3 include small experiments that complement the work presented in Chapters 2-4, but that will not be included in any manuscripts. Finally, Appendix 4 preliminarily explores the relationship between water movement and corresponding vegetative tissue deformation in both Selaginella lepidophylla and Myrothamnus flabellifolius.

Chapter 2: This chapter is published. I performed the microscopy, timelapse video capture, and weight loss quantification, while Dr. Ahmad Rafsanjani performed the computational modelling. We both performed tensile testing experiments. The manuscript was co-written by Dr. Rafsanjani and I, and Drs. Tamara Western and

Damiano Pasini reviewed and provided feedback on the manuscript.

Chapter 3: This chapter is in preparation for publication and is expected to be submitted within the next month. I carried out all the presented work, with the following exceptions:

Dr. Rafsanjani performed the micro-computed x-ray tomography experiments described

7 in the text and presented in Figure 3.2. Dr. Meisam Asgari performed the atomic force microscopy imaging and indentation presented in Figure 3.3. Again, Drs. Tamara

Western and Damiano Pasini reviewed and provided feedback on the manuscript.

Chapter 4: This chapter is in preparation for publication for NanoLetters (which combines results and discussion into a single section), and is expected to be submitted within the next month. Dr. Asgari performed the atomic force microscopy experiments while I carried out the other microscopy (brightfield, immunofluorescence, and scanning electron microscopy) presented in the chapter. Drs. Tamara Western and Damiano Pasini reviewed and provided feedback on the manuscript.

Appendix 1: This work is published. Dr. Tamara Western and I wrote the bulk of the manuscript, with contributions from Drs. Damiano Pasini and Ahmad Rafsanjani.

Appendices 2-3: I performed all these experiments myself.

Appendix 4: The data presented here will be prepared for publication, along with x-ray tomography data collected from the Canadian Light Source (CLS), and data collected from the LemnaTec HTS. I carried out all the presented work, with the following exceptions: Mike Ogden and I shared work on data collection at CLS, and Dr. Emilio

Vello helped me with programming image capture on the LemnaTec HTS. Mike Ogden will be involved with writing the x-ray tomography results, and Drs. Tamara Western and

Damiano Pasini will review and provide feedback on the manuscript. This manuscript is expected to be ready for submission by the end of the summer/early fall.

TABLE OF CONTENTS

ABSTRACT 2 RÉSUMÉ 4 ACKNOWLEDGMENTS 6 AUTHOR CONTRIBUTION AND CHAPTER OVERVIEW 7 TABLE OF CONTENTS 9 LIST OF ABBREVIATIONS 14 LIST OF FIGURES 15 LIST OF TABLES 17 CHAPTER 1 – Literature Review: Biomimetics, Water-Responsive Actuation, and Resurrection Plants as Models for Hierarchical Actuation 18 INTRODUCTION 19 SECTION 1: BIOMIMETICS, ACTUATION, AND APPLICATIONS 21 1.1: Biomimetics And Actuation 21 1.2: Applications For Actuation 22 1.3: Fluid-Driven Actuation 24 SECTION 2: ACTUATORS IN THE PLANT KINGDOM 27 2.1: Swelling And Shrinking 27 2.2: Osmotic Gradients 34 2.3: Plant-Inspired Actuator Prototypes 36 2.4: Limitations Of Current Fluid-Driven Hydroactuators 39 2.5: A Potential New Model For Studying Hygroactuation 41 SECTION 3: RESURRECTION PLANTS AS COMPLEX ACTUATORS 42 3.1: Resurrection Plants 42 3.2: Molecular Responses To Desiccation 46 3.3: Physiological Responses To Desiccation 51 3.4: Resurrection Plants Investigated In This Research Project 55 3.4.1: Selaginella Lepidophylla 55 3.4.2: Myrothamnus Flabellifolius 61

SECTION 4: USING RESURRECTION PLANTS TO ELUCIDATE PROPERTIES OF HYGROACTUATION 65 4.1: Availability, Type, And Strength Of Materials For Hydroactuators 65 4.2: Hygroactuator Longevity And Conformational Reversibility 66 RESEARCH OBJECTIVES 68 Objective 1: Characterization Of Plant Movement 68 Objective 2: Morphological And Compositional Features Responsible For Organ-Level Actuation 69 Objective 3: Water Dynamics 69 LINK BETWEEN CHAPTER 1 AND 2 81 CHAPTER 2 – Hydro-Responsive Curling of the Resurrection Plant Selaginella lepidophylla 82 ABSTRACT 83 INTRODUCTION 84 RESULTS 87 Plant Morphology 87 Shape Transformation of Stems 87 Morphology And Composition Of Stem Sections 88 Curvature Characterization Of Stems 89 Mechanical Response Of Stems To Dehydration 90 Curling Mechanisms Of Inner And Outer Stems 91 DISCUSSION 94 MATERIALS AND METHODS 97 SUPPLEMENTARY INFORMATION 109 S1. Preparation of Spurr’s Resin-embedded Sections 110 S2. Preparation of Paraffin-embedded Sections 111 S3. Discrete Curvature Characterization 113 S4. Geometrical Model Based on Euler Spiral 114 S5. Finite element Model for Bilayer Stems 115 LINK BETWEEN CHAPTER 2 AND 3 121

CHAPTER 3 – Morphological and Compositional Gradients Direct Stem Curvature and Deformation in the Resurrection Plant Selaginella lepidophylla 122 ABSTRACT 123 INTRODUCTION 124 RESULTS 126 Mechanical Properties of S. lepidophylla Stems 126 Morphological Differences Between Adaxial and Abaxial Stem Tissue 127 Gradients of Cell Wall Properties Along the Length of Inner Stems 129 DISCUSSION 132 Directional Deformation of S. lepidophylla Stems is Associated with Differential Tissue morphology and Cell Wall Properties 132 Lengthwise Gradients of Cell Wall Thickening, Lignification and Hemicellulose are Associated with Linear Curling in Inner S. lepidophylla Stems 134 Comparison of Deformation in S. lepidophylla to Other Established Plant Models 137 MATERIALS AND METHODS 140 SUPPLEMENTARY INFORMATION 158 LINK BETWEEN CHAPTER 3 AND 4 167 CHAPTER 4 – Characterizing Cell Wall Stiffness in the Resurrection Plant Selaginella lepidophylla 168 ABSTRACT 169 INTRODUCTION 170 RESULTS AND DISCUSSION 172 Cell Wall Stiffness Differs between Adaxial and Abaxial Stem Sides 172 Cell Wall Layering Differs between Adaxial and Abaxial Stem Sides 173 Cell Wall Composition Resembles Gelatinous Fibers of Tension Wood, Coiling Vines, and Ice Plant Seed Capsules 175 CONCLUSION 177 MATERIALS AND METHODS 179 SUPPLEMENTARY INFORMATION 190

CONCLUSION AND FUTURE WORK 202 LITERATURE CITED 206 APPENDICES 223 APPENDIX 1 – HIERARCHIES OF STIFFNESS 224 ABSTRACT 225 INTRODUCTION 226 TOP-DOWN INVESTIGATION OF PLANT STEM STIFFNESS HIERARCHIES IN THE REFERENCE PLANT ARABIDOPSIS THALIANA 228 Whole Plant 228 Organ Level 229 Tissue Level 231 Cell Level 234 Cell Wall Level 238 CAN ARABIDOPSIS BE USED AS A MODEL TO UNDERSTAND THE ROLE OF PARAMETERS THAT GOVERN PLANT STIFFNESS? 248 FUNCTIONAL GRADIENTS IN PLANT STIFFNESS 253 Stiffness Gradients and their Effects on Load-Bearing Capacity 253 Plant Stiffness Gradients and Plant Actuation 255 HARNESSING OF PLANT STIFFNESS: BIOMECHANICAL TAILORING OF PLANTS 257 CONCLUDING REMARKS AND FUTURE DIRECTIONS 260 BOX 1. TERMINOLOGY RELATED TO MECHANICS OF BIOLOGICAL SYSTEMS IN THE CONTEXT OF THIS PAPER 272 BOX 2. MECHANICAL TEST TYPES COMMONLY USED FOR TESTING PLANTS 274 REFERENCES 287 APPENDIX 2 – LIGNIN IDENTITY IN S. LEPIDOPHYLLA CORTICAL STEM TISSUE 296 APPENDIX 3 – RAMAN CONFOCAL SPECTROSCOPY ANALYSIS OF S. LEPIDOPHYLLA CORTICAL STEM TISSUE 302

APPENDIX 4: WATER MOVEMENT DURING ORGAN DEFORMATION IN THE RESURRECTION PLANTS SELAGINELLA LEPIDOPHYLLA AND MYROTHAMNUS FLABELLIFOLIUS 309 4.1. Water Movement Through S. lepidophylla Stems 311 4.2. Kinematics of Hydrating Myrothamnus flabellifolius Leaves 314

LIST OF ABBREVIATIONS

AFM: atomic force microscopy CLS: Canadian light source E: Young’s modulus (elasticity/stiffness) FAA: formaldehyde, alcohol, acetic acid FE: finite element G-fiber/-layer: gelatinous fiber/layer GFP: green fluorescent protein G-lignin: guaiacyl lignin LR: London resin MFA: microfibril angle

OsO4: osmium tetroxide PD: post-dehydration PEG: polyethylene glycol PO: propylene oxide

PO4: phosphate RFP: red fluorescent protein ROS: reactive oxygen species RWC: relative water content S-lignin: syringyl lignin TBO: Toluidine Blue O TBST: tris-buffered saline, tween TEM: transmission electron microscopy TOMCAT: tomographic microscopy and coherent radiology experiments TX2: texas red YFP: yellow fluorescent protein

LIST OF FIGURES

Figure 1.1. Hygroscopic Movement of Wheat and Pinecones 72

Figure 1.2. Gelatinous fibers in Coiling Tendrils/Vines and Tension Wood 74

Figure 1.3. Osmotic Gradients in Venus Flytrap and Touch-Me-Nots 76

Figure 2.1. Morphology and Composition of the Resurrection Plant Selaginella lepidophylla 101

Figure 2.2. Response of Outer and Inner Stems to Dehydration 103

Figure 2.3. Curling Mechanisms of Inner and Outer Stems 105

Figure 2.4. Curling Pattern of Bilayer Stems 107

Supplementary Figure 2.1. Discrete Representation of a Smooth Curve 112

Supplementary Figure 2.2. Finite Element Mesh for a Bilayer stem 117

Supplementary Figure 2.3. Comparison between FE Simulations and Theoretical Timoshenko Bimetallic Model for Normalized Curvature of Bilayer Stems 119

Figure 3.1. S. lepidophylla Stem Deformation, and Stem and Tissue Stiffness 147 Figure 3.2. Morphology and Tissue Structure in S. lepidophylla 149 Figure 3.3. Cell Wall Properties in S. lepidophylla 151 Figure 3.4. Summary of Gradients Responsible for Directional Stem Bending and The Degree of Stem Curling in S. lepidophylla 153

Supplementary Figure 3.1. Tissue Lignification as Detected with Basic Fuchsin 159 Supplementary Figure 3.2. LM10 Binding Pattern 161 Supplementary Figure 3.3. Antibody Binding Pattern in Outer Stem Cross-Sections 163 Figure 4.1. Water-Responsive Deformation in S. lepidophylla 182 Figure 4.2. Stiffness of Cortical Cell Walls in S. lepidophylla Stems 184 Figure 4.3. Cell Wall Layering in S. lepidophylla Cortex 186 Figure 4.4. S. lepidophylla Cortical Cell Wall Composition 188

Supplementary Figure 4.1. Topological Scans of Adaxial and Abaxial Cortical Cells And Cell Walls in the Middle Region of Inner S. lepidophylla Stems 191 Supplementary Figure 4.2. Cell Wall Layering Visualized by Transmission Electron Microscopy 193 Supplementary Figure 4.3. Cellulose Cell Wall Distribution 195 Supplementary Figure 4.4. Hemicellulose Cell Wall Distribution 197 Supplementary Figure 4.5. Pectin Cell Wall Distribution 199 Figure 5.1. A Schematic of the Hierarchy of Mechanical Parameters Influencing Stiffness Across Different Length Scales in Plants 263

Figure 5.2. Examples of Uniaxial Mechanical Forces that Act on Objects, and Common Tests used to Investigate the Mechanical Properties of Objects 265

Figure 5.3. Examples of Arabidopsis Inflorescence Stem Treatments and Mutants that Affect Hierarchical Parameters Governing Stiffness 267

Figure 5.4. The Role of Stiffness Gradients in Plant Function 270

Figure 5.5. Phloroglucinol Staining of S. lepidophylla Apical Cortex 298 Figure 5.6. Mäule Staining of S. lepidophylla Apical Cortex 300 Figure 5.7. Raman Confocal Spectroscopy of S. lepidophylla Apical Cortex 303 Figure 5.8. Analysis of S. lepidophylla Apical Cortex Peaks 305 Figure 5.9. Visualization of Water Movement through S. lepidophylla Stems 316 Figure 5.10. Change in S. lepidophylla Stem Hydration Over Time 318 Figure 5.11. Conformational Change in Isolated M. flabellifolius Leaves 320

LIST OF TABLES

Table 1.1. Commonly Studied Biological Actuators 78 Table 1.2. Commonly Studied Resurrection Plant Species 79 Table 3.1. Mechanical Properties of S. lepidophylla Stems 155 Table 3.2. Mechanical Properties of S. lepidophylla Stem Tissue 156 Table 3.3. Adaxial and Abaxial Cortical Tissue Cell Dimensions 157 Supplementary Table 3.1. S. lepidophylla Stem Dimensions 165 Supplementary Table 3.2. S. lepidophylla Tissue and Cell Wall Composition 166 Supplementary Table 4.1. S. lepidophylla Cortical Cell Wall Stiffness 201 Table 5.1. Mechanical Properties of Arabidopsis Stem Structure and Cell Wall Mutants 278 Table 5.2. Comparison of Mechanical Properties of Primary Cell Wall, Secondary Cell Wall and Stem Structural Mutants of Arabidopsis 284 Table 5.3. Cell Wall Peaks from Plant Raman Confocal Spectroscopy Literature 307 Table 5.4. Raman Confocal Spectroscopy of S. lepidophylla Apical Cortex and Possible Peak Identity 308 Table 5.5. Kinematic Analysis of Dehydrating M. flabellifolius Leaves 322 Table 5.6. Statistical Analysis of M. flabellifolius Leaf Deformation Across Cycles of Wetting and Drying 323

CHAPTER 1

Literature Review: Biomimetics, Water-Responsive Actuation, and

Resurrection Plants as Models for Hierarchical Actuation

INTRODUCTION

Biomimetics, also termed biomimicry, studies biological systems as sources of inspiration for human innovation. Without observing the way birds fly, for example, human flight would have remained a dream rather than a reality (Bar-Cohen, 2006).

Observing the function and appearance of various biological systems has led to bioinspired devices used both to solve human problems (e.g., engineering and medicine), as well as to advance human technology (e.g., flight and artificial intelligence) (Bar-Cohen,

2006).

However, imitating a biological system based on appearance alone cannot fully replicate its function or structure. Properties at microscopic (micro and nano) length scales are often responsible for the complex functions and structures observed at the macro (i.e., whole organism) length scale. Thus, bio-inspired devices based solely on macro-scale features do not capture the complexity or breadth of function that is observed in their biological counterparts (Bar-Cohen, 2006; Erb, Sander, Grisch, & Studart, 2013;

C. Lv et al., 2018). Current research in biomimetics has recognized this problem, and biological systems are now being studied at various length scales to gain a more comprehensive understanding of their function and structure (Q. Chen, Shi, Gorb, & Li,

2014; Faisal, Rey, & Pasini, 2013; Guo et al., 2017; Jensen & Fozard, 2015; K. Kim, Yi,

Zamil, Haque, & Puri, 2015).

While multi-length scale biological models exist (e.g., (Jensen & Fozard, 2015)), more models are countinually sought that display novel features. New models, such as resurrection plants (Brulé, Rafsanjani, Pasini, & Western, 2016), provide an opportunity

19 to examine biological systems with more complex functions and structures, the study of which could ultimately lead to bio-inspired devices with innovative properties.

The following literature review explores the themes presented in this thesis, including biomimetics, plant-based biomimetic models, and resurrection plants. Section 1 provides an overview of biomimetics and actuation, with a focus on fluid-driven actuation. Section 2 explores actuation in the plant kingdom and highlights the main plant models that have been studied for their mechanisms of fluid-driven actuation.

Section 3 focuses specifically on resurrection plants as fluid-driven actuators, and describes the two resurrection plant species investigated in this thesis (Selaginella lepidophylla and Myrothamnus flabellifolius). Finally, Section 4 discusses how studying resurrection plants can help to further our understanding of fluid-driven actuation.

SECTION 1: BIOMIMETICS, ACTUATION, AND APPLICATIONS

1.1: BIOMIMETICS AND ACTUATION

The field of biomimetics studies the inherent structural and compositional properties of biological systems that allow organisms to perform complex functions, and seeks to mimic these functions in synthetic materials and structures for human-based applications (Burgert & Fratzl, 2009a; Kempaiah & Nie, 2014; Meng & Li, 2013; Vaia &

Baur, 2008). An excellent – and relatively simple – example of biomimicry is Velcro.

Inspired by the hooking action of burrs to animals and clothing, George de Mestral created a synthetic material composed of two strips of fabric, one with small hooks that could secure to another layer created from small loops of material (De, 1955). Examples of biomimetic study can be traced as far back as ancient Greece (Aziz, 2016), but research in biomimicry only gained traction during the 1990s due to technological advancements that broadened the scope and depth of possible experimental techniques available to study biological systems (Arnarson, 2011; Benyus, 1997; Kennedy, Fecheyr-

Lippens, Hsiung, Niewiarowski, & Kolodziej, 2015; Sarikaya, 1994; Srinivasan, Haritos,

& Hedberg, 1991). Since then, the field of biomimetics has become an active area of research in a variety of disciplines (e.g., biology, chemistry, and engineering) (Kennedy et al., 2015).

More recently, biomimetic research has focused on the process of actuation in biological systems. Actuation is the generation of a mechanical response (e.g., twisting, folding, and oscillating) to a non-mechanical, external stimulus (e.g., pH, temperature, and electrical current) that results in conformational changes within a material or

21 structure (Erb et al., 2013; Johnson, Bonser, & Jeronimidis, 2009; Kempaiah & Nie,

2014; Meng & Li, 2013). The interaction of stimulus and material generates a mechanical response by either releasing stored mechanical energy, or by converting the energy of the stimulus into mechanical energy, which can then elicit conformational changes.

Depending on the material’s properties, deformation can occur at multiple length scales

(nanometre through to metre), can be reversible or irreversible, and can range in degree of complexity (simple uniaxial bending through to walking or swimming motions) (Erb et al., 2013; Studart, 2015; Studart & Erb, 2014).

A variety of naturally occurring biological actuators exists, including plant, animal and bacterial organisms (Table 1.1) (Fratzl & Barth, 2009; Meng & Li, 2013;

Srinivasan et al., 1991; Vaia & Baur, 2008). In these organisms, actuation accomplishes functions such as movement (e.g., medusoid jellyfish (Green, Ben-Nissan, Yoon,

Milthorpe, & Jung, 2016)), predation (e.g., Venus flytrap (Forterre, Skotheim, Dumais, &

Mahadevan, 2005)), predator evasion (e.g., cuttlefish (Rossiter, Yap, & Conn, 2012)), and species propagation (e.g., ice plant seed capsule (Harrington et al., 2011)). This has led to an enriched understanding of the mechanisms underlying actuation in biological systems, as well as inspiration for the creation of synthetic actuators for human-based applications.

1.2: APPLICATIONS FOR ACTUATION

The main goal of investigating biological actuators is to produce ‘smart’ objects: synthetic materials or structures capable of sensing changes in their surrounding environment and adapting accordingly by autonomously modifying their shape to

22 perform a specific function for human-based applications (Fratzl & Barth, 2009; Meng &

Li, 2013; Srinivasan et al., 1991; Vaia & Baur, 2008). This is accomplished by first identifying properties within biological actuators that are responsible for actuation, then further exploring these properties through computational models and trials with proof-of- concept prototypes. Actuation is a highly sought after feature because it lends versatility and shape-changing anonymity to materials and objects. In addition, actuators can be multifunctional, exhibiting a variety of pre-programmed responses to different stimuli

(Dunlop, Weinkamer, & Fratzl, 2011; Ionov, 2014; Meng & Li, 2013; Srinivasan et al.,

1991; Xuanhe Zhao, 2014).

Synthetic actuators are therefore very useful to solving human-based problems and are relevant to a number of sectors. Currently, synthetic actuators are used in medical

(e.g., drug release, artificial muscles, and wound healing) and aerospace technology (e.g., self-deploying satellites, military stealth, and drones) sectors (Di Luca, Mintchev, Heitz,

Noca, & Floreano, 2017; Schleicher, Lienhard, Poppinga, Speck, & Knippers, 2015;

Willie et al., 2010; Xin Zhao et al., 2017). There are also emergent prototypes in textiles

(e.g., colour-changing materials) and architecture (e.g., self-opening windows and structures) sectors (Karshalev et al., 2018; Magna et al., 2013; Milwich, Speck, Speck,

Stegmaier, & Planck, 2006; Reichert, Menges, & Correa, 2015; Sarkar, Bosneaga, &

Auer, 2009; Yuk et al., 2017). Recently, models and prototypes have focused on fluid- driven actuation because fluid as a stimulus provides the following advantages: (1) simplicity (demands less complex building and/or fewer materials to react to the stimulus); (2) generates large deformation with low energy cost; (3) multifunctionality; and (4) economically cost-efficient (Kempaiah & Nie, 2014; Kothera, Woods, Sirohi,

Wereley, & Chen, 2010; S. Li, Vogt, Rus, & Wood, 2017; M. Ma, Guo, Anderson, &

Langer, 2013; Reichert et al., 2015; Yuk et al., 2017).

1.3: FLUID-DRIVEN ACTUATION

Fluid-driven actuation involves the use of a fluid (water, blood, oil, etc.) as a stimulus for prompting and/or driving deformation in an actuator. The primary stimulus in biological systems is water, which imparts natural composite materials and structures

with a variety of functions. This is especially true in plants, in which water possesses

multiple functions, from biochemical reactions, to turgidity, as well as acting as a

mechanism for initiating changes in conformation of vegetative tissue (Dumais &

Forterre, 2012; Fratzl & Barth, 2009; Srinivasan et al., 1991). In the literature, the main

types of fluid-driven actuators that have been created are polymer bilayers, and hydrogels

(C. Lv et al., 2018; Mao et al., 2016; Meng & Li, 2013; Sehaqui, Zhou, & Berglund,

2011).

Polymer bilayers are composed of two materials with different expansion

properties (Ionov, 2011; Stoychev, Zakharchenko, Turcaud, Dunlop, & Ionov, 2012).

These layers can either be overlaid, joined side-by-side, or blended to form a single

homogenous polymer with non-homogenous expansion properties (Kempaiah & Nie,

2014). Depending upon the arrangement of the different materials, various three-

dimensional shapes can be achieved, as well as multi-stage folding so that a single

material can deform into various shapes through pre-determined, step-wise folding that is

dependent on the amount, duration or type of stimulus applied (Ionov, 2011; Kempaiah &

Nie, 2014; Stoychev et al., 2012). Polymer bilayers respond very well to fluid-stimulated

24 deformation; using two materials with differential expansion results in different extents of swelling and shrinking between the layers of the bilayer, which – depending on material arrangement – can lead to an array of deformation modes (Ionov, 2011;

Kempaiah & Nie, 2014; Stoychev et al., 2012). In fact, hydrogels are often water- responsive polymer bilayers (Erb et al., 2013; Ionov, 2014).

Hydrogels are matrices composed of one or polymers capable of swelling and shrinking (sometimes swelling up to or more than ten times their original volume) in response to water absorption/loss (Ionov, 2013, 2014; Koetting, Peters, Steichen, &

Peppas, 2015; Xuanhe Zhao, 2014). Depending on the polymer type, polymers can lend structural support and pre-determine the direction of hydrogel deformation, can provide flexibility to allow for deformation, or can dictate the degree of hydrogel swelling (Ionov,

2013, 2014; Meng & Li, 2013; Xuanhe Zhao, 2014). Hydrogels are unique in that they possess both a solid and liquid component, a combination that imparts this type of composite material with multiple functions, similar to water-driven actuation in plants

(Ionov, 2013; Koetting et al., 2015; Xuanhe Zhao, 2014).

Hydrogels are a particularly advantageous form of actuator, and extensive research has explored how to optimize the extent of movement capable by hydrogels, their energy efficiency (i.e., low energy input, large deformation output), and material combinations for best actuating performance (Koetting et al., 2015). The key features that make hydrogels such attractive actuating devices are: (1) responsiveness to multiple types of stimuli; (2) high load-bearing ability; (3) fracture and fatigue-resistant; (4) versatility in terms of matrix architecture (density, polymer length, and polymer interaction), leading to control over the degree of hydrogel expansion, hydrogel load-bearing ability,

25 and deformation pattern (direction, rate, and range of movement); (5) reversible deformation; (6) biocompatible and biodegradable; and (7) amenable to 3D printing, which makes hydrogels easier to produce at an industrial scale (Bai, Yang, Morelle,

Yang, & Suo, 2018; Dunlop et al., 2011; Ionov, 2013, 2014; Koetting et al., 2015;

Chunxin Ma et al., 2018; Velders, Dijksman, & Saggiomo, 2017; Xuanhe Zhao, 2014;

Zolfagharian et al., 2016; Zolfagharian, Kouzani, Khoo, Nasri-Nasrabadi, & Kaynak,

2017).

Plant cell walls are considered biological hydrogels because of their complex polymer matrices and ability to absorb large amounts of water (Zheng et al., 2018). Thus, the concepts underlying fluid-driven actuation and hydrogel prototyping are primarily derived from studying actuator models in the plant kingdom (Rivka Elbaum & Abraham,

2014; C. Lv et al., 2018; Zheng et al., 2018).

SECTION 2: ACTUATORS IN THE PLANT KINGDOM

As indicated above, many models of actuation have been derived from the plant kingdom (Dunlop et al., 2011; Rivka Elbaum & Abraham, 2014; Forterre, 2013; Zheng et al., 2018). In plants, actuation is primarily water-based, involving either differential hygroscopic swelling and shrinking of tissues or osmotic gradients affecting cell turgor pressure (Rivka Elbaum & Abraham, 2014; Forterre et al., 2005; Kempaiah & Nie,

2014). Implementing hygroscopic actuation (herein referred to as hygroactuation) into synthetic actuators has become increasingly popular for many reasons: (1) water is a readily available resource and is economically inexpensive; (2) it is biocompatible and environmentally sustainable; and (3) its use as a stimulus to elicit a mechanical response is less detrimental to the actuating material than other conventional stimuli used in existing synthetic actuators (e.g., high heat and extreme pH changes) (Fratzl & Barth,

2009; Ionov, 2013; Sehaqui et al., 2011; Xuanhe Zhao, 2014).

As stated above, there are two main types of actuating mechanisms used by plants: differential swelling and shrinking among tissue types, and osmotic gradients. The plant models that use each of these mechanisms are discussed in detail below.

2.1: SWELLING AND SHRINKING

The following are plant models in the literature used to investigate differential hygroscopic tissue swelling and shrinking (Table 1.1). Hygroactuation in wheat awns and pinecones is driven by tissue bilayers, and by cell wall bilayers in ice plant seed capsules, twisting tendrils and vines, and tension wood.

2.1.1: WHEAT AWNS

Wheat uses hygroactuation to facilitate seed dispersal (R. Elbaum, Zaltzman,

Burgert, & Fratzl, 2007). The actuating organ in this plant species is the seed dispersal unit, composed of two awns joined at the base by a seed capsule. In response to daily, cyclic humidity changes, the awn tissue swells and shrinks, creating a swimming motion.

When seed dispersal units drop from the wheat plant, this swimming motion pushes the dispersal unit away from the adult plant, thereby dispersing seeds into the surrounding area (R. Elbaum et al., 2007). Investigation into the hygroactuation properties of the seed dispersal unit showed that movement arises because of differential tissue swelling in the awns. The awn can be divided into two distinct tissue layers: the active layer (i.e., awn ridge), which is responsible for driving movement, and the resistance layer (i.e., awn cap), which is responsible for controlling the direction of movement.

Differential swelling and shrinking in these two tissue types was determined to be a result of changes in cellulose microfibril angle (MFA) in the cell wall (R. Elbaum,

Gorb, & Fratzl, 2008; R. Elbaum et al., 2007). Cellulose is a key polysaccharide in the plant cell wall (both primary and secondary wall types) that provides structural support and shape to the wall (Cosgrove, 2005; Cosgrove & Jarvis, 2012; Höfte & Voxeur, 2017;

McFarlane, Döring, & Persson, 2014). Cellulose microfibrils are stiff bundles of crystalline cellulose chains held together primarily by hydrogen bonds. These microfibrils are deposited in specific orientations within cell wall layers, and act as the primary load-bearing polymer in the plant cell wall (Cosgrove & Jarvis, 2012; McFarlane et al., 2014).

Wide-angle X-ray scattering of wheat samples determined that active and resistance awn tissue layers exhibit different microfibril angles in their cell walls.

Relative to the longitudinal axis of the cell, the overall MFA in the resistance layer was close to 0°, whereas microfibrils in the active layer were oriented in all directions (R.

Elbaum et al., 2008; R. Elbaum et al., 2007). The variation in cellulose MFA gives rise to differential directions of expansion between the two awn tissue layers: cells in the resistance layer are limited in longitudinal expansion and shrinkage, due the parallel alignment of cellulose with the longitudinal axis of the cell, while cells in the active layer expand and shrink in all directions, due to the random MFA orientation.

Interestingly, in-depth analysis (experimental and computational modelling) revealed that the resistance layer of the awn demonstrated a similar cellulose deposition pattern to that found in plywood (R. Elbaum et al., 2008; Zickler et al., 2012). While the overall MFA in resistance tissue was estimated to be 0°, closer examination at the cell wall level showed that cellulose in one lamina (i.e., a single cell wall layer) was deposited at a ninety-degree angle relative to the lamina above and underneath it. This pattern of perpendicular laminar deposition increases plywood stiffness and rigidity, and was proposed to act in the same way in the resistance layer of wheat awns. Thus, while limited longitudinal swelling and shrinking does occur in the resistance layer, its main role is to act as a rigid surface that restricts the swelling of the active layer – and hence awn movement – to a specific direction (described further Figure 1.1).

2.1.2: PINE CONE

Similar to wheat awns, pinecones also depend on conformational changes for seed dispersal, and are responsive to fluctuations in air humidity. The actuating organ in the pinecone is the scale. As pinecones dry, scales open and release seeds, and as air humidity increases, scales absorb moisture and close so that they are pressed against the body of the pinecone (Dawson, Vincent, & Rocca, 1997; Dunlop et al., 2011; Reyssat &

Mahadevan, 2009). Again, pinecones bear semblance to wheat awns in the fact that scales can be considered bilayer structures with active and resistance layers that expand differentially due to changes in cellulose organization between the two tissue layers

(Dawson et al., 1997; Dunlop et al., 2011; Reyssat & Mahadevan, 2009). Like the structure in wheat awns, average cellulose MFA in the resistance layer of pinecone scales is approximately parallel to the longitudinal axis of the cell, while MFA in the active layer is oriented in every direction, leading to the active layer being responsible for movement, and the resistance layer responsible for the direction of movement (Figure

1.1) (Dunlop et al., 2011; Reyssat & Mahadevan, 2009).

The opening and closing motion of pinecones is repeatable over multiple drying/wetting cycles, and conformational change between open and closed positions is consistent and reproducible across cycles (Reyssat & Mahadevan, 2009). As well, the degree of scale curvature and the rate of opening and closing was found to be identical between different sizes and species of pinecone, implying that the mechanical mechanism has been conserved across species and is most likely significant for effective seed dispersal (Reyssat & Mahadevan, 2009).

2.1.3: ICE PLANT SEED CAPSULE

Ice plant is a general name for a variety of species of the Aizoaceae genus from arid and semi-arid environments that have adapted hygroactuation-driven seed dispersal mechanisms (Rivka Elbaum & Abraham, 2014; Harrington et al., 2011). In dry conditions the seed capsule is in a closed conformation, with five protective valves covering five interior seed compartments. Upon rehydration, these valves open, peeling back on themselves by 150° (Harrington et al., 2011). The process of changing from a closed to open conformation requires two coordinated steps, termed flexing and packing

(Harrington et al., 2011). Both processes rely on cell wall swelling and geometric constraints to generate directional movement to open seed compartments. Similar to wheat awns and pinecones, movement is generated by differential swelling and shrinking.

In ice plants, however, this process occurs at the cell level rather than at the tissue level.

(Harrington et al., 2011; Studart, 2015).

Specialized cells along the upper side of the valve, which are collectively referred to as the keel, are responsible for both the flexing and packing movement of the valves.

Valve deformation results from differential swelling among cell wall layers in keel cells.

Keel cell walls resemble those of gelatinous fibers in coiling tendrils/vines and tension wood (described below in 2.1.4 and 2.1.5) in that they possess a hydrophilic wall layer, and a hydrophobic lignin-rich wall layer (Harrington et al., 2011; Studart, 2015). In keel cell walls, a lignin-rich layer surrounds an inner layer primarily composed of cellulose

(Rivka Elbaum & Abraham, 2014; Harrington et al., 2011). Usually in cell walls, pectin and hemicellulose polysaccharides are considered more hydrophilic than cellulose.

However, compared to lignin, cellulose is hydrophilic. Thus, in keel cell walls, the inner

cellulosic layer acts as the active layer, swelling and shrinking in response to water, and

the lignin layer acts as the resistance layer, controlling the extent of swelling that occurs.

Combined with anisotropic cell geometry, the keel cell structure (morphology and

composition) determines the direction of valve opening and closing (Harrington et al.,

2011).

2.1.4: TENDRILS AND TWINING VINES

Many species of tendrils and vines (e.g., Redvine and English Ivy) display unique

twisting and coiling features that allow them to climb and grow to great heights.

Gelatinous fibers (G-fibers) are necessary for the twisting and coiling process. Species

that do not coil do not contain G-fibers (Bowling & Vaughn, 2009; Meloche, Knox, &

Vaughn, 2007), underlining the significant role that this fiber type plays in producing

coiling motions in tendrils and vines. G-fibers are files of long fiber cells that were first

identified in tension wood (Bowling & Vaughn, 2009).G-fibers are unique in their cell

wall architecture; the secondary cell wall is highly lignified (i.e., hydrophobic) and

provides the fiber with structural rigidity and support, while the tertiary cell wall is highly

enriched in pectic polysaccharides, which creates a gelatinous matrix, and is non-lignified

(Bowling & Vaughn, 2009; Meloche et al., 2007). This pectic cell wall layer is highly

hydrophilic, and can absorb large amounts of water (Bowling & Vaughn, 2009). Like ice plant seed capsules, this cell wall bilayer (secondary cell wall = resistance layer, tertiary cell wall = active layer; Figure 1.2) leads to differential expansion in response to water, allowing for controlled and directed vine growth (Bowling & Vaughn, 2009).

Three general vine/tendril coiling shapes result from different patterns of G-fiber

deposition: (1) tendrils that coil in any direction display a hollow cylinder of G-fibers that

surround the centre of the tendril; (2) tendrils that only coil in one direction possess G-

fibers arranged as a curved plate positioned between the epidermis and xylem on the

inner side of the tendril/vine; and (3) tendrils with “sticky” ends that adhere to branches

for climbing contain a band of G- fibers running through the centre of the tissue (adjacent

to the vascular bundle) that, upon sticking to a surface and anchoring the tendril/vine,

causes forces build up in the growing tendril/vine that causes coiling (Bowling &

Vaughn, 2009; Meloche et al., 2007). In addition to the position of G-fiber cells within a

tendril/vine stem, the abundance of G-fibers can affect the degree of coiling observed.

Supercoiled tendrils/vines possess a high abundance of G-fibers relative to other tissue

types present in the plant while loosely coiled tendrils/vines possess a lower abundance

of G-fibers (Bowling & Vaughn, 2009).

2.1.5: TENSION WOOD (ANGIOSPERM TREES)

While movement follows a very long timeline and deformation is only observed

as minor changes in shape, tension wood is a well-studied model for hygroscopic

swelling and shrinking. Similar to twisting tendrils/vines, tension wood employs G-fibers

to direct movement during limb growth in tension wood tree species (Bowling &

Vaughn, 2008; Lafarguette et al., 2004; Pilate et al., 2004). However, at the cell wall

level, the composition of tension wood G-fibers more closely resembles that of ice plant

seed capsule. A tertiary hydrophilic cellulose layer swells and shrinks in response to

water while a surrounding lignified, hydrophobic layer limits the extent of tertiary layer

33 swelling (Bowling & Vaughn, 2008; Pilate et al., 2004). The differential swelling between the lignified secondary cell wall layer and cellulosic tertiary cell wall layer generates tensile forces in tree limbs that work to keep the limb horizontal (counteract gravity), or bend it upward (a response to environmental factors that stimulate limb growth) (Figure 1.2) (Bowling & Vaughn, 2008; Lafarguette et al., 2004).

2.2: OSMOTIC GRADIENTS

Plant species using osmotic gradients to change conformation often display rapid movements (Burgert & Fratzl, 2009a; Ionov, 2013; Zheng et al., 2018). Below are two well-studied species that take advantage of osmotic gradients for predation (i.e., Venus flytrap) and predator evasion (i.e., Touch-Me-Not) (Table 1.1).

2.2.1: VENUS FLYTRAP

Unlike the plant species described above where water stimulates shape change, the stimulus for Venus flytrap movement is mechanosensory. However, the resulting deformation undertaken by the plant is dependent upon hygroactuation mechanisms. The leaf surfaces of the Venus flytrap are covered in small hairs (mechanosensors) that are sensitive to touch (Braam, 2005; Forterre et al., 2005; Volkov, Adesina, Markin, &

Jovanov, 2008). When triggered, these mechanosensors initiate osmotic gradients and drastic changes in turgor pressure (caused by water movement in and out of vacuoles) that result in open (convex shape) leaves snapping tightly shut (concave shape) against each other (Braam, 2005; Forterre et al., 2005; Hodick & Sievers, 1989; Volkov et al.,

2008). The exact pathways and mechanisms involved in the deformation process have not

yet been fully elucidated (Forterre et al., 2005; Volkov et al., 2008), but some studies

have highlighted critical components involved in leaf movement (outlined below and in

Figure 1.3).

The rapid snapping movement observed in Venus flytraps is caused by releasing

stored elastic energy within the cells of the leaf surface, which itself is the result of cell

shape (determined by cellulose MFA) and level of turgor pushing on the cell wall

(Forterre et al., 2005; Hodick & Sievers, 1989). When triggered (e.g., by an insect

landing on the interior surface of the leaf), a chemical cascade causes cells of the interior

leaf surface (high turgor, large volumes of water in the vacuole) to rapidly lose turgor

(water is channelled to vacuoles in cells on the exterior leaf surface through water

channels called aquaporins). This leads to changes in turgor pressure in both leaf

surfaces, which releases stored energy in the leaves that then drives convex-to-concave

leaf deformation (Forterre et al., 2005; Volkov et al., 2008).

2.2.2: TOUCH-ME-NOT

Like the Venus flytrap, Touch-Me-Not plants (e.g., Mimosa pudica) rapidly fold

their leaves in response to mechanosensory stimulation to small touch-sensitive hairs on

their leaf surfaces (Allen, 1969; Braam, 2005; Burgert & Fratzl, 2009a; Volkov, Foster,

Ashby, et al., 2010). The underlying mechanism for movement is also similar, but Touch-

Me-Not plants do no store elastic energy in their leaves like Venus Fly traps. Instead,

they rely uniquely on osmotic gradients and turgor pressure to change shape. In Touch-

Me-Not plant species, flexor and extensor cell types in the pulvinus, a hinge-like structure at the base of the leaf that is connected to the stem, are responsible for leaf

movement (Allen, 1969; Braam, 2005; Temmei et al., 2005; Volkov, Foster, Ashby, et

al., 2010). Three types of pulvini exist: primary, secondary, and tertiary, which

respectively move the petiole, pinna and leaflets (Volkov, Foster, & Markin, 2010). Upon stimulation, extensor cells (underside of leaf) lose turgor pressure (water movement out of the vacuole) and – through aquaporin channelling – flexor cells (upper side) gain turgor (water movement into the vacuole). This change in pressure between the two sides of the pulvinus causes leaves to droop toward the stem. Upon removal of stimulation, turgor is gradually re-established in extensor cells (i.e., water movement from flexor to extensor cells) and leaves resume a normal, open position (Allen, 1969; Burgert & Fratzl,

2009a; Temmei et al., 2005; Zheng et al., 2018).

2.3: PLANT-INSPIRED ACTUATOR PROTOTYPES

Below are a few examples of engineering prototypes that were created to mimic the hygroresponsive properties of pinecone, wheat awn, Venus Flytrap, and Touch-Me-

Not plant models.

2.3.1: WHEAT AWN AND PINE CONE-INSPIRED

Based on the tissue bilayering observed in pinecones, Reyssat & Mahadevan 2009 developed a plastic and paper bilayer prototype that curls as a result of differential expansion/contraction in response to water. In this prototype, the paper acts as the active layer, and the plastic as the passive layer. Reyssat & Mahadevan 2009 were able to demonstrate that their prototype mimicked reversible pinecone scale movement and that

36 the prototype was functional over repeated cycles of wetting and drying (Reyssat &

Mahadevan, 2009).

Wang et al. 2015 produced a small-scale film bilayer based on differences in cellulose microfibril angles in wheat awn and pinecone tissue bilayers. Unlike the bilayer created by Reyssat & Mahadevan 2009, the prototype produced by Wang et al. 2015 is responsive to changes in air humidity and also bends at much faster rates (M. Wang,

Tian, Ras, & Ikkala, 2015).

Like Wang et al. 2015, Reichert et al. 2015 also produced a humidity-responsive prototype based on differential expansion of thin material layers, but they did so on a much larger scale. Using layers of plywood stacked in different grain (i.e., different cellulose MFA) directions, they were able to create large structures (metre length scale) that deformed in response to air humidity. Because of its large length scale, the prototype shows promise for further construction of ‘smart’ buildings (i.e., those that can adapt and respond to the environment) (Reichert et al., 2015).

Velders et al. 2017 developed a bilayer actuator based on an existing product: hydrogel beads available for commercial use. Using the hydrogel beads as the active layer, Velders et al. 2017 glued them to a restrictive polymer layer. Alone, the hydrogel beads swell in every direction. However, when glued to the restrictive layer, swelling was constrained to a specific direction, just as movement is in wheat awns and pinecones. As well, by gluing beads at different distances to each other (e.g., close together or far apart),

Velders et al. 2017 could modify the degree of bending of their prototype (beads closer together curled more tightly, while beads farther apart curled less). This prototype offers

37 a cheap alternative to produce hydroactuators with complex deformation (depending on bead size and bead clustering) (Velders et al., 2017).

2.3.2: VENUS FLYTRAP AND TOUCH-ME-NOT-INSPIRED

Sinibaldi et al. 2014 created a fluid-driven prototype based on the osmotic gradients found in Touch-Me-Nots and Venus Flytraps. Their prototype was composed of two metal compartments separated by an osmotic membrane. Upon stimulation with electricity, an osmotic solution flowed from one compartment to the other. An expandable elastic disc was fitted onto the second compartment. Because of the unidirectional flow of the osmotic fluid through the membrane, the pressure in the second compartment forced the elastic disc to bulge. While they did not generate a reversible actuator, Sinibaldi et al. 2014 created an actuator with high energy efficiency, high mechanical force response and applicability to biorobotics (Sinibaldi, Argiolas, Puleo, &

Mazzolai, 2014).

Zheng et al. 2018 developed a bilayer hydrogel actuator based on the osmotic gradients observed in Mimosa pudica plants. What makes their actuator prototype interesting is that it can deform and function in open-air environments. One of the previous limiting factors of hydrogels is that they only function when immersed in water.

By containing the water within the hydrogel and using self-circulating osmotic gradients between layers in the hydrogel, Zheng et al. 2018 created a prototype that can function in multiple environments (aqueous and non-aqueous). Thus, while this hydrogel does not exchange water with the environment (i.e., no overall volume gain/loss), it bypasses the need for hydrogels to remain immersed in an aqueous environment to function, which

38 expands their utility and the types of applications for which they can be used (Zheng et al., 2018).

2.4: LIMITATIONS OF CURRENT FLUID-DRIVEN HYDROACTUATORS

There are currently synthetic hydroactuators used in various sectors (e.g., medical and textile) (S. Li et al., 2017). However, there are limitations to their actuating ability.

Most noticeably, the majority of existing hydroactuators (especially hydrogels) are based upon large-scale observations (e.g., bending of plant organs) and are fabricated to only respond to fluid stimuli at this level (Dunlop et al., 2011; C. Lv et al., 2018; Meng & Li,

2013). This is limiting in two senses: (1) it restricts the materials that can be used to create hydroactuators to those that react at larger scales; and (2) it also restricts the range of motions and conformational changes possible for synthetic actuators (i.e., less complexity and more brute motion) (C. Lv et al., 2018; Meng & Li, 2013; Zheng et al.,

2018). In addition to these limitations, other concerns have arisen from observing the use of synthetic actuators over time. The two main limitations are outlined below:

2.4.1: AVAILABILITY, TYPE, AND STRENGTH OF MATERIALS FOR

HYDROACTUATORS

The current types of materials available for producing synthetic composites are limited, both in terms of their compatibility to act as actuating materials, as well as the cost of using certain materials for industrial-scale applications (C. Lv et al., 2018;

Sehaqui et al., 2011). In addition, efforts are being made to choose materials that are biocompatible (e.g., for use in medical sectors) and ecologically safe (Fratzl & Barth,

2009; Zolfagharian et al., 2017). Future hygroactuating devices should be easily degradable or reusable, and any by-products they produce should not be toxic (Fratzl &

Barth, 2009; C. Lv et al., 2018; Zolfagharian et al., 2017). Also, the strength of these materials must be taken into consideration; many current actuators demonstrate weakness and structural failure at smaller length scales (nanometre to millimetre), since they were originally modelled from large-scale observations (Ionov, 2011; Kempaiah & Nie, 2014;

Sehaqui et al., 2011).

2.4.2: HYDROACTUATOR LONGEVITY AND CONFORMATIONAL

REVERSIBILITY

Two very important, and interconnected, attributes to be considered when designing actuators are conformational reversibility and the longevity of the structure/material (i.e., how stable it is over a set period of use and/or multiple cycles of stimulation and deformation) (Bai et al., 2018). These are two challenges that have yet to be addressed in the literature, apart from highlighting that they are limiting factors that need to be overcome to improve existing hydroactuators (Ionov, 2011; Xuanhe Zhao,

2014). While a structure or material may exhibit large-scale reversibility when transitioning between two conformational shapes, it does not imply that irreversible deformation has not occurred at smaller length scales within the structure/material. This can lead to limitations in the number of cycles over which a material/structure can deform before macroscopic damage becomes apparent and permanent deformation occurs.

As described in the literature, what is needed are synthetic materials that: (1) reversibly change shape; (2) carry out complex conformational alterations; (3) exhibit

conformation changes at multiple length scales; and (4) possess high load-bearing ability

(Erb et al., 2013; Sehaqui et al., 2011).

2.5: A POTENTIAL NEW MODEL FOR STUDYING HYGROACTUATION

Currently plant models of hygroactuation fail to meet the four traits mentioned

above. Wheat awns and pinecones rely on tissue bilayers, and tension wood, ice plant

seed capsules and coiling tendrils/vines rely on cell wall bilayers, all of which generate

more or less simplistic deformation. As well, the changes in shape of tension wood and

coiling tendrils/vines are irreversible. Ideally, a model plant system displaying at least three, it not all four, of the desired characteristics for synthetic actuators is required for investigation.

It has recently come to light that resurrection plants could be models for hygroactuation. Resurrection plants reversibly deform, display – depending on the species – simple to complex modes of deformation, and also exhibit extensive folding at multiple length scales (organ, cell and cell wall levels) (Dunlop et al., 2011; Fratzl &

Barth, 2009; Rascio & Rocca, 2005).

SECTION 3: RESURRECTION PLANTS AS COMPLEX ACTUATORS

Below is an overview of resurrection plants (Section 3.1), including the molecular

(Section 3.2) and physiological (Section 3.3) mechanisms they employ to survive desiccation tolerance, as well as an in depth description of the two resurrection plants investigated as part of this research project (Section 3.4 and 3.5)

3.1: RESURRECTION PLANTS

The term resurrection plant includes over 350 species ranging from simple plants such as algae, bryophytes and mosses, to more complex plant types such as ferns and angiosperms (Alpert, Bone, & Holzapfel, 2000; J. M. Farrant & Moore, 2011; Morse,

Rafudeen, & Farrant, 2011). Resurrection plants are found in a diverse array of habitats

(e.g., arctic, arid and temperate regions) (Alpert et al., 2000; Oliver, Tuba, & Mishler,

2000; Rascio & Rocca, 2005). The unifying feature of resurrection plant species is that they are exposed to extreme environmental conditions (e.g. severe temperatures, drought and wind), and have adapted responses to survive prolonged durations in these conditions

(Leprince & Buitink, 2010; Rascio & Rocca, 2005; Scott, 2000). Over the past few decades, resurrection plants have served as models to investigate various biological questions (Table 1.2), spanning from the analysis of biospheres and ecological niches to the advancement and improvement of both crop engineering and biomechanical engineering (Leprince & Buitink, 2010; Morse et al., 2011; Toldi, Tuba, & Scott, 2009;

Vicre, Lerouxel, Farrant, Lerouge, & Driouich, 2004).

Resurrection plants are desiccation tolerant, and are able to withstand losing large volumes of relative water content from their vegetative tissue (Moore, Vicre-Gibouin,

Farrant, & Driouich, 2008; Morse et al., 2011; Scott, 2000; Toldi et al., 2009). As well, they are able to remain in a desiccated state for prolonged periods of time without impeding their survival (Moore, Vicre-Gibouin, et al., 2008; Rascio & Rocca, 2005).

When water once again becomes available in the surrounding environment, resurrection plants are able to rehydrate with minimal tissue damage and can continue normal biological processes such as growth and reproduction without any functional impairment

(J. M. Farrant & Moore, 2011; Morse et al., 2011; Scott, 2000). Survival during prolonged drought is dependent upon three main factors: (1) minimizing the amount of damage incurred during dehydration of vegetative tissue; (2) maintaining cellular stability

(structure and metabolism) during periods of desiccation; and (3) possessing the ability to quickly repair desiccation-induced damage during the rehydration process (Alpert et al.,

2000; J. M. Farrant & Moore, 2011; Oliver et al., 2000; Rascio & Rocca, 2005).

It is important to distinguish desiccation tolerance from drought tolerance.

Drought tolerance refers to plants that are able to withstand mild water-deficit stress (i.e., losing ≤20% relative water content) (Alpert et al., 2000; Moore, Vicre-Gibouin, et al.,

2008; Oliver et al., 2000; Rascio & Rocca, 2005). However, at lower relative water contents (>20% water loss), these plants do not possess mechanisms to cope with the biochemical and physiological stresses that accompany such high levels of water loss and their viability is compromised (Oliver et al., 2000; Rascio & Rocca, 2005; Toldi et al.,

2009). In addition, drought tolerant species cannot survive prolonged periods of mild water deficit without experiencing irreparable tissue damage, which ultimately leads to

plant death (Moore, Vicre-Gibouin, et al., 2008; Oliver et al., 2000; Rascio & Rocca,

2005; Toldi et al., 2009). In contrast, desiccation tolerant plants are able to withstand

losing up to 95% of their relative water content without impairing their viability (Oliver

et al., 2000; Rascio & Rocca, 2005; Toldi et al., 2009). Desiccation tolerant plants

employ various mechanisms to prevent and/or repair damage incurred during dehydration

and subsequent rehydration of their vegetative tissue (Morse et al., 2011; Toldi et al.,

2009). These mechanisms are described in detail below in Sections 3.2 and 3.3.

There are two main categories of resurrection plant species: (1) fully desiccation

tolerant; and (2) modified desiccation tolerant (Alpert et al., 2000; Moore, Vicre-Gibouin,

et al., 2008; Oliver et al., 2000; Rascio & Rocca, 2005). Full desiccation tolerance

represents a very primitive method of surviving prolonged drought conditions and is

considered to be true desiccation tolerance (J. M. Farrant & Moore, 2011; Oliver et al.,

2000; Rascio & Rocca, 2005). The first terrestrial plants to invade land were fully

desiccation tolerant, and today only very primitive species (e.g., bryophytes, algae, and

lichen) display this type of tolerance (J. M. Farrant & Moore, 2011; Oliver et al., 2000).

Rather than rely upon preventative mechanisms to avoid tissue damage, fully desiccation tolerant species instead depend upon a wide array of cellular mechanisms to repair damage incurred during tissue drying and rehydration (Oliver et al., 2000; Rascio &

Rocca, 2005). These mechanisms are constitutive, meaning that fully desiccation tolerant

species are primed for drought, even when the plant is fully hydrated, which allows these

species to dry out very rapidly, on the scale of minutes to hours (Morse et al., 2011;

Rascio & Rocca, 2005). In contrast, modified desiccation tolerance represents tolerance

to drought that has re-evolved within genera over evolutionary time (Alpert et al., 2000;

Rascio & Rocca, 2005). This type of tolerance is most commonly found in higher genera, such as ferns and angiosperms (Alpert et al., 2000; Oliver et al., 2000). Due to the complexity of their vegetative tissues, modified desiccation tolerant species employ protective mechanisms that prevent tissue/cellular damage during the process of dehydration and rehydration (J. M. Farrant & Moore, 2011; Rascio & Rocca, 2005).

These mechanisms are not constitutive and are activated by water loss (Oliver et al.,

2000). Since these mechanisms take time to establish themselves in the plant, the vegetative tissue of modified resurrection species cannot dry out rapidly; if it does, the plant will not survive desiccation (Oliver et al., 2000; Rascio & Rocca, 2005). Therefore, these types of species dehydrate much more slowly than fully desiccation tolerant species, with dehydration time ranging from hours to days (J. M. Farrant & Moore, 2011;

Rascio & Rocca, 2005).

Resurrection plants must contend with both mechanical and biochemical stresses that arise during dehydration and rehydration of their vegetative tissues; therefore, resurrection plants have adapted physiological and molecular protective/repair mechanisms (Dinakar, Djilianov, & Bartels, 2012; Morse et al., 2011; Rascio & Rocca,

2005). The most common types of desiccation-induced stresses experienced by resurrection plants and their subsequent responses are discussed below.

3.2: MOLECULAR RESPONSES TO DESICCATION

There are two main biochemical stresses arising from vegetative tissue

dehydration: the effects of water loss on the cell, and accumulation of toxic levels of

reactive oxygen species (ROS).

3.2.1: WATER LOSS STRESS

As cells dehydrate cell volume shrinks, bringing intracellular compartments and

macromolecules that normally do not interact into closer proximity (Moore, Vicre-

Gibouin, et al., 2008; Vicre et al., 2004). As water is stripped away from organelles and

macromolecules, they are left with open sites for hydrogen bonding, which energetically

destabilizes them (Moore, Vicre-Gibouin, et al., 2008; Toldi et al., 2009). To fulfill these

missing hydrogen bonds, macromolecules (e.g., proteins) and organelles (e.g.,

membranes) will hydrogen bond with any other nearby macromolecule or organelle

(Moore, Vicre-Gibouin, et al., 2008; Toldi et al., 2009). This leads to unfavourable

interactions such as membrane fusion, protein aggregation and protein denaturation, all of

which are irreversible processes that can result in cell death (Moore, Vicre-Gibouin, et

al., 2008; Toldi et al., 2009; Vicre et al., 2004).

3.2.2: WATER LOSS STRESS RESPONSES

1) Preferential hydration: to prevent energetically unfavourable interactions as cells lose cytoplasmic volume, remaining water is diverted toward maintaining the hydrogen bonds of macromolecules and organelles that should not come into contact with each other under normal, hydrated conditions (Morse et al., 2011; Vicre et al., 2004).

Thin shells of water surround macromolecules and organelles, acting as shields to prevent unfavourable hydrogen bonding, which maintains intracellular stability (Morse et al.,

2011; Vicre et al., 2004).

2) Osmoprotectants: at very low water contents (less than 5% relative water content), preferential hydration cannot be maintained as there is too little water in the cell to create water shells around macromolecules and organelles (Alpert et al., 2000; Toldi et al., 2009; Vicre et al., 2004). In place of water, osmoprotectants are used to protect macromolecules and organelles from unfavourable interactions (Moore, Vicre-Gibouin, et al., 2008; Vicre et al., 2004). Osmoprotectants are non-aqueous solvents or solutes that form hydrogen bonds (J. M. Farrant & Moore, 2011; Scott, 2000; Toldi et al., 2009;

Vicre et al., 2004). The most common osmoprotectant found in all researched resurrection plants is sugar, and in particular, sucrose and trehalose (J. M. Farrant &

Moore, 2011; Morse et al., 2011; Vicre et al., 2004). In addition, certain proteins also function as osmoprotectants (Leprince & Buitink, 2010; Scott, 2000). These include late embryogenesis abundant and small heat shock proteins, both of which have increased abundance in most tested species of resurrection plants undergoing dehydration (Dinakar et al., 2012; J. M. Farrant & Moore, 2011; Morse et al., 2011).

3) Vitrification: another mechanism employed by resurrection plants to prevent unfavourable interactions between intracellular components is to vitrify the cytoplasm

(Toldi et al., 2009). Essentially, at very low relative water content, the cytoplasm thickens, “gluing” components in place within a glass-like solvent, thus preventing movement and trafficking of molecules within the cell (Toldi et al., 2009). Sucrose acts as a medium for crystallization of the cytosol, while late embryogenesis abundant and

small heat shock proteins facilitate the process of vitrification, though their exact role in

this process is still unclear (J. M. Farrant & Moore, 2011).

3.2.3: ACCUMULATION OF TOXIC REACTIVE OXYGEN SPECIES

Under desiccation conditions, the metabolism of resurrection plants is altered. As

a result, processes such as respiration and photosynthesis are affected due to the changes

in metabolite production and processing (Morse et al., 2011; Rascio & Rocca, 2005). As

dehydration continues, respiration and photosynthesis become impaired, and in the

absence of their regular activity, reactive oxygen species (ROS) are not processed as

quickly as normal and can accumulate to toxic levels in the cell (Dinakar et al., 2012;

Vicre et al., 2004). This results in photo-oxidative damage, such as DNA breakage,

membrane lipid oxidation and protein oxidation, leading to impairment of cellular

function and subsequent cellular death (Dinakar et al., 2012; Morse et al., 2011; Vicre et

al., 2004).

3.2.4: TOXIC REACTIVE OXYGEN SPECIES STRESS RESPONSE

The main sources of ROS production in the plant cell are chloroplasts,

mitochondria and peroxisomes (Dinakar et al., 2012). Resurrection plants have adapted

multiple mechanisms to minimize ROS production and damage during dehydration of

their vegetative tissue (Rascio & Rocca, 2005; Vicre et al., 2004).

1) Downregulation of photosynthesis and mitochondrial respiration: the metabolism of resurrection plants is altered during desiccation, resulting in limited

processing of photosynthetic and respiratory metabolites and their by-products (Morse et

al., 2011; Toldi et al., 2009). In response, resurrection plants slowly downregulate

photosynthetic activity to reduce the build up of unprocessed metabolites that would

otherwise lead to ROS accumulation (Toldi et al., 2009). Respiration is also limited to

prevent mitochondrial ROS accumulation. Downregulation of photosynthesis and

respiration occurs very soon after plants start to lose water, beginning – in most species

of resurrection plants – around 20% relative water content loss (Morse et al., 2011).

2) Poikilochlorophylly and homoiochlorophylly: ROS can be produced in chloroplasts that are overexcited by bright light. Resurrection plants have overcome this by adapting one of two very different mechanisms: (1) poikilochlorophylly, or (2) homoiochlorophylly (Morse et al., 2011; Rascio & Rocca, 2005; Scott, 2000). The former is commonly observed in monocot resurrection plant species (e.g., grasses such as

Xerophyta viscosa) and involves complete dismantling of chloroplasts (i.e., thylakoid membranes and chlorophyll are degraded) (Dinakar & Bartels, 2013; Dinakar et al.,

2012; Morse et al., 2011).

Poikilochlorophylly is an effective manner of evading ROS production since it essentially abolishes the ROS producing centres in chloroplasts (thylakoid membranes and chlorophyll) (Morse et al., 2011; Rascio & Rocca, 2005; Scott, 2000). However, it has one key drawback; since the chlorophyll and thylakoid membranes are entirely absent in desiccated plants, upon rehydration, these components must be re-synthesized (Rascio

& Rocca, 2005; Vicre et al., 2004). This takes time and it has been well documented that poikilochlorophyllous resurrection plants do not regain photosynthesis as quickly as

49 homoiochlorophyllous species (Dinakar et al., 2012; Kranner, Beckett, Wornik, Zorn, &

Pfeifhofer, 2002; Vicre et al., 2004).

In contrast, homoiochlorophyllous resurrection species retain their chloroplast

(i.e., intact thylakoid membranes and chlorophyll) throughout the process of dehydration, desiccation and subsequent rehydration (Dinakar & Bartels, 2013; Dinakar et al., 2012;

Morse et al., 2011). Homoiochlorophyllous resurrection plants are more prone to ROS damage than poikilochlorophyllous plants. However, they recover photosynthesis more rapidly upon rehydration of their vegetative tissue (Dinakar et al., 2012; Rascio & Rocca,

2005). Since the chloroplasts of homoiochlorophyllous species remain intact during desiccation, resurrection plants must employ protective measures to prevent the build up of toxic ROS within these organelles (Kranner et al., 2002). Thus, most species accumulate anti-ROS pigments in their chloroplasts, including anthocyanins and carotenoids that quench excessive ROS before they accumulate and cause damage

(Dinakar et al., 2012).

3) Antioxidants: antioxidant enzymes (e.g., super oxide dismutase) and macromolecules (e.g., vitamins and polyphenols) are found throughout the plant kingdom and are not unique to resurrection plants (Morse et al., 2011). However, in response to dehydration, resurrection plants (both poikilochlorophyllous and homoiochlorophyllous) upregulate the normal activity of antioxidant enzymes and accumulate higher levels of antioxidant pigments (Alpert et al., 2000; Morse et al., 2011; Vicre et al., 2004). This effectively acts as ‘sunscreen’ that quenches ROS species to prevent photo-irradiation damage.

3.3: PHYSIOLOGICAL RESPONSES TO DESICCATION

Resurrection plants are unique in that they exhibit reversible conformational changes in response to dehydration and rehydration. Most notably, deformation is hierarchical, occurring at multiple length scales (metre-to-nanometre) (Morse et al., 2011;

Vicre et al., 2004).

3.3.1: EXPOSED PHOTOSYNTHETIC SURFACES AND ORGAN-LEVEL

DEFORMATION

In hydrated conditions, the vegetative tissue of resurrection plants is fully exposed to the surrounding environment, including photosynthetic surfaces (Alpert et al., 2000).

During dehydration and subsequent downregulation of photosynthesis, having exposed leaf surfaces increases the risk of photoirradiation damage since plants cannot process light energy as efficiently as they do when they are fully hydrated (Lebkuecher &

Eickmeier, 1991, 1993; Sherwin & Farrant, 1996). This is especially problematic for resurrection plant species in arid habitats with little to no shade (Alpert et al., 2000;

Oliver et al., 2000). To circumvent photo-irradiation damage to photosynthesizing tissues, resurrection plants fold their vegetative tissue to reduce the surface area exposed to the surrounding environment. Folding is primarily seen in modified desiccation tolerant species, such as ferns and angiosperms (J. M. Farrant & Moore, 2011); however, certain less-complex species such as Selaginella lepidophylla also undergo vegetative tissue folding in response to dehydration (Lebkuecher & Eickmeier, 1991, 1993). Various genera of resurrection plants display different folding techniques such as two- or three-

dimensional trajectory curling, or fan-like folding (J. M. Farrant & Moore, 2011).

3.3.2: TURGOR LOSS, CELL COLLAPSE AND CELL-LEVEL

DEFORMATION

As mentioned in Section 3.1.1, organelles and compartments are brought into

close proximity upon cytoplasmic volume loss during dehydration. This reduction in size,

if not tightly controlled, can lead to negative turgor pressure, cell collapse and subsequent

cell death (Vicre et al., 2004). Since water loss cannot be prevented in drought

conditions, resurrection plants employ two strategies at the cell level to prevent

irreversible cell collapse:

1) Vitrification: not only does vitrification prevent unfavourable macromolecule

and organelle interaction, but the crystalline sucrose-rich structure also provides

mechanical stability that helps prevent the cell from collapsing upon itself at very low

relative water contents (J. M. Farrant & Moore, 2011; Scott, 2000).

2) Vacuole storage/vacuole fragmentation: another method used by resurrection plants that helps to stabilize the cell is to fill the vacuole with osmoprotectants (Morse et al., 2011; Rascio & Rocca, 2005; Vicre et al., 2004). Osmoprotectants provide stability and prevent the vacuole from collapsing. Since the majority of plant cell volume is taken up by the vacuole, this is a critical mechanical stabilizer in the cell (Rascio & Rocca,

2005; Vicre et al., 2004). In addition, certain species of resurrection plants fragment their main vacuole into smaller vacuoles that can be dispersed about the cytosol, providing further mechanical support to the cellular structure (Rascio & Rocca, 2005).

3.3.3: PLASMALEMMA TEARING AND CELL WALL-LEVEL

DEFORMATION

As cell volume decreases, so does turgor pressure, reducing tensile strain on the

cell membrane and possibly even generating compressive strain (J. M. Farrant & Moore,

2011; Rascio & Rocca, 2005). In desiccation-sensitive plants, loss of turgor results in

tearing of the plasmalemma away from the cell wall, causing electrolyte leakage and

compromising the viability of the cell (Morse et al., 2011; Vicre et al., 2004). This also

holds true for the reverse process; rapid water uptake during rehydration can cause

plasmalemma rupture and electrolyte leakage (J. M. Farrant & Moore, 2011; Morse et al.,

2011; Vicre et al., 2004). To prevent plasmalemma rupture, resurrection plants reversibly

fold vegetative tissue cell walls, and some species also modify cell wall composition to

increase folding capability:

1) Cell wall folding/unfolding: these processes involve modifying the cell wall so that it is able to extensively (un)fold while maintaining its connection to the plasmalemma as cell volume (increases)decreases (Rascio & Rocca, 2005; Vicre et al.,

2004). Plant cells walls are composed of three main polysaccharides: (1) cellulose, which provides structural support and rigidity to the wall; (2) hemicellulose, which cross-links polysaccharides (Barbacci, Lahaye, & Magnenet, 2013; Cosgrove, 2005); and (3) pectin, which acts as a glue surrounding the other polysaccharides and polymers in the cell wall

(Barbacci et al., 2013; Cosgrove, 2005; Mikkonen, 2013). Cellulose microfibrils, such as those discussed in wheat awns and pinecones (Sections 2.1.1 and 2.1.2) align and form laminar sheets within the cell wall (Barbacci et al., 2013). Between these cellulose sheets is sandwiched a pectinaceous gel matrix embedded with hemicellulose polysaccharides,

that latter of which link laminar layers together (Barbacci et al., 2013; Cosgrove, 2005).

The pectin matrix is hydrophilic; when water is abundant, pectin polysaccharides bind

water, causing the matrix to swell, spreading out the laminar cellulosic layers (Barbacci

et al., 2013; Cosgrove, 2005). In resurrection plants, as they dehydrate, water is lost from

the pectin matrix, allowing components such as polysaccharide side chains and

hemicellulose backbones to compact and fold down, bringing the cellulose laminar sheets

closer together, thereby compacting the cell wall as a unit (Moore, Farrant, & Driouich,

2008). Depending on cellulose microfibril angle (MFA), both among individual cell wall

layers and average MFA for the cell wall as a whole, cell walls will fold to different

extents and in different directions (Erb et al., 2013; Moore, Farrant, et al., 2008; Rascio &

Rocca, 2005).

2) Cell wall compositional modifications: in combination with the intrinsic folding properties of their cell walls, some resurrection plants, such as Myrothamnus flabellifolius, also modify the composition of their cell wall in response to desiccation to make the wall more flexible and plastic, allowing for more extensive folding (Moore,

Farrant, et al., 2008; Moore et al., 2013; Morse et al., 2011; Vicre et al., 2004). These modifications also ensure that polysaccharide side chains that are brought into close proximity during folding do not accidentally cross-link and permanently alter the conformation of the cell wall (Moore, Farrant, et al., 2008; Rascio & Rocca, 2005; Vicre et al., 2004). In resurrection plants, a number of different modifications occur, such as:

(1) loosening cell wall components (accomplished by proteins such as expansins and dehydrins) to allow components to slide across each other for tighter folding; (2) changes in calcium content to prevent undesirable polysaccharide side chain cross-linking; and (3)

54 increasing the abundance of hemicellulose and pectinaceous polysaccharides to create a more pliable cell wall (Moore, Farrant, et al., 2008; Moore et al., 2013; Morse et al.,

2011; Rascio & Rocca, 2005; Vicre et al., 2004).

3.4: RESURRECTION PLANTS INVESTIGATED IN THIS RESEARCH PROJECT

Two resurrection plants displaying different modes of deformation were investigated in this thesis: Selaginella lepidophylla and Myrothamnus flabellifolius.

Below are descriptions of these two resurrection plant species. Since the focus of this project is to investigate the physiological mechanisms responsible for vegetative tissue deformation, the mechanisms that S. lepidophylla and M. flabellifolius employ to prevent/repair ROS-induced damage, as well as the biochemical pathways leading to vitrification and preferential hydration will not be discussed.

3.4.1: SELAGINELLA LEPIDOPHYLLA

Selaginellaceae is a family of approximately 700 spikemoss species that belong to the class of Lycopsida (i.e., lycophytes) (Banks, 2009; Korall & Kenrick, 2002;

Lebkuecher & Eickmeier, 1991, 1993). Lycophytes and euphyllophytes represent two divergent vascular lineages that arose approximately 400 million years ago (Banks, 2009;

Weng et al., 2010). Euphyllophytes gave rise to modern day, seed-bearing vascular plants, while lycophytes gave rise to more primitive vascular plant genera, including

Selaginella (Banks, 2009). Lycophytes reproduce via sporangia, which in the case of most Selaginella, are found on the adaxial (upper) surface of the microphylls (simple leaves) (Banks, 2009). Lycophytes also differ from euphyllophytes in terms of their

55 vasculature; microphylls possess a single, non-branching vascular bundle, whereas euphyllophytes possess branching vasculature within their leaves (Banks, 2009; Soni et al., 2012; Weng et al., 2010).

Selaginella inhabit a broad spectrum of environments, including tropical, temperate, arctic and arid regions (Banks, 2009; A. Yobi et al., 2012). It is believed that at least three independent evolutionary events led to desiccation tolerance within the

Selaginellaceae family, producing approximately ten extant desiccation tolerant species and many other species demonstrating drought tolerance (Dinakar et al., 2012; Pandey et al., 2010). Among these, S. lepidophylla is the most widely studied desiccation tolerant species of the Selaginellaceae family. Various studies have analyzed its biochemical and molecular behaviour in response to desiccation, and to some extent the physiological mechanisms required to survive desiccation and subsequent rehydration (Table 1.2).

S. lepidophylla is native to the Chihuahuan desert, a region that spans between both the United States and northern Mexico (Bergtrom, Schaller, & Eickmeier, 1982;

Brighigna, Bennici, Tani, & Tani, 2002). The climate in this area is very arid, receiving less than 25cm of rain during the year. The majority of this rainfall occurs in June and

July, when the Chihuahuan desert experiences humid, monsoon-like conditions. S. lepidophylla experiences its highest activity levels during this time, when there is water available for growth and reproduction (Lüttge, Beck, & Bartels, 2011). In a completely hydrated state, S. lepidophylla appears as a flat rosette of stems arranged in a spiral phyllotaxy, with the youngest branches positioned near the center of the rosette

(Brighigna et al., 2002; Lebkuecher & Eickmeier, 1991, 1993). Upon dehydration, the branches of S. lepidophylla curl into a tight ball that varies in diameter (6-8cm), with

56 older, mature stems curling over younger inner stems (Brighigna et al., 2002; Korall &

Kenrick, 2002; Lebkuecher & Eickmeier, 1991, 1993). Under laboratory conditions, it takes approximately 24 hours for S. lepidophylla to fully hydrate, while it takes approximately 48 hours for plants to air-dry to a fully dehydrated state (i.e., ~5% relative water content) (Brighigna et al., 2002; A. Yobi et al., 2013).

Due to its unique evolutionary position as a stepping-stone species between fully desiccation tolerant (e.g., bryophyte) and modified desiccation tolerant (e.g., angiosperm) species, S. lepidophylla relies upon a combination of both constitutive and inducible mechanisms (A. Yobi et al., 2013). This involves cross talk between molecular/biochemical and physiological mechanisms that function together to minimize and repair any damage incurred by the plant during the process of dehydration and rehydration (Harten & Eickmeier, 1986). The physiological mechanisms employed by S. lepidophylla species are described below.

3.4.1.1: ORGAN-LEVEL DEFORMATION

The actuating organ of S. lepidophylla is the stem which curls (tip toward base) during dehydration. Stem curling is critical for surviving desiccation (Lebkuecher &

Eickmeier, 1991, 1993). Two studies, one performed under controlled laboratory conditions (Lebkuecher & Eickmeier, 1991) and another in fieldwork conditions

(Lebkuecher & Eickmeier, 1993), demonstrated that plants inhibited from closing during drying (by heavy mesh nets) died as a result of photo-irradiation damage. In both studies, photosynthetic electron transport activity, changes in chlorophyll content, and changes in the rates of carbon dioxide exchange were monitored. Restrained plants (those inhibited

57 from curling in response to drying) demonstrated decreases in photosynthetic capacity, as well as a decrease in the amount of carbon dioxide uptake as compared to unrestrained control plants. Plants that were prevented from curling (via heavy mesh nets placed over hydrated plants) did not recover photosynthetic ability when rehydrated. It was determined that the curling of branches is responsible for reducing photo-irradiation damage to photosynthetic surfaces by 99.7% in S. lepidophylla (Lebkuecher &

Eickmeier, 1991, 1993).

Stem curling coincides with the time point between the activation and establishment of inducible protective mechanisms within S. lepidophylla, an interval at which plants are especially vulnerable to photo-irradiation damage (Brighigna et al.,

2002). They therefore rely upon the physical shading of photosynthetic surfaces, which is accomplished by curling, to reduce any damage incurred while inducible mechanisms are activated and established. Stem uncurling also occurs at a second critical time point. The slow uncurling of living stems allows time for repair processes to mend any damage accumulated during desiccation. Therefore, the physiological act of curling and uncurling is necessary for S. lepidophylla to survive desiccation. This was also confirmed by looking at other members of the Sellaginellaceae family; despite evolving desiccation tolerance independently of each other, all desiccation tolerant Selaginella display similar branch folding. To a lesser extent, drought tolerant Selaginella also undergo slight curling of their branches, though it is significantly less drastic than what is observed in desiccation tolerant Selaginella (Korall & Kenrick, 2002).

3.4.1.2: CELL-LEVEL DEFORMATION

Few studies have analyzed changes in cellular morphology that occur upon

dehydration and rehydration of S. lepidophylla. Among these, most studies have only

analyzed the changes in cellular organization as plants are rehydrated. Below is a

description of the main changes observed at various time points after initial wetting of S.

lepidophylla (Bergtrom et al., 1982; Brighigna et al., 2002; Thomson & Platt, 1997):

(1) Change in cell volume: desiccated cortex cells in the stem were found to be compacted and optically dense. Upon wetting, cells in the stem of S. lepidophylla became rounder and increased in volume, corresponding to hydration of cytosol and vacuole.

While cells do shrink in volume and organelle compartments become condensed, cell organization and integrity is maintained (Platt, Oliver, & Thomson, 1997; Thomson &

Platt, 1997). No damage or alterations to cell organization were observed during gradual rehydration either (Platt et al., 1997).

(2) Change in vacuole shape and vacuolar content: in S. lepidophylla fragmented,

optically dense vacuoles (most likely due to the presence of osmoprotectants in the

vacuoles) were observed. As cells rehydrated, the vacuoles grew less optically dense,

corresponding to water replacing macromolecules and solutes that were stored in the

vacuole while the plant was in a dehydrated state (Bergtrom et al., 1982; Brighigna et al.,

2002; Platt et al., 1997; Thomson & Platt, 1997).

3.4.1.2: CELL WALL-LEVEL DEFORMATION

The cell walls of S. lepidophylla microphylls fold extensively upon dehydration

(Platt et al., 1997; Thomson & Platt, 1997). To date, no studies in the literature have examined cell wall folding of the different stem tissues, however. Similarly, little work has been conducted to understand the composition of S. lepidophylla cell walls. A study looking at the cell wall composition of various plant types, including S. lepidophylla, showed the presence of high amounts of mannosyl and glucuronoarabinoxylan, hemicellulose-specific saccharides in S. lepidophylla stem tissue (Nothnagel &

Nothnagel, 2007). While these components may be of importance in modulating cell wall folding, further investigation is required to confirm if they play a role in surviving desiccation.

3.4.2: MYROTHAMNUS FLABELLIFOLIUS

The monophyletic family Myrothamnaceae possesses only two species:

Myrothamnus flabellifolius and Myrothamnus moschatus, both of which are desiccation tolerant, though only M. flabellifolius has been studied (Lüttge et al., 2011). M. flabellifolius is a woody shrub native Africa, and is the tallest and largest known resurrection plant (it can reach heights of 0.5-1.5m) (Lüttge et al., 2011; Moore et al.,

2006). It is believed that M. flabellifolius is one of the first angiosperms species to have evolved desiccation tolerance; it diverged from Gunnera (an angiosperm genus) 120 million years ago and has not diverged any further since (Moore, Lindsey, Farrant, &

Brandt, 2007; Moore et al., 2006; Wanntorp, Wanntorp, Oxelman, & Källersjö, 2001). M. flabellifolius prefers rocky inselbergs and outcroppings, and occupies mountainous regions in south and east Africa as either single shrubs or groups (termed colonies) (John

P Moore et al., 2007; J. P. Moore et al., 2007). M. flabellifolius is dependent on a long and extensive root system for water acquisition (John P Moore et al., 2007; J. P. Moore et al., 2007).

Unlike S. lepidophylla in which stems curl, M. flabellifolius shows a more complex mode of deformation. In this species of resurrection plant, the actuating organ is the leaf. In a fully hydrated state, leaves rest perpendicularly to the stem with two leaves positioned on either side of the stem (John P Moore et al., 2007; J. P. Moore et al., 2007;

Moore et al., 2006). The next pair of leaves is oriented at a ninety-degree angle to the pair above or below it (i.e., decussate) (John P Moore et al., 2007; J. P. Moore et al., 2007;

Moore et al., 2006). As M. flabellifolius desiccates, leaves collapse on themselves like a fan, and also fold upward so that they lie parallel to the stem (Kranner et al., 2002; Moore

61 et al., 2006). Similar to S. lepidophylla, it takes approximately 24-48 hours for M. flabellifolius leaves to fully rehydrate (Jill M Farrant & Kruger, 2001; Sherwin & Farrant,

1996), and approximately 48 hours for M. flabellifolius to dehydrate down to ~5% relative water content (J. Farrant, Vander Willigen, Loffell, Bartsch, & Whittaker, 2003).

As an angiosperm resurrection species, M. flabellifolius primarily relies upon inducible mechanisms that prevent cellular damage during dehydration and rehydration of its vegetative tissues (Dinakar & Bartels, 2013; Dinakar et al., 2012). However, it does possess some constitutive mechanisms as well. Both constitutive and inducible mechanisms are described below.

3.4.2.1: ORGAN-LEVEL DEFORMATION

As previously indicated, the leaves of M. flabellifolius collapse on themselves and also fold upward so that they lie parallel along the stem (Moore et al., 2006). The leaves are composed of sclerenchyma ribs joined together by pliable parenchymal tissue. The pliable parenchymatous tissue folds upon itself during dehydration, bringing the sclerenchyma ribs closer together and minimizing the leaf surface area available for photo-irradiation (J. Farrant et al., 2003; Moore et al., 2006; Sherwin & Farrant, 1996). In addition to bending up toward the stem, dehydrating leaves also curve as they move toward the stem, changing from a flat to concave shape. While not assessed in the literature, it is possible that curving helps to pull the leaves from a perpendicular to parallel conformation relative to the stem axis during dehydration, and likewise may help to push the leaves back down to a perpendicular position during rehydration.

3.4.2.2: CELL-LEVEL DEFORMATION

Similar to what was observed for S. lepidophylla, vacuoles in desiccated M. flabellifolius cells contain optically dense non-aqueous solutes and macromolecules, which was again proposed to provide mechanical stabilization for the cell and prevent cell wall collapse (Morse et al., 2011). While not discussed in the literature, this is probably very important for leaf parenchymal cells since they fold extensively and lose nearly half their cellular volume (44%) during dehydration (Jill M Farrant, 2000).

3.4.2.3: CELL WALL-LEVEL DEFORMATION

Cell wall folding occurs only in specific tissues of M. flabellifolius. The parenchymal cells of the leaf show cell wall folding, while the sclerenchymal rib and vascular cell walls do not (Jill M Farrant, 2000; Moore et al., 2006). As previously mentioned, M. flabellifolius employs inducible and (some) constitutive mechanisms to survive dehydration and rehydration processes. It has been postulated that no de novo changes in cell wall architecture arise as a result of dehydration in M. flabellifolius parenchymal cells, but rather the cell wall composition primes the cell wall for folding

(Moore, Farrant, et al., 2008; Moore et al., 2006). A higher content of soluble pectin is observed, as well as a higher abundance of arabinose, which is proposed to be closely linked to pectins in the wall of M. flabellifolius (Dinakar et al., 2012; Moore et al., 2006;

Vicré, Sherwin, Driouich, Jaffer, & Farrant, 1999). This specific cell wall composition is proposed to plasticize the cell wall, with the soluble pectin and arabinan side chains acting as cell wall modulators to increase the flexibility of the cell walls during desiccation and rehydration so that polymers can more easily slide past each other and

(un)compact the cell wall (Moore et al., 2006; Oliver et al., 2000). In addition, the arabinan side chains, found on rhamnogalacturonan I, a common pectin polysaccharide, are also thought to promote reversible folding of the cell wall by preventing unfavourable polymer interaction and irreversible cross-linking during cell wall compaction (Moore,

Farrant, et al., 2008; Moore et al., 2013). To ensure cell walls can unfold during rehydration, it has been suggested that arabinan side chains form ‘egg-box’ structures that prevent unfavourable interactions between polymers that are at risk of irreversibly binding as the cell wall dehydrates (Moore, Farrant, et al., 2008; Moore et al., 2013).

These ‘egg-box’ structures also provide mechanical stability to the compacted wall, helping to prevent cell collapse and plasmalemma tearing (Moore, Farrant, et al., 2008;

Moore et al., 2013).

SECTION 4: USING RESURRECTION PLANTS TO ELUCIDATE

PROPERTIES OF HYGROACTUATION

Below are potential solutions to the challenges outlined in Section 2.4 that could be addressed by studying resurrection plants as models for hygroactuation.

4.1: AVAILABILITY, TYPE, AND STRENGTH OF MATERIALS FOR

HYDROACTUATORS

(1) Water as a solvent: though not specific to resurrection plants, using water as a stimulus for actuation offers the potential for more material combinations of shape changing polymers than those driven by changes in temperature or electricity, which require materials that can withstand drastic changes in temperature or high voltages (H.

Lv, Leng, Liu, & Du, 2008). In addition, water is readily available and cost effective for industrial-scale applications (H. Lv et al., 2008; Zhou, Zhao, & Tian, 2012). As well, unlike wheat awns, pine cones, and tension wood in which differential swelling/shrinking occurs in dead vegetative tissue, the vegetative tissue of certain resurrection plant species

(e.g., S. lepidophylla) can undergo deformation in both living and dead states. Therefore, comparing and contrasting fluid-driven actuation in both vegetative tissue states within the same plant system could possibly identify actuation mechanisms that are specific to either living or dead states, as well as those mechanisms that are common to both states.

This could provide insight into the interaction of living versus dead material with water, leading to a better understanding of hydrodynamics within different actuating material types.

4.2: HYGROACTUATOR LONGEVITY AND CONFORMATIONAL

REVERSIBILITY

(1) Hierarchical Conformational Reversibility: as previously mentioned, many existing synthetic actuators were generated from large length scale observations without taking into account smaller scale (i.e., micrometer to nanometer) deformation properties and structural damage (Erb et al., 2013; C. Lv et al., 2018). To achieve more complex modes of deformation (degree of deformation, stepwise deformation through multiple stable conformations, etc.), further study of reversible conformation at smaller length scales is required (Erb et al., 2013; C. Lv et al., 2018). While plant models exist that demonstrate deformation at various length scales (e.g., wheat awn/pinecone at the tissue level, tension wood and coiling tendrils/vines at the cell wall level), there are currently no models that explore deformation at multiple length scales within the same plant system.

Resurrection plants offer an opportunity to examine the relationship among various length scale deformation properties within a single plant species.

(2) Longevity: resurrection plants are able to undergo multiple, repetitive cycles of transitioning between wet and dry states without permanent deformation or damage to their vegetative tissue structure (Brighigna et al., 2002; Moore, Vicre-Gibouin, et al.,

2008; Potts & Penfound, 1948). Mimicking long functional lifespan in synthetic hygroactuators, especially hydrogels, has been challenging. Even fracture-resistant hydrogels eventually show permanent deformation after a set number of deformation cycles (Bai et al., 2018; Xuanhe Zhao, 2014). In contrast, resurrection plants do not display a limited number of cycles; plants either deform until vegetative tissue death

(e.g., M. flabellifolius), or continue to deform in response to water even after vegetative

66 tissue death (e.g., S. lepidophylla). Therefore, examining the mechanisms enabling resurrections plants to continually deform without apparent structural damage at multiple length scales could provide insight into properties that could improve the functional lifespan of synthetic actuators.

RESEARCH OBJECTIVES

Given the adaptive physiological features of resurrection plants outlined above in

Sections 3 and 4, I hypothesize that resurrection plants can be used as new models to

study hierarchical, water-driven actuation. In this thesis, the resurrection plant Selaginella

lepidophylla is explored as a model to study the mechanical, morphological, and

compositional features leading to reversible, hierarchical actuation in this species.

Myrothamnus flabellifolius is also preliminarily explored (Appendix 4). S. lepidophylla

and M. flabellifolius were chosen as they represent different types of resurrection plants

(S. lepidophylla is a primitive spikemoss, whereas M. flabellifolius is an angiosperm)

with different modes of deformation (curling versus fan-like folding). The following

objectives address the hypothesis:

OBJECTIVE 1: CHARACTERIZATION OF PLANT MOVEMENT

Some studies have examined the overall folding patterns of S. lepidophylla plants

(Harten & Eickmeier, 1986; Lebkuecher & Eickmeier, 1991, 1993). However, to date, no

studies have characterized the specific folding pattern of individual S. lepidophylla stem

types, which is difficult to observe when stems are attached to the plant. Here, the

deformation patterns of individual S. lepidophylla stem types are characterized, and

microtensile testing is used to explore the underlying mechanical forces driving stem

movement. Data addressing Objective 1 with respect to S. lepidophylla are presented in

Chapter 2 and Chapter 3.

Similarly, some studies have characterized M. flabellifolius leaf folding patterns

(Korte & Porembski, 2012; John P Moore et al., 2007; J. P. Moore et al., 2007; Moore et

al., 2006). However, none of these studies have fully characterized the kinematics of leaf

folding. Here, the deformation patterns of individual M. flabellifolius leaves are

characterized. Preliminary data addressing Objective 1 with respect to M. flabellifolius

are presented in Appendix 4.

OBJECTIVE 2: MORPHOLOGICAL AND COMPOSITIONAL FEATURES

RESPONSIBLE FOR ORGAN-LEVEL ACTUATION

In terms of tissue and cell wall morphology and composition, S. lepidophylla

stems have not been extensively studied (Bergtrom et al., 1982; Nothnagel & Nothnagel,

2007). Most groups have focused on microphylls rather than stems (Bergtrom et al.,

1982; Brighigna et al., 2002). None of the research in the literature explores the

mechanical properties of S. lepidophylla stem tissues and cell walls. This thesis

investigates S. lepidophylla stem morphology and composition at both tissue and cell

wall levels. The mechanical properties at each level are also investigated. Data addressing

Objective 2 are primarily presented in Chapter 3, with preliminary data presented in

Chapter 2, and additional data presented in Chapter 4 and Appendices 2-3.

OBJECTIVE 3: WATER DYNAMICS

To gain a comprehensive understanding of actuation in S. lepidophylla and M. flabellifolius, the interaction of water with plant tissue needs to be examined. To date, no studies have investigated water movement through S. lepidophylla other than measuring drying rates for whole plants (Bergtrom et al., 1982; Brighigna et al., 2002; Lebkuecher

& Eickmeier, 1991, 1993). In M. flabellifolius, water transport has been primarily investigated in the stems rather than the leaves of the plant (Canny, 2000; Schneider et al., 2003; Schneider et al., 1999; Sherwin, Pammenter, February, Vander Willigen, &

Farrant, 1998). The investigations conducted in this thesis carry these water movement experiments further, and examine the role that water plays in driving S. lepidophylla stem and leaf M. flabellifolius movement. Preliminary data addressing Objective 3 are presented in Appendix 4.

FIGURES

Figure 1.1.

72 Figure 1.1. Hygroscopic Movement of Wheat and Pinecones. Movement and deformation direction result from differences in active and passive tissue layers. At the cell wall level, cellulose in the active tissue layer is oriented in all directions. This results in anisotropic swelling (blue arrow) and shrinking (pink arrow) of cells. Cellulose in the resistance tissue layer is oriented parallel to the longitudinal cell axis, resulting in isotropic swelling and shrinking along this axis. Thus, active tissue swells and shrinks anisotropically, while resistance tissue swells and shrinks isotropically. However, when active and resistance tissue layers are joined side by side, resistance tissue limits the direction of swelling and shrinking of active tissue (inhibits swelling in the directions marked by white Xs), thereby controlling the direction of movement of the two tissue layers. In wheat awns and pinecone scales, this directional bending allows for changes in organ shape, leading to seed dispersal (Dawson et al., 1997; Dunlop et al., 2011; R.

Elbaum et al., 2008; R. Elbaum et al., 2007; Reyssat & Mahadevan, 2009).

Figure 1.2.

74 Figure 1.2. Gelatinous Fibers in Coiling Tendrils/Vines and Tension Wood. G-fiber cells possess a normal secondary cell wall layer (green), as well as a cellulose-enriched tertiary G-layer (purple). This G-layer can swell to great volumes upon water uptake.

Swelling causes the G-layer to expand and place pressure on the more rigid, lignified secondary cell wall layer. Combined with cell shape, this causes G-fiber cells to laterally expand and longitudinally contract. (Note: Specialized keel cells in ice seed plant capsules function in a similar manner due to a highly hydrophilic, swellable cellulosic inner layer in their cell walls). Different modes of organ deformation arise from the presence of G-fibers (purple) in plant tissue, and these modes depend on the amount and distribution of G-fiber cells within tissue. In coiling tendrils/vines, there are three common distributions of G-fibers: (1) a ring near the center of the stem, which leads to multi-directional coiling at the organ level; (2) a plate of G-fibers on one side of the stem, leading to unidirectional coiling; and (3) a G-fibers at the center of the stem, leading to sticky stem ends for attachment and coiling. In tension wood, G-fibers are normally found on the upper side of the tree limb, resulting in limbs curving upward, as indicated by the pink arrow and branch at the bottom of the figure (Bowling & Vaughn, 2009;

Harrington et al., 2011; Lafarguette et al., 2004; Meloche et al., 2007; Pilate et al., 2004).

Figure 1.3.

76 Figure 1.3. Osmotic Gradients in Venus Flytrap and Touch-Me-Nots. Venus Flytrap and Touch-Me-Not operate via the same principle as deformation is dependent on rapid changes in turgor (mediated by ion channels and aquaporins) between cell types (flexor and extensor in Touch-Me-Not species) or plant surfaces (inner and outer in Venus

Flytraps). At the cell wall level in Venus Flytraps in open (convex) conformation, cells on the inner surface are very turgid while cells on the outer surface are less turgid. Upon stimulation, cells on the outer surface rapidly swell and cells on the inner surface lose turgidity. This causes inner surface tissue shrinking and outer surface tissue expansion, leading to a change in conformation (open/convex to closed/concave). In Touch-Me-

Nots, flexor and extensor cells follow a similar pattern of swelling and shrinking, leading to tissue contraction and expansion, and at the organ level raising or lowering of individual leaves (Allen, 1969; Braam, 2005; Burgert & Fratzl, 2009b; Forterre et al.,

2005; Hodick & Sievers, 1989; Temmei et al., 2005; Volkov et al., 2008; Volkov, Foster,

Ashby, et al., 2010).

Table 1.1. Commonly Studied Biological Actuators Organism Actuating Organ Living/Dead Stimulus References

Bacteria

E. coli Flagellae Living Chemical 1-3 S. marcescens Flagellae Living Chemical 2-5 M. gryphiswaldense Flagella-like magnetosomes Living Magnetotatic 6-7 T. periformis Flagella-like magnetosomes Living Magnetotatic 8

Animals

Arachnids Leg (lyriform) Living Mechanosensory 9-11 Cephalopods Chromatophores Living Electrical 12-14 Annelids Muscle fibers Living Electrical 15-16

Plants

Mechanosensory; 17-19 Venus flytrap Leaves Living electrical Mechanosensory; 20-22 Touch-Me-Not Leaves Living electrical Pine cone Scales Dead Water 23-25 Wheat awn Awns Dead Water 26-28 Ice plant seed capsule Protective valves Dead Water 28-30 Coiling tendrils/vines Stems Living Water 31-33 Tension wood Limbs Dead Water 32, 34-35 Resurrection plants Stems; leaves Living; dead Water 36-38 Orchid seed pods Seed pods Dead Water 38-39

(1) (Passino, 2002); (2) (Julius et al., 2009); (3) (Mutlu, Alici, & Li, 2011); (4) (Sakar et al., 2011); (5) (D. Li et al., 2015); (6) (Martel, Tremblay, Ngakeng, & Langlois, 2006); (7) (Martel et al., 2006); (8) (D. H. Kim, Cheang, Kőhidai, Byun, & Kim, 2010); (9) (Hößl, Böhm, Rammerstorfer, Müllan, & Barth, 2006); (10) (Hößl, Böhm, Rammerstorfer, & Barth, 2007); (11) (Fratzl & Barth, 2009); (12) (Florey & Kriebel, 1969); (13) (Rossiter et al., 2012); (14) (Deravi et al., 2014); (15) (S. Kim, Laschi, & Trimmer, 2013); (16) (Xu et al., 2017); (17) (Hodick & Sievers, 1989); (18) (Forterre et al., 2005); (19) (Forterre, 2013); (20) (Allen, 1969); (21) (Temmei et al., 2005); (22) (Volkov, Foster, Ashby, et al., 2010); (23) (Dawson et al., 1997); (24) (Reyssat & Mahadevan, 2009); (25) (Dunlop et al., 2011); (26) (R. Elbaum et al., 2007); (27) (R. Elbaum et al., 2008); (28) (Rivka Elbaum & Abraham, 2014); (29) (Harrington et al., 2011); (30) (Studart, 2015); (31) (Bowling & Vaughn, 2008); (32) (Bowling & Vaughn, 2009); (33) (Vaughn & Bowling, 2010); (34) (Lafarguette et al., 2004); (35) (Pilate et al., 2004); (36) (Rascio & Rocca, 2005); (37) (Brulé et al., 2016); (37) (Dinakar & Bartels, 2013); (38) (Erb et al., 2013); (39) (Ionov, 2017)

Table 1.2. Commonly Studied Resurrection Plant Species Mechanical Species "Omics" Analysis* Type References Analysis

Lycophytes

Metabolomic; proteomic; Kinematics; Selaginella lepidophylla + 1-7 genomic material elasticity Metabolomic; proteomic; Selaginella tamariscina - - 8-10 genomic Selaginella bryopteris Proteomic - - 11-13

Bryophytes

Tortula ruralis Genomics - - 14-17

Epiphytes

Kinematics; Polypodium polypodioides Proteomic + 18-20 material elasticity

Dicots

Myrothamnus flabellifolius Genomic + Kinematics 21, ** Metabolomic; proteomic; Craterostigma plantagineum - - 22-26 Genomic Craterostigma wilmsii Metabolomic; proteomic + Tensile 27-29 Lindernia brevidens Genomic - - 30-31 Metabolomic; proteomic; Boea hygrometrica - - 32-34 genomic Metabolomic; proteomic; Haberlea rhodopensis - - 35-41 genomic

Monocots

Xerophyta viscosa Proteomic; genomic - - 34, 42-46 Xerophyta humilis Metabolomic; genomic + Tensile 29, 42, 47-48 Metabolomic; Sporobolus stapfianus + Tensile 29, 49-51 proteomic;genomic * Genomics includes transcriptomics ** Appendix 4.2 (1) (Harten & Eickmeier, 1986); (2) (Iturriaga, Cushman, & Cushman, 2006); (3) (Nothnagel & Nothnagel, 2007); (4) (Weng et al., 2010); (5) (A. Yobi et al., 2012); (6) (A. Yobi et al., 2013); (7) (Rafsanjani, Brulé, Western, & Pasini, 2015); (8) (M. S. Liu, Chien, & Lin, 2008); (9) (X. Wang et al., 2010); (10) (Deeba, Pandey, Pathre, & Kanojiya, 2009); (11) (Pandey et al., 2010); (12) (Antony & Thomas, 2011); (13) (Deeba, Pandey, & Pandey, 2016); (14) (Wood, Duff, & Oliver, 1999); (15) (X. Chen, Zeng, & Wood, 2002); (16) (Oliver, Dowd, Zaragoza, Mauget, & Payton, 2004); (17) (Oliver et al., 2010); (18) (Helseth & Fischer, 2005); (19) (Helseth, 2008); (20) (Layton et al., 2010); (21) (Chao Ma et al., 2015); (22) (Zonneveld, Leitch, & Bennett, 2005); (23) (Rohrig et al., 2006); (24) (Röhrig et al., 2008); (25) (Rodriguez et al., 2010); (26) (Gasulla et al., 2013); (27) (Sherwin & Farrant, 1996); (28) (Cooper & Farrant, 2002); (29) (Hedderson, Balsamo, Cooper, & Farrant, 2009); (30) (Smith-Espinoza, Bartels, & Phillips, 2007); (31) (Giarola & Bartels, 2015); (32) (Jiang et al., 2007); (33) (Y. Zhao et al., 2014); (34) (R.-Z. Sun et

79 al., 2018); (35) (Georgieva, Christov, & Djilianov, 2012); (36) (Apostolova et al., 2012); (37) (Gechev, Dinakar, Benina, Toneva, & Bartels, 2012); (38) (P Mladenov et al., 2013); (39) (Moyankova et al., 2014); (40) (Petko Mladenov et al., 2015); (41) (Xiao et al., 2015); (42) (Collett et al., 2004); (43) (Ingle, Schmidt, Farrant, Thomson, & Mundree, 2007); (44) (Abdalla, Baker, & Rafudeen, 2010); (45) (Abdalla & Rafudeen, 2012); (46) (Costa, Cooper, Hilhorst, & Farrant, 2017); (47) (Dace, Sherwin, Illing, & Farrant, 1998); (48) (Collett, Butowt, Smith, Farrant, & Illing, 2003); (49) (Neale et al., 2000); (50) (Oliver et al., 2011); (51) (Abou Yobi et al., 2017)

LINK BETWEEN CHAPTER 1 AND 2

Chapter 1 outlines recent work in the field of actuation, specifically examining plant-based actuation and taking into account limitations and recent strides made in this domain of study. Resurrection plants are proposed as a model to investigate more complex deformation patterns at different length scales, and two species are detailed in terms of their physiology and function. Chapter 2 continues to examine resurrection plants as a model for water-based actuation, and focuses on the study of Selaginella lepidophylla at the organ level. This study aims to characterize the movement and patterns of deformation observed in S. lepidophylla organs, and to investigate some of the underlying properties leading to these patterns of deformation. The contents of Chapter

2 were published in 2015 in Scientific Reports (Rafsanjani et al., 2015).

CHAPTER 2

Hydro-Responsive Curling of the Resurrection Plant Selaginella lepidophylla

ABSTRACT

The spirally arranged stems of the spikemoss Selaginella lepidophylla, an ancient resurrection plant, compactly curl into a nest-ball shape upon dehydration. Due to its spiral phyllotaxy, older outer stems on the plant interlace and envelope the younger inner stems forming at the plant centre. Stem curling is a morphological mechanism that limits photoinhibitory and thermal damage the plant might experience in arid environments.

Here, we investigate the distinct conformational changes of outer and inner stems of S. lepidophylla triggered by dehydration. Outer stems bend into circular arcs in a relatively short period of desiccation, whereas inner stems curl slowly into spirals due to hydro- actuated strain gradient along their length. This arrangement eases both the tight packing of the plant during desiccation and its fast opening upon rehydration. The insights gained from this work shed light on the hydro-responsive movements in plants and might contribute to the development of deployable structures with remarkable shape transformations in response to environmental stimuli.

INTRODUCTION

Resurrection plants are vascular plants tolerant to extreme vegetative desiccation that are able to resume normal growth and metabolic activity upon rehydration (Rascio &

Rocca, 2005). The spikemoss Selaginella lepidophylla is an ancient resurrection plant native to Chihuahuan desert (Mexico and United States) that shows dramatic curling and uncurling with changes in plant hydration (Banks, 2009; Eickmeier, 1980). When dehydrated, the spirally arranged stems of S. lepidophylla tightly curl to form a rough sphere. As a result of this morphology, the outer stems serve to substantially reduce solar radiation (99.7%) exposed to inner stems at the centre of the plant (Harten & Eickmeier,

1986). The morphological and anatomical traits of S. lepidophylla in relation to the curling of its stems were examined at the turn of the twentieth century (du Sablon, 1888).

At that time, it was elucidated that the movements of the tissues are entirely physical rather than biophysical, and depend upon the hygroscopic capacities of the tissues

(Uphof, 1920).

While fascinating for botanists, adaptive movements in plants can inspire material scientists and engineers to exploit the underlying mechanisms for the development of innovative biomimetic materials and actuating devices that show intriguing shape transformations in response to environmental stimuli (de Haan, Verjans, Broer,

Bastiaansen, & Schenning, 2014; Erb et al., 2013; Guiducci, 2014; Guiducci, Fratzl,

Bréchet, & Dunlop, 2014; Ionov, 2013; Razghandi et al., 2014; Xuanhe Zhao, 2014).

Nastic movements in plants are generally driven by hydration motors of osmotic, colloid

or fibrous design, where the direction of the movement is determined by integrated

features of mobile tissue, rather than by stimulus direction (Burgert & Fratzl, 2009b;

Stahlberg, 2009). In particular, the plant motion in fibrous motors relies on the relatively

slow variation of the water content within the internal capillary spaces of parallel-

arranged cellulose fibres in the cell walls (Skotheim & Mahadevan, 2005). Changes in

hydration result in anisotropic swelling and shrinkage strains that emerge in the direction

transverse to the long axis of the cells. By combining different layup of cellulose layers,

plants generate a wide variety of water-controlled actuators that can trigger a diverse

range of complex movements. The swelling/shrinkage-induced movements are strongly

dependent on the stresses generated through moisture uptake, which may also occur in

non-living remnants of plants. Thus, sometimes they can be viewed from a purely

mechanical perspective (Bertinetti, Fischer, & Fratzl, 2013). Opening of pine cones

(Dawson et al., 1997), water responsive movements of the skeleton of the desert plants

Anastatica hierochuntica (Friedman, Gunderman, & Ellis, 1978; Hegazy, Barakat, &

Kabiel, 2006) and Asteriscus pygmaeus (Gutterman & Ginott, 1994), the unfolding

mechanism of the seed capsules of the desert ice plants from Aizoaceae family

(Harrington et al., 2011), self- burial mechanism of hygroscopically responsive Erodium awns (Abraham et al., 2012; Aharoni, Abraham, Elbaum, Sharon, & Kupferman, 2012;

Evangelista, Hotton, & Dumais, 2011), twisting of seed pods (Armon, Efrati, Kupferman,

& Sharon, 2011) and the walk and jump of Equisetum spores (Marmottant,

Ponomarenko, & Bienaimé, 2013) are examples of plant movements in response to environmental stimuli.

In this work, we investigate the moisture responsive curling of the stems of the spikemoss S. lepidophylla through a multidisciplinary combination of experiments, theory and numerical simulations. We aim at understanding the underlying mechanisms of nastic movements in this resurrection plant as a paradigm for the design of novel mechanisms of water-controlled actuation and structural deployment.

RESULTS

Plant Morphology

In the hydrated state, the spirally arranged stems of S. lepidophylla are flat and outstretched and, upon dehydration, they compactly curl into a spherical nest-ball shape with an average diameter of 6 to 8 cm (Figure 2.1A-B). The exterior stems are typically fragile and have a grey-brown colour. Protected within these older outer stems may be several layers of inner stems forming the centre of the flattened, green rosette of the plant that uncurl and are photosynthetically active upon rehydration in their native state

(Harten & Eickmeier, 1986).

Shape Transformation of Stems

The curling and uncurling of isolated stems taken from outer and inner portions of the spiral rosette of S. lepidophylla were filmed during hydration and dehydration. Figure

2.1C and D respectively show the curling sequences of outer and inner stems over the course of dehydration from the initial wet condition to the final dry state. When hydrated, both inner and outer stems were flattened, but when dehydrated, they exhibited large deformation with distinct curling patterns. Upon dehydration, the outer stems bend into an approximately circular arc in a relatively short period of desiccation, whereas the inner stems curl slowly and form a spiral configuration.

Morphology and Composition of Stem Sections

The observation of curling mechanisms of outer and inner stems naturally leads to the question of whether their distinct curling patterns are correlated to the morphology and composition of their underlying tissue. The stem of S. lepidophylla is composed of an annular-shaped layer of cortical tissue and an inner, central vascular bundle (protostele), which is separated from the cortical layer by an air-filled canal (Figure 2.1E). The vascular tissue comprises a central bundle of xylem cells surrounded by a sheath of phloem cells (i.e., an amphicribral bundle) (Brighigna et al., 2002). Specialized cells, termed trabeculae, are found within the air-filled canal. Due to their thick, cellulosic cell walls, trabeculae provide structural support, bridging the airspace between the central vascular bundle and the cortical tissue (Harholt et al., 2012).

The composition of the cell walls was investigated by staining Spurr’s resin- embedded stem sections (Supplementary Information 2.1) with Toluidine Blue O (TBO) for the detection of lignin and pectin under bright field microscopy. Figure 2.1E shows a cross-section of an inner stem at the root-base interface, and it reveals that, while lignified cell walls (green-blue) are found throughout the inner cortex tissue, there is an asymmetry in cell density, whereby cells appear to be smaller and/or denser on the abaxial side of the cross section. Further investigation with basic Fuchsin staining of paraffin-embedded sections (Supplementary Information 2.2) allowed detection of variation of lignified tissues at different sections along the stem length. Consistent with the TBO results, lignification is found throughout the inner cortex in the basal sections

(Figure 2.1F), whereas the middle of the stem shows lignified tissues in the abaxial side of the stem (Figure 2.1G). The lignified tissues are reduced to a narrow strip at the apical

tip of the stem (Figure 2.1H). For outer stems, the staining pattern is similar to that seen

at the basal section of inner stems without significant variation along the stem length

(Figure 2.1E and E; data not shown). The spatial variation of lignification along the stem

length reduces from the base to the apical tip of inner stems of S. lepidophylla, and it may

play a crucial role in their spiralling curling mechanism.

Curvature Characterization of Stems

The variation of curvature along the length of outer and inner stems (Figure 2.1C-

D) was characterized at different time intervals from an analysis of the discrete

curvatures (Langer, Belyaev, & Seidel, 2005) of the stem centreline (Figure 2.2A-B and

Supplementary Information 2.3). The stems had a small natural twist, which was

neglected in this analysis. The curvature and stem length were normalized with the total

length l respectively as κ* = κl and s* = s/l. Starting from an almost straight shape, after one hour of drying, the outer stem started to curl with a progressive curling front which was initially close to the tip before propagating toward the base. The final dehydrated

outer stem resembles an arc with a constant curvature. A propagating curling front was

also observed for the inner stem in the course of dehydration where its apical tip

exhibited more mobility and compliance compared to its base. The maximum achievable

curvature of the inner stem increases linearly along the stem length, i.e. k ∝ s

representing the geometry of an Euler (Cornu) spiral.

Mechanical Response of Stems to Dehydration

We investigated the mechanical aspects of the water-controlled curling of outer

and inner stems in more detail. The absolute displacements of the apical tip (ux and uy) of outer and inner stems were tracked during drying (Figure 2.2C-D). Both stem types experienced large deformation. Furthermore, fully hydrated stems were clamped at both ends and the reaction forces that prohibited stem curling were measured (Figure 2.2E). In a given period of time during dehydration, the weight loss w* of the wet stems was also determined (Figure 2.2F), a measure that allowed us to translate dehydration time into weight loss.

The mechanical response of each stem was strongly dependent on its moisture content. For the outer stem, a switch-like behaviour was observed. At early times of dehydration, the mobility of the stem tip was not notable and the generated axial force in the restrained stem was small. Once the outer stem reached a certain weight loss level, i.e. w* ≅ 45%, its movement accelerated and its tip moved downward. At the same water content, a very sharp increase in the force response of the restrained stem was observed.

On the other hand, the inner stem responded smoothly to dehydration. Initially, the stem remained still until w* ≅ 35%, then a moderately uniform transition was observed in both displacement and restraining force profiles. The change of the weight associated to the main vertical displacement of the apical tip for the outer stem was Δw* ≅ 5%, whereas

Δw* ≅ 25% for the inner stem. The maximum vertical velocity of the tip (duy/dt) of the

outer stem (≅ 2:7 mm/min) during dehydration was threefold larger than the one of the

inner stem (≅ 0:9 mm/min). Also, the resulting restraining force during dehydration in the

outer stem was about threefold larger than the force in the inner stem. The outer stem

reached its equilibrium weight loss, i.e. w* = 50%, after 1 hour whereas for the inner

stem it took about 5 hours to reach equilibrium, i.e. w* = 60%. The drying rate of the

outer stem was faster than the inner stem. A power law behaviour for the early drying

rate before reaching the equilibrium was observed with an exponent of 0.65 for the outer

stem and 0.41 for the inner one, suggesting a diffusive transport mechanism.

Curling Mechanisms of Inner and Outer Stems.

To elucidate the spiralling pattern of inner stems, we adopted a geometrical model

based on the normalized Euler (Cornu) spiral, which is defined as a curve with curvature

k that linearly changes with its curve length s, i.e. κ = a2s. In this setting, the Cartesian

coordinates of a normalized Euler spiral shape are given by the following Fresnel-like

integrals:

where s~ ∈ [0, a] is the curve parameter and a ∈ [0, η] represents a desiccation factor, where a a = 0 indicates the initial fully hydrated state and η the hydro-actuation strain gradient generated in the stem section upon dehydration. Generalization of the above geometrical model to a hypothetical class of spirals with power-law curvature (defined by the Cesàro equation κ = ar+1sr) allows us to investigate the role of material variation

along the stem length (Figure 2.3A and Supplementary Information 2.4). The mobility of

the stems decreases by increasing the exponent r, an outcome that indicates the curling

pattern reflects the hydro-actuation capacity of the tissue. Alternatively, stem spiralling

can be simulated via finite element (FE) modelling of inextensible elastic beams by

introducing hydro-actuation strain gradient through the beam cross section (Figure 2.3B-

C). The curling of spirally arranged stems of S. lepidophylla triggered by dehydration has

been examined with a simplified FE model of the whole plant (Figure 2.3D). The curling

of exterior stems with constant curvature allows them to envelope spiralled inner stems in

the centre of the plant.

From a structural point of view, the stem of S. lepidophylla can be seen as a

bilayer composed of non-lignified active (a) and lignified passive (p) layers that curls

upon dehydration due to a mismatch in the eigenstrains developed in each layer. In outer

stems, the relative thickness of the constituent layers is constant, whereas in inner stems

the relative thickness of the active (passive) layer increases (decreases) from the base to

the apical tip of the stem (Figure 2.4A and Figure 2.1F-H). The curvature of a bilayer in

terms of elastic modulus, geometry and actuation eigenstrains (εa and εp) of the

constituent layers is given by the Timoshenko (Timoshenko, 1925) bimetallic model as:

where h = ha + hp is the total thickness of the bilayer with thickness ratio m = hp/ha and elastic moduli ratio n = Ep/Ea. The conformations of several bilayer stems (h/l = 0.02,

Ep/Ea = 2) subjected to an actuation contraction strain (εa = – 0.2) where the constituents’ thickness varies linearly along the stem length are computed by large deformation FE simulations (Figure 2.4A). In particular, when the ha/h ratio increases at the base of the stem, a transition from spiral to circular configuration is observed. In addition, for

92 increasingly higher values of the hp/h ratio, the number of turns of each stem concept reduces during deformation. The curvature of the representative bilayer models of an outer stem with constant relative thickness and an inner stem with linear thickness profile is calculated by Eq. (2) and FE simulations (Figure 2.4B). The FE model accounts for geometric nonlinearities that large actuation strains trigger, as Supplementary Fig 2.3 in the supplemental information further explains, while the Timoshenko bimetal model is only accurate for small deformation. The results compare well with the deformed shape of the stems, and the predicted curvature follows similar trends observed in experiments.

From this analysis, we can gather that the proportion of the constituent passive to active tissues along the stem length governs the resulting shape of the stem.

DISCUSSION

Stem curling triggered by desiccation is a morphological mechanism in the desert plant S. lepidophylla. This movement has an ecophysiological importance that limits photoinhibitory and thermal damage to the plant and provides a way to overcome bright- light, high-temperature, and water-deficit stresses (Lebkuecher & Eickmeier, 1993). We observed distinct large deformation mechanisms (Figure 2.1C-D), mechanical responses

(Figure 2.2A-D) and time scales (Figure 2.2F) in moisture responsive curling of outer and inner stems of S. lepidophylla, which may ameliorate their photo-protection function. Our study reveals that two mechanisms are potentially effective (Figure 2.3D) in moisture responsive curling of the spikemoss S. lepidophylla: (i) stiff outer stems located in the exterior portion of the plant behave as classical bilayers curling with an approximately constant curvature, and (ii) compliant inner stems located in the centre of the plant introduce a class of rod-like hygromorphs (Reyssat & Mahadevan, 2009) with planar spiralling capacity. This arrangement, when paired with the plant’s spiral phyllotaxy that yields a gradient of stem curling patterns, can adaptively regulate the packing and deployment of the whole plant by according the closure of the plant in dry periods and its opening at times of water retrieval.

One of the main differences between inner and outer stems of S. lepidophylla lies in their shape transformation mechanisms. In both cases, the hydro-responsive movements are mechanical and governed by the properties of the tissues and chemical composition of the cell walls. At the level studied, the main difference affecting bending in the outer stems appears to be the presence of asymmetric cell density between the

94 abaxial and adaxial sides of the stem. For inner stems, however, microscopic studies also revealed the presence of an asymmetric lignification in the cortical tissue towards the abaxial side of the stem (Figure 2.1E). Furthermore, in inner stems, the distribution of lignified cells varies along the stem length (Figure 2.1F-H). The presence of the lignified cells alters the hydro-actuation capacity of the stem and locally increases the stiffness of the cortical tissue. Upon dehydration, differential shrinkage strains are induced in the tissue, and are relaxed through conformational changes of the stems (Figure 2.1C-D). The spatial hydro-actuation strain gradient along the stem length plays an important role in the spiralling pattern of dehydrated stems (Figure 2.3A). Conformations similar to those of the inner stems have been observed in the growth process of coiling tendrils (Y. Wang,

Chantreau, Sibout, & Hawkins, 2013). In the balloon vine tendril Cardiospermum hallachum, gelatinous fibres with highly lignified secondary walls generate a contractile force that converts elongated, straight tendrils into a planar spiral shape (Bowling &

Vaughn, 2009).

The stems of S. lepidophylla can robustly curl/uncurl over several dehydration/hydration cycles without structural damage. The reversibility of the movement indicates that the shape of the stems mirrors their water content. The stems respond to dehydration once the water loss exceeds a certain threshold (Figure 2.2C-D).

This behaviour can be attributed to the existence of free water in liquid form in the interior of the cells and bound water within the fibrous cell walls, a phenomenon also observed in wood cells (Siau, 1984). During dehydration, free water molecules can easily leave the cell cavities (though perhaps more slowly from the cytoplasm if cells are living), and despite their significant volume, they appear to only have minor influence on

95 the deformation of the stem (Figure 2.2C-D). On the other hand, the bound water molecules are strongly absorbed into the microscopic voids of the fibrous cell walls.

During drying desiccation, the fibrous cell walls shrink and produce the driving force for deformation, thereby leading to stem curling. Differences in the degree of curling and the rate of water loss can be affected by the lignification status of cell walls, as cellulose can hydrogen-bond water molecules and lignin is hydrophobic.

In summary, the spikemoss S. lepidophylla exploits simple but effective strategies to carefully pack and deploy many of its rod-like stems (Figure 2.3D). The spiral phyllotaxy is organized outward in order of ascending length and increasing lignification along the length of the stems (Figure 2.1B). This arrangement, in cooperation with the identified curling mechanisms (Figure 2.3B-C), facilitates tight packing and fast opening of the plant with a minimized interlocking between stems despite their large deformations. The insight gained from this study might set the stage for the design and development of deployable structures (e.g., in communication antennas and morphogenetic architectural designs) and actuating devices that can yield programmable shape transformations (e.g., stent deployment devices) in constant feedback and interaction with their functioning environment.

METHODS

Plant materials

Mature dehydrated S. lepidophylla were obtained from Canadian Air Plants in

New Brunswick, Canada. Plants were maintained unplanted in a desiccated state under laboratory room conditions (25°C, 50% relative humidity).

Time-lapse Video Capture

Time-lapse videos were captured using a Logitech C920 HD Pro Webcam

(1080p, Carl Zeiss optics) and Video Velocity Time-Lapse Studio software (Candylabs).

S. lepidophylla plants were allowed to rehydrate for 24 hours, and individual inner stems

(chosen from 1/4 to 1/3 of the way along the spiral to match the length of mature stems) and outer stems (chosen from approximately 3/4 the way along the spiral) of ~5 cm length were cut from the plant at the root-stem interface. Stems were secured into individual metal clamps, which were affixed to the base of a square Petri dish. Stems were then allowed to air-dry for approximately 6 hours, during which time changes in their curvature were captured via time-lapse filming at frame rate of 1 min-1. Stems were taken from three different S. lepidophylla plants. In total, four inner stems and four outer stems were tested and displayed similar curling/uncurling patterns. The models presented in this article were based upon a single representative stem from both inner and outer sample pools.

Drying Force Measurement

Rehydrated stems were secured at both ends by clamps and drying induced tensile

force was measured using an ADMET MicroEP machine. Stems (8 inner and 8 outer [one

per type from eight different plants]) were then allowed to air-dry and the change in load

over time was calculated every five minutes for a total of 220 minutes.

Weight Measurement

The change in weight between rehydrated and dehydrated states of stems was

calculated using a Mettler AJ100 analytical balance. Stems were rehydrated overnight

and excess water was blotted from the stem surface. Individual outer and inner stems were placed upon the balance and allowed to air-dry for 220 and 350 minutes, respectively. The weight loss percentage was calculated with respect to the weight of the

stem at its fully hydrated state as w* = 100 x (wwet - w)/ wwet.

Staining and microscopy

Toluidine Blue O (TBO) used for the detection of lignin and pectin. Spurr’s resin-embedded sections were stained with 0.05% TBO in 0.01 M PO4 buffer (pH 5.7) for 10 s on a hotplate (60°C) and were rinsed with deionised water. Prepared sections were observed with bright field microscopy. Basic fuchsin was used to confirm and detect lignin (Dharmawardhana, Ellis, & Carlson, 1992). Paraffin-embedded sections were stained in 0.0001% basic fuchsin (in 70% ethanol) for 5 minutes, washed in 70% ethanol for 2 minutes, and rinsed briefly in deionised water. Samples were mounted in

50% glycerol and observed with the TX2 (RFP filter set). Prepared sections were viewed

using a Leica DM6000B epifluorescence microscope.

Finite Element Simulations

Finite element (FE) simulations of simplified outer and inner stem models were

performed using the commercial package ABAQUS (Rising Sun Mills, Providence, RI,

USA). The hydro-actuation strain was modelled with thermal expansion, where

temperature represented moisture content. The rod-like stems were modelled as

inextensible elastic beams and were discretised with hybrid quadratic beam elements

(B32H). The curling was induced by introducing thermal strain gradient through the

beam cross section. The spatial variation of hydro-actuation capacity along the stem

length was introduced into the model using an analytic field, which was derived from

curvature profiles.

The FE bilayer model was discretised with elastic plane stress quadratic elements

(CPS6). The stem was clamped at its base. The geometrical nonlinearities were taken into

account, but the self-contact of the stem was neglected. The curling of the stems was simulated by imposing a negative thermal strain in the active layer. A mesh size sensitivity analysis was performed and based on that a mesh with about 7 elements along the stem thickness (~5000 elements) was chosen. This choice led to fairly consistent results in the range of the parameters considered in this work and enabled us to smoothly extract curvature values during the post-processing stage. Further information on the FE model of bilayer stems is provided in Supplementary Information S5.

FIGURES

100

Figure 2.1.

101

Figure 2.1. Morphology and Composition of the Resurrection Plant Selaginella lepidophylla. A plant in (A) desiccated and (B) hydrated states. S. lepidophylla exhibits a spiral phyllotaxy with the youngest stems near the centre of the plant and the oldest stems near the outermost edge of the plant. Curling sequence of fully hydrated (C) outer and

(D) inner stems of S. lepidophylla during dehydration. Both stems exhibit large deformation with distinct curling patterns. (E) Cross section of a toluidine blue O (TBO) stained, Spurr’s resin-embedded inner stem (basal region) showing the location of the cortex (C), xylem (X), phloem (P), and specialized trabeculae cells (T) in the air space

(A) separating the protostele from the cortex. While lignified cell walls (dark blue) are found throughout the inner zone of the cortical tissue, there appears to be a difference of cell density between the adaxial and abaxial zones. Basic fuchsin was used to visualize lignin within the cell walls of the stem tissue. The cross-sections of a basic fuchsin stained, paraffin-embedded, inner stem at (F) the basal section reveal lignification (bright red) in all the ground tissue, whereas (G) the middle of the stem, shows lignified tissues in the abaxial side of the stem. This staining pattern is reduced to a narrow strip at (H) the apical tip of the stem.

102

Figure 2.2.

103

Figure 2.2. Response of Outer and Inner Stems to Dehydration. The normalized

curvature k* = kl as a function of the normalized arc-length s* = s/l of (A) outer and (B)

inner stems of S. lepidophylla at different time intervals. Mechanical response of initially wet stems to dehydration for (C) the outer and (D) the inner stems. Plotted as a function of the weight loss w* of the stem during dehydration, the absolute displacements of the stem tip, ux and uy, and the reaction force of a similar stem clamped at its both ends

illustrate that stem curling correlates with induced-stress buildup. The vertical lines in gray represent the associated dehydration time. (E) Force measurement setup. (F)

Dehydration weight loss w* of outer and inner stems as a function of time. The early drying behaviour before reaching equilibrium is fitted with a power-law function.

104

Figure 2.3.

105

Figure 2.3. Curling Mechanisms of Inner and Outer Stems. (A) The curling patterns

of virtual stems with power-law hydro-actuation strain gradient (κ = ar+1sr) for different r exponents when η = 24. The dotted line shows the initial undeformed state. The deformed shape of the dehydrated (B) outer and (C) inner stems of S. lepidophylla are reproduced using large deformation FE simulations by setting η = κ* and are overlaid (solid lines) on stem images. The curvature of the non-inner stem model is constant (κ* = 3.93), whereas for the inner stem model, the curvature varies linearly with the stem length (κ* =1.6 +

13.64 s*). (D) A simplified FE model for hydro-responsive curling of S. lepidophylla plants showing the cooperative packing of outer and inner stems at different stages of dehydration. The contours represent the absolute rotation of the stems in radian. Note that in actual S. lepidophylla plants, the spiral phyllotaxy would yield a gradient of curvature moving from the centre of the plant. Also, both the properties of the inner stems, and their shading by the more rapidly closing outer stems would slow their dehydration process, thus the timing of the movement of the inner stems would be delayed in real plants.

106

Figure 2.4.

107

Figure 2.4. Curling Pattern of Bilayer Stems. (A) Conformations of bilayer stems (h/l

= 0.02) composed of elastic active (a) and passive (p) layers (Ep/Ea = 2) subjected to an

actuation contraction strain εa = –0.2 obtained with large deformation FE simulations for

layer profiles with linearly varying thickness along the stem length. (B) Normalized

curvature κ* of bilayer models of an outer stem with a constant relative thickness (ha/h = hp/h = 0.5) and an inner stem with a linearly varying relative thickness (ha/h = 0.1 at the base and hp/h = 0.5 at the tip of the stem) under action of shrinkage strains (εa = –0.05 for the outer stem and εa = –0.2 for the inner stem) in the active layer. Results obtained with theoretical Timoshenko model (Eq. (2)) and FE simulations. The inset shows the corresponding axial strain contours of deformed stems.

108

SUPPLEMENTARY INFORMATION

109

S1. Preparation of Spurr’s Resin-embedded Sections

Inner and outer stem sections were fixed in 3% glutaraldehyde in 0.1M PO4

buffer (pH 7.0) for 16 hours at 4°C on a nutator. Samples were then washed 3 x 10

minutes in 0.05M PO4 buffer. Samples were post fixed in a 1% OsO4 solution (in 0.05M

PO4 buffer) for 2 hours at room temperature. Samples were rinsed in deionized water and subjected to an ethanol series (10 minutes in 30% ethanol, 1 hour each in 50%, 70% and

85%, 95% and 2x 100% ethanol) at room temperature. Samples were washed in 100% propylene oxide for 30 minutes, after which they were changed into a mixture of propylene oxide and Spurr's resin (2 hours each of 3:1 PO:Spurr's, 1:1 PO:Spurr's, and

1:3 PO:Spurr's). Samples were left in 100% Spurr's overnight at room temperature. The next day, samples were changed into fresh Spurr's twice and left overnight. Samples were changed into fresh Spurr's the next day and were polymerized in open tubes at 60°C for

48 hours. Samples were sectioned using a Leica EM UC6 Ultramicrotome and were mounted on regular glass slides.

110

S2. Preparation of Paraffin-embedded Sections

Inner and outer stem sections were fixed in FAA (4% formaldehyde, 5% acetic acid and 50% ethanol) for seven days at 4°C on a nutator. The FAA solution was changed for fresh FAA on day 3 of fixation. Stems were then washed twice for 30 minutes each in

50% ethanol 4°C and left overnight in 70% ethanol at 4°C on a nutator. Samples were dehydrated through an ethanol series (1hr each 85%, 95%, 2 x100%) at room temperature. After dehydration, samples were transferred to a mixture of xylene and ethanol (1hr each: 75% ethanol:25%xylene, 50% ethanol:50% xylene, 25% ethanol:75% xylene, 2 x 100% xylene). Samples were placed in fresh xylene in scintillation vials and paraplast chips were added (1:4 volume of xylene) overnight at room temperature. The next day, the vials were placed at 42°C for 30 minutes. Another 1:4 volume of paraplast chips was added and samples were incubated at 60°C for 6 hours. Paraplast:xylene solution was discarded and replaced by molten paraplast(i.e. paraplast chips that had been melted at 60°C for 24 hours). Samples were then left at 60°C for one week and the molten paraplast was changed for fresh solution twice a day.

Samples were embedded in fresh paraplast in small Petri dishes and the paraplast was allowed to harden at room temperature overnight. Samples were sectioned using a Leica

RM22452 Microtome and were mounted on positively charged slides. Samples were deparaffinized prior to staining with 2 x 100% xylene (15 minutes each).

111

(S3) Supplementary Figure 2.1.

112

(S3) Supplementary Figure 2.1. Discrete Curvature Characterization

The curvature of the stems of S. Lepidophylla (Figure 2A-B) are characterized by accurate estimations of the curvature of a smooth curve from its discrete approximation.

Supplementary Figure 2.1. shows a segment of a smooth curve which is represented by a polyline with five points P1 to P5, with the corresponding edges PiPi+1 (i = 1, 2, 3, 4) denoted by c, d, e and f, and their lengths are c, d, e and f. The curve can be represented by a Taylor series expansion (Langer et al., 2005). A linear approximation for the true curvature vector k can be obtained by finite difference approach and if all edges have equal length, the convergence is even quadratic.

113

S4. Geometrical Model Based on Euler Spiral

Towards rationalizing the spiralling behaviour of the living stems, we adopt a geometrical model based on the definition of the normalized Euler (Cornu) spiral. By definition, an Euler spiral is a curve whose curvature k changes linearly with its curve length s, i.e. k = a2s where a is a constant. This definition can be generalized to consider the role of material variation along the stem length. For a general class of power-law curvature (e.g., induced by the functionally graded hydro-actuation capacity of the tissue) defined by the Cesàro equation k = ar+1sr, the parametric equations for the spiral profile read:

~ where s ∈ [0, a], a ∈ [0, η] and pFq(ap; bq; z) is the generalized hypergeometric function

(Askey & Daalhuis, 2010). These equations were evaluated in the computational software

Mathematica (Wolfram) using the built-in function HypergeometricPFQ. The simplified model above is used to investigate the role of material variation along the stem length as illustrated in Figure 2.3A in the article.

114

S5. Finite Element Model for Bilayer Stems

Finite element (FE) simulations of bilayer stem models performed using the commercial package ABAQUS 6.11 (SIMULIA, Rising Sun Mills, Providence, RI,

USA). A Python script is written to systematically create stem models. The bilayer is composed of a soft active (a) and a stiff passive (p) elastic layer, which have respectively the elastic moduli of Ea and Ep and the actuations strains of Ea and Ep. The Poisson’s ratio

for both layers is νa = νp = 0.3. The hydro-actuated strain is modelled with thermal expansion, where temperature represents moisture content. The stem is clamped at its base. Geometric nonlinearities are taken into account by activating NLGEOM option in

ABAQUS, which allows for large-deformation analysis. A mesh size sensitivity analysis is performed and based on that a mesh with about 7 elements along the stem thickness (~

5000 triangular plane stress quadratic elements, CPS6) gives consistent results in the range of the parameters considered in this work. Supplementary Figure 2.2. shows the mesh for a bilayer stem in its deformed state. This mesh resolution allows us to characterize curvature smoothly along the stems centerline following the procedure introduced in S3.

In Supplementary Figure 2.3, the FE predictions of the normalized curvature of bilayer models for constant and varying thickness ratios are compared to those obtained by the Timoshenko bimetallic theory (Timoshenko, 1925) at different actuation strains.

For both cases, at small actuation strains, FE results are in very good agreement with theory; however, as the actuation strain increases the FE results deviates from the

Timoshenko bimetallic model, which is derived, based on small deformation assumption.

In FE simulations, we have taken into account geometric nonlinearities, which allows

115 nonlinear analysis of stems under large deformation, as observed in this work. Therefore, while Timoshenko model is still a fairly good model, it is not accurate for large actuation strains. According to Eq. (2) in the article, the predicted curvature of Timoshenko bimetal model scales linearly with actuation strain. In contrast, FE results suggest that curvature does not magnify linearly with actuation strain and the location of maximum curvature shifts as the actuation strain increases. To summarize, to model the large deformation induced by stem curling upon hydration, we perform multiple simulations that account for geometric nonlinearities. The bimetallic Timoshenko model was introduced as a limiting case, which – exact and sufficient for small deformation – still provides quite reasonable predictions for large strains.

116

Supplementary Figure 2.2.

117

Supplementary Figure 2.2. Finite element mesh for a bilayer stem. A bilayer stem (h/l

= 0.02, ha/h = 0.1 at base and = 0.5 at tip) is meshed with 4857 triangular plane stress quadratic elements (CPS6) in ABAQUS.

118

Supplementary Figure 2.3.

119

Supplementary Figure 2.3. Comparison between FE simulations and theoretical

Timoshenko bimetallic model for normalized curvature of bilayer stems. (a) constant thickness and (b) variable thickness.

120

LINK BETWEEN CHAPTER 2 AND 3

Chapter 2 characterizes the deformation patterns observed in the resurrection

plant Selaginella lepidophylla. Two stem types – inner and outer, corresponding to

developing and mature stems respectively – display different deformation and curvature

patterns, rates of deformation, and corresponding differences in mechanical properties

upon dehydration. Inner stems curl tightly upon drying, show lower contractile force

during dehydration, possess linear curvature in a dry conformation, and require six hours

to alternate between hydrated and dehydrated conformations. In contrast, outer stems do

not curl as tightly (form an arc shape), show higher contractile force during dehydration

than inner stems, possess constant curvature along the length of the stem in a dry

conformation, and alternate between wet and dry conformations in only one hour.

Preliminary investigation at the tissue level indicated that the different patterns of

deformation might result from differences in the degree of stem lignification between

inner and outer stem types, and that deformation in both stem types might be a result of

changes in apparent cell density between the two stem sides (adaxial versus abaxial).

Chapter 3 explores the inherent properties of S. lepidophylla stems that allow stems to change shape, and what properties allow stems to curl to different degrees. The investigation compares the properties of inner and outer stems types at the tissue level, then focuses on inner stem cell walls to examine specific features of the cell wall leading

to the deformation patterns observed at the organ level.

121

CHAPTER 3

Morphological and Compositional Gradients Direct Stem Curvature and Deformation in

the Resurrection Plant Selaginella lepidophylla

122

ABSTRACT

Various bio-inspired materials have been derived from studying stimuli-responsive deformation in the plant kingdom. Of the plant models studied, desiccation tolerant resurrection plants are intriguing because they exhibit reversible, hierarchical water- responsive deformation. Here, we explore morphological and compositional properties at tissue and cell wall levels in the desiccation tolerant spikemoss Selaginella lepidophylla that lead to linear curling in inner rosette stems and constant curvature in outer stems.

Directional bending in both inner and outer stems is associated with cross-sectional gradients of tissue density, cell orientation and secondary cell wall composition.

Deformation towards the upper side of the stem results from a juxtaposition of cells of varying stiffness and angle between the adaxial and abaxial stem sides. In inner stems, a gradient of adaxial cortex secondary cell wall thickness, as well as a gradient of secondary cell wall composition (lignin, hemicellulose), affect tip to base stiffness and swelling/shrinking, allowing for more complex curling as compared to outer stems.

123

INTRODUCTION

The plant kingdom has long been a source of inspiration for biomimetic materials

and actuating devices (Burgert & Fratzl, 2009a; Holstov, Bridgens, & Farmer, 2015;

Moulia, 2013; Naleway, Porter, McKittrick, & Meyers, 2015; Paris, Burgert, & Fratzl,

2010). Various plants and plant organs, such as the pinecone, Dionaea muscipula (Venus

flytrap), and Mimosa pudica, undergo a set of well-defined shape transformation as part

of a physiological response to environmental stimuli (Burgert & Fratzl, 2009a; Reyssat &

Mahadevan, 2009). Deformation allows plant species to carry out various functions,

including seed dispersal, predation, and predator evasion (Burgert & Fratzl, 2009a; Rivka

Elbaum & Abraham, 2014). In resurrection plants, species that can tolerate extreme

drought conditions, deformation is essential for survival. The vegetative tissue of

resurrection plants reversibly deforms in response to changes in relative water content

(Dinakar et al., 2012; J. M. Farrant & Moore, 2011; Morse et al., 2011). Curling of stems

and folding of leaves as a result of desiccation limits the amount of photo, thermal and

water-deficit stresses the resurrection plant is exposed to during periods of drought

(Dinakar et al., 2012; Lebkuecher & Eickmeier, 1993; Morse et al., 2011).

Selaginella lepidophylla, a resurrection spikemoss native to Northern and Central

America, is composed of hundreds of stems connected together by an extensive root system (Lebkuecher & Eickmeier, 1991, 1993). These stems are arranged in a spiral

phyllotaxy with developing (inner) stems at the center of the plant, and sequentially more

mature (outer) stems spiraling outward from the center (Figure 3.1A). When hydrated, S.

lepidophylla stems are completely uncurled and the plant appears as a flattened rosette.

124

Upon dehydration, stems curl and the whole plant deforms into a spherical shape, with outer stems curling over inner stems (Brighigna et al., 2002; Rafsanjani et al., 2015).

Inner and outer S. lepidophylla stems curl to different degrees that, in combination with a spiral phyllotaxy, allow for tight and precise stem packing during desiccation-induced deformation. Preliminary investigation at the tissue level showed that asymmetric cell density and lignin distribution might be responsible for the different degrees of curling and mechanical responses exhibited by inner and outer stem types (Rafsanjani et al.,

2015).

Here, we adopt a broad experimental approach to address two related questions:

(1) what properties are primarily responsible for directional stem deformation, and (2) how do these properties contribute to the different degrees of curling observed in inner and outer S. lepidophylla stem types? We take advantage of an array of techniques to explore morphology (micro-computed x-ray tomography and transmission electron microscopy), composition (brightfield and fluorescence microscopy), and mechanical properties (microtensile testing and nanoindentation) at the tissue and cell wall levels leading to deformation in S. lepidophylla.

125

RESULTS

Mechanical Properties of S. lepidophylla Stems

The degree of curling of individual S. lepidophylla stems varies with their location in the plant’s spiral phyllotaxy (Figure 3.1A-B) (Rafsanjani et al., 2015). Time- lapse observation of a wedge-shaped cross-section of a dehydrating plant reveals that inner stems curl slowly into tight spirals, whereas outer stems rapidly curl into an arc shape (Figure 3.1B). Given these specific curling profiles, we expect the mechanical forces involved in deformation to also vary between inner and outer stem types. We performed uniaxial tensile tests on hydrated inner (Figure 3.1C) and outer stems (Figure

3.1D), with outer stems being divided between those that have lost their leaflets

(microphylls) and those that retained them, as leaflets can influence surface water movement along the stem (data not shown). Mechanical properties (described in Box 1 of

Appendix 1) were calculated from stress-strain plots obtained during testing (Table 3.1).

Of particular interest is the property of Young’s Modulus (E), which measures the ability of a material or object to resist deformation under load (Brulé et al., 2016; Mirabet, Das,

Boudaoud, & Hamant, 2011). High E values denote stiff materials that resist deformation, while low values are common for elastic materials that readily deform under load.

Mechanical testing of S. lepidophylla stems revealed that outer stems are significantly stiffer than inner stems (Table 3.1, Figure 3.1E).

During dehydration, S. lepidophylla stems curl towards their adaxial or upper side

(as opposed to their lower or abaxial side) (Figure 3.1B-D). As this movement can be computationally mimicked with a simple bilayer model (Rafsanjani et al., 2015), we

126 hypothesized that adaxial and abaxial stem sides differ in their relative stiffness.

Hydrated, inner stems were cut lengthwise, and adaxial and abaxial sides were subjected to uniaxial tensile testing (Table 3.2). The abaxial region was significantly stiffer than the adaxial region (Figure 3.1F). As a control, inner stems were cut lengthwise into left and right sides and tested. No significant difference in stiffness was observed (Table 3.2).

Due to the brittleness of outer stems, they could not be cut to be similarly tested.

Thus, at the organ level, the degree of stem curling observed in inner and outer stems appears to be associated with differences in stiffness (i.e., inner stems are less stiff than outer stems). In addition, a stiffness gradient between adaxial and abaxial stem sides seems to contribute to directional bending in inner stems, whereby the stem curls toward the less stiff (adaxial) side. Given the possible contribution of these stiffness gradients to the direction and extent of stem curling, we decided to investigate the underlying features leading to differences in stiffness between inner and outer stems, and also between adaxial and abaxial stem sides.

Morphological Differences Between Adaxial and Abaxial Stem Tissue

S. lepidophylla stems are composed of four main tissue types (Figure 3.2A).

Microphylls are attached to a thin epidermal layer covering a thick cortical tissue layer.

The cortex is shaped like a hollow cylinder surrounding an amphicribral vascular bundle that runs through an airspace in the center of the stem. The vascular bundle is connected to the cortex by large, thin-walled trabeculae cells. As the bulk of the stem is made up of cortex, we focused on comparing the morphology of this tissue between inner and outer stems, and between adaxial and abaxial stem sides.

127

Light microscopy revealed two differences in adaxial and abaxial cortical tissue

across all observed regions (apical, middle and basal) of inner and outer stems. First, the

adaxial cortex is thicker than the abaxial cortex (Figure 3.2B; (Rafsanjani et al., 2015)).

This was also observed when recording stem dimensions of adaxial and abaxial regions

for mechanical testing (Supplementary Table 3.1). Second, the abaxial cortex appears denser (more cells per square area) when compared to the adaxial cortex (Figure 3.2B

insets).

Transmission electron microscopy (TEM) was performed to determine if cell size

and shape differed between adaxial and abaxial cortex. Average cross-sectional cell area,

cell wall area, and lumen area were calculated for apical, middle and basal inner stem

regions. Adaxial cortical cells had significantly larger total cell area and lumen area

across all stem regions when compared to abaxial cortical cells (Table 3.3, Figure 3.2C-

D). In contrast, cell shape did not change significantly between adaxial and abaxial

cortical cells, or across cells in apical, middle and basal stem regions (Figure 3.2C-D).

Three-dimensional stem tissue morphology was assessed using high-resolution

synchrotron radiation phase-contrast X-ray tomographic microscopy, and three-

dimensional image analysis. Four radial sections were analysed across apical and basal

regions of inner stems: adaxial, abaxial, left and right (Figure 3.2E-F). In each region,

which stretches from the periphery toward the centre of the cortex, we determined the

orientation of the cells with respect to the long axis of the stem as well as the cell lumen

volume. While there was little difference between left and right sides of the stem cortex

in terms of three-dimensional structure, changes were seen comparing adaxial and abaxial

sections (Figure 3.2E-H). First, abaxial cortical cells are aligned parallel with the

128

longitudinal stem axis, whereas adaxial cortical cells range in angle from 30-55° relative

to this axis. Second, there are alterations in lumen volume. Both adaxial and abaxial

cortex cells have a range of lumen volumes (Figure 3.2F, H). While abaxial cells have larger volumes in general (Figure 3.2F), they are also longer than adaxial cells, making it hard to assess cross-sectional tissue density between the two stem regions (Figure 3.2F).

Thus, lumen volume was normalized over cell length, revealing that, while there is a

range of lumen sizes in both adaxial and abaxial cortex, adaxial cells are generally larger

in cross-sectional lumen area, as seen with TEM (Figure 3.2C-D; Table 3.3). This is also

consistent with the lower cell density observed in adaxial tissue with light microscopy

(Figure 3.2B).

Taken together, structural differences exist between adaxial and abaxial stem

sides, and are observed from tip to base in both inner and outer stems, though only the

apical region of inner stems is shown here. Given the distinctive tissue structure of

adaxial and abaxial cortex, including changes in cell size and angle, we decided to

examine adaxial and abaxial cortical cell walls, to identify whether or not there were any

differences in adaxial and abaxial stem sides visible at the cell wall level.

Gradients of Secondary Cell Wall Properties Along the Length of Inner Stems

Cell wall thickness is a known factor that affects plant stiffness (Brulé et al.,

2016; Gibson, 2012). Quantification of cell wall thickness using TEM revealed that,

between adaxial and abaxial cortex cells along the length of the stem, abaxial cell walls

are significantly thicker in apical cells, but that this difference between stem sides is lost

in the middle and basal portions of inner stems (Table 3.3). Topological scans of hand-

129

cut, dried inner apical sections using atomic force microscopy (AFM) revealed a similar

change in cell wall thickness between adaxial and abaxial cortical cells as observed with

TEM (Figure 3.3A). These AFM scans also showed similar results to those seen with

TEM with regard to cell shape and lumen size (compare Figure 3.2C-D with Figure

3.3A). In addition to examining cell morphology, nano-indentation was performed to

explore cell wall stiffness. As seen at the gross level for longitudinally-cut whole stems,

apical adaxial cell walls are significantly less stiff than apical abaxial cell walls

(~340MPa and ~870MPa respectively, p<0.05).

Cell wall stiffness is also affected by types, quantities, localization and

interactions between various wall polymers. These include not only cellulose, but also the

matrix polysaccharides (pectins and hemicelluloses) and the polyphenolic lignin

(Cosgrove & Jarvis, 2012; Mirabet et al., 2011; Speck & Burgert, 2011). The presence

and distribution of these polymers was examined using a combination of fluorescence

microscopy and immunohistochemistry. Lignin has previously been shown to be present

in S. lepidophylla stems, with differences not only between adaxial and abaxial cortex,

but also across apical, middle and basal inner stem regions (Appendix 2; (Rafsanjani et

al., 2015; Weng et al., 2010)). At the stem apex, basic fuchsin staining detects lignin in the abaxial cortex near the stem periphery; in the stem middle, lignin is observed throughout the abaxial cortex; and at the stem base, both adaxial and abaxial cortex are uniformly lignified (Supplementary Figure 3.1). Lignification in outer stems was consistent from stem tip to base, and was distributed throughout both adaxial and abaxial cortical tissue (Rafsanjani et al., 2015).

130

To evaluate the presence, location and amount of pectic and hemicellulosic

polysaccharides, inner S. lepidophylla stems were subjected to immunostaining with a

battery of pectin and hemicellulose-detecting antibodies. Of the eleven pectin and nine

hemicellulose antibodies used, ten bound to S. lepidophylla sections. In most cases, the

antibodies either bound only to the vascular bundle and/or showed uniform binding

across adaxial/abaxial cortex and/or along the stem (Supplementary Table 3.2). However, two antibodies detecting secondary cell wall hemicellulose, LM10 and LM11 (Harholt et al., 2012; McCartney, Marcus, & Knox, 2005), differed in epitope binding both between adaxial and abaxial cortex, and along the apical-basal stem axis. LM10

(unsubstituted/low substituted xylan) and LM11 (unsubstituted/low/highly substituted xylan) have overlapping binding patterns in S. lepidophylla stems (Figure 3.3;

Supplementary Figure 3.2). In apical regions, both antibodies bind strongly to adaxial cells and only weakly to abaxial cells. While antibody binding to adaxial cells decreases to a certain extent in middle and basal stem cross-sections, it still remains significantly higher than that seen in the abaxial side (Figure 3.3C-D; Supplementary Figure 3.2).

LM10 and LM11 antibodies show a similar binding pattern in the adaxial and abaxial cortex of both inner and outer stems. However, outer stems do not show decreasing tip- base antibody binding. Rather, binding remains consistent among apical, middle and basal cross-sections and is similar to that seen in basal sections from the inner stem

(Supplementary Figure 3.3).

131

DISCUSSION

In response to drought, the stems of the spikemoss S. lepidophylla curl over themselves as a physiological response to reduce light-induced oxidative stress

(Lebkuecher & Eickmeier, 1991, 1993). Bending occurs toward the adaxial stem side, and the degree of curling varies between inner and outer stems of the rosette: inner stems curl into a spiral shape with linear curvature, while outer stems deform into an arc with constant curvature (Rafsanjani et al., 2015). While this mode of actuation can be reproduced through a simple bilayer model, the structural properties responsible for the direction of bending and the degree of curling are poorly understood. Using a combination of light, electron and atomic force microscopy, as well as three-dimensional

X-ray tomography, our results suggest that directional curling is a result of structural gradients across the stem cross-section, from adaxial to abaxial sides. In the case of inner stems that show a more complex curling pattern, gradients also exist along the length of the stem.

Directional Deformation of S. lepidophylla Stems is Associated with Differential

Tissue Morphology and Cell Wall Properties

Mechanical testing of dissected adaxial and abaxial (but not left and right) sides of S. lepidophylla stems revealed that the adaxial portion of the stem is less stiff (Figure

3.1F). This is consistent with the proposed bilayer model that supposes the juxtaposition of two layers with differential swelling (shrinking) in response to stem hydration

(dehydration). The active layer generally acts as the driver for tissue/organ movement,

132

while the passive layer constrains and controls the direction of movement (Burgert &

Fratzl, 2009a; Dunlop et al., 2011; Rivka Elbaum & Abraham, 2014; Erb et al., 2013;

Forterre, 2013; Reyssat & Mahadevan, 2009). All things being equal (i.e., identical

cylindrical cells with hoop-type cellulose reinforcement leading to cell elongation

primarily in the longitudinal axis), the less stiff adaxial tissue would comprise the active

tissue whose greater swelling/shrinking in response to hydration status would push or pull

the stiffer, passive abaxial tissue into a different conformation. What makes the abaxial

tissue stiffer? Are there other structural properties that can influence this directional

bending? Detailed analysis of the cortical tissue of S. lepidophylla at several length-scales and in both two and three dimensions suggests that there is a complex morphological and biochemical hierarchy involved.

Observed structural differences between adaxial and abaxial cortical tissue in S. lepidophylla can be divided into cell size and shape, secondary cell wall composition, and cell angle with respect to the primary stem axis (Figure 3.4). While the abaxial cortical cells are elongated, thick-walled cylinders with relatively narrow lumens, adaxial cells tend to be shorter and wider (larger cross-sectional lumen area). This leads to higher tissue density in the abaxial cortex, which can contribute to increased stiffness.

Conversely, immunohistochemistry revealed that across stem types, adaxial cells have a greater proportion of hemicellulose xylans, which could contribute to increased swelling of the adaxial cortex (see below for more discussion of cell wall composition). In addition, adaxial cells are set at an angle (30-55°) to the primary stem axis, while abaxial cells lie parallel to the longitudinal axis. This establishes differential swelling/shrinking angles between the two sides of the stem, leading to bending. Put together, in response to

133 changes in water status, the adaxial cortex would be expected to shrink/swell on an angle and, given larger lumen cross-sectional area and lower tissue density, to a greater extent than the abaxial cortex. This would result in pulling towards the adaxial side with dehydration (shrinkage), and thus adaxial bending, and pushing towards the abaxial side with hydration (swelling) leading to straightening of the stem.

Lengthwise Gradients of Cell Wall Thickening, Lignification and Hemicellulose are

Associated with Linear Curling in Inner S. lepidophylla Stems

The younger inner stems of S. lepidophylla curl into a tight spiral upon dehydration, unlike the arc-shape seen for older outer stems (Rafsanjani et al., 2015).

Structural, immuno- and histochemical analysis suggests that the tissue morphology is identical between inner and outer stems, and that the base of inner stems is equivalent to the whole length of the outer stem. What differs between stem types are the characteristics of the cell walls along the length of the inner stems. Namely, tip to base gradients are seen in cell wall thickness and composition. In the apical region of inner stems, abaxial cortex cells have thick secondary cell walls, while adaxial cell walls are thinner. By the middle of the stem, this difference is lost (Table 3.3). Compositionally, there is a gradient of lignification along the inner stems in which lignin is most strongly detected along the outer curve of the apical abaxial cortex, throughout the whole abaxial cortex mid-stem, and across the entire (adaxial and abaxial) cortex at the stem base

(Figure 3.4; Supplementary Figure 3.1). Conversely, antibody detection of hemicellulose xylans reveals an opposite gradient (Figure 3.3; Figure 3.4; Supplementary Figure 3.2).

Xylans stain strongly in the adaxial portion of the cortex at the apical region and become

134 less apparent lower in the stem (they are consistently detected at a low level in the abaxial cells). Considered from a developmental viewpoint, secondary cell wall differentiation could be seen as delayed in the adaxial cortex.

Both secondary cell wall thickness and composition affect cell and tissue stiffness

(Cosgrove & Jarvis, 2012; Gibson, 2012). In walls of equal composition, thicker walls act to strengthen and stiffen. The deposition of lignin plays a similar role. A polyphenolic polymer, lignin coats the polysaccharides and fills in the pores of cell walls, making them both stiffer and hydrophobic (Gibson, 2012; Grabber, 2005). Thus, lignified tissue is less elastic and less able to swell/shrink in response to changes in hydration status. The significance of the gradient of xylan detection is less obvious. It is possible that it is present throughout all cortical cell walls in an equal amount, and the strong binding of the antibodies seen in the apical adaxial region of inner stems (and to a lesser extent in all adaxial regions) is a reflection of decreased levels of lignin in these regions – i.e., the presence of lignin is masking the hemicellulose epitopes. Barring antigen masking by lignification in abaxial tissue, a possible explanation for the pattern of xylan staining is that xylan is acting as a plasticizer to promote tighter and reversible cell wall compaction during dehydration, as suggested for other resurrection plant species (Moore et al., 2013;

Moore, Vicre-Gibouin, et al., 2008). If it is acting as a plasticizer in S. lepidophylla, xylan could be binding to cellulose fibrils during dehydration to replace hydrogen bonds that cellulose previously formed with water. This would prevent cellulose microfibrils from binding to each other, thereby allowing for reversible cell wall compaction leading to more tissue shrinking and larger deformation at the organ level (Busse-Wicher,

Grantham, Lyczakowski, Nikolovski, & Dupree, 2016; Grantham et al., 2017; Moore et

135 al., 2013). Having a higher abundance of xylan in adaxial cortex makes sense with this stem side’s role as an active layer that pushes and pulls the abaxial, passive layer during stem deformation. A decreasing tip to base gradient of xylan abundance in adaxial cortex also fits with the observed pattern of tight curling at the stem tip, and less curling at the stem base. This seems like a reasonable function for xylan with respect to S. lepidophylla stem curling because outer stems show lower adaxial xylan abundance and consistent xylan staining from tip to base, and they are unable to curl to the same degree as inner stems. Further testing using Raman confocal spectroscopy (attempted analysis shown in

Appendix 3) to quantify xylan and lignin abundance and map their spatial distribution in adaxial and abaxial tissue is needed to more precisely determine the presence of xylan in

S. lepidophylla cortical cell walls and its function in stem deformation.

Whether or not xylan is acting as a plasticizer, the higher accessibility of hydrophilic hemicellulose epitopes in adaxial tissue suggests that it is more available for water binding and able to swell/shrink (Busse-Wicher et al., 2016). This is consistent with the reduced stiffness of adaxial cortex secondary cell walls when probed with AFM

(Figure 3.3B). The sum of these gradients suggests that inner stems have a stiffness gradient running from tip to base such that they are less stiff at the tip where the adaxial cortex cell walls are thinner and lignification is only significant on the most abaxial edge of the cortex, and that they become progressively stiffer moving toward the stem base where the adaxial cell walls thicken and become lignified. In addition to stiffness, the apical adaxial cells would be expected to have increased differential swelling/shrinking.

Together these features would allow for tighter curling at the tip and progressively less curling moving downward, resulting in a spiral shape with stem drying.

136

Comparison of Deformation in S. lepidophylla to Other Established Plant Models

Bilayer structure is observed in a number of water-driven actuating plant species, including – but not limited to – pinecones (Dawson et al., 1997; Fratzl & Barth, 2009;

Reyssat & Mahadevan, 2009), wheat awns (R. Elbaum et al., 2008; R. Elbaum et al.,

2007; Erb et al., 2013), and orchid tree seedpods (Armon et al., 2011; Erb et al., 2013).

The juxtaposition of active and passive tissue layers in these species gives rise to differential tissue swelling and shrinking, leading to movement. However, unlike these species with very distinct bilayers, in S. lepidophylla there is an added degree of complexity in that the bilayers are graded, either from stem tip to base or between stem sides (adaxial/abaxial). In this aspect, S. lepidophylla stems resemble bamboo; both show transverse and longitudinal gradients leading to specific mechanical behaviours to counteract environmental stresses imposed upon the plant. In bamboo, cellulose fibers confer mechanical support to the culm. Fibers are arranged in a radial pattern, with increasing fiber density moving from the center to the periphery of the culm and a corresponding stiffness gradient (low to high stiffness from culm center to periphery)

(Habibi, Samaei, Gheshlaghi, Lu, & Lu, 2015; Tan et al., 2011). This gradient gives bamboo its characteristic flexural response, allowing it to resist bending in high winds. In

S. lepidophylla, rather than a radial pattern, stiffness changes from stem tip to base and between adaxial and abaxial sides, giving rise to the specific curling patterns of inner stems. Thus, it is interesting to observe how changing the geometry of functional gradients can give rise to different mechanical and deformational behaviours. This is a useful consideration for building actuating devices with complex shape change and mechanical responses.

137

The combination of tissue and cell wall gradients in S. lepidophylla also highlight

appealing features that could be deliberately used to produce synthetic actuators with

superior functionality. Careful incorporation of spatial inhomogeneity into active bilayer

systems can confer characteristics that can direct water-responsive conformational

changes. Directional deformation attained through this basic strategy could be integrated

with other paradigmatic concepts, such as biomimetic origami, to potentially generate

different conformational states depending on the type or level of stimulus (e.g., water)

applied to the structure (Harrington et al., 2011; Ionov, 2011, 2014; S. Li et al., 2017).

Curvature changes, shape-shifting, and dimensional transformations can serve multiple

sectors, where the requirements of folding, packaging, and deployment are paramount,

such as in aerospace components (e.g., self-deploying satellites) (Huang et al., 2013;

Srinivasan et al., 1991), self-folding medical devices and drug delivery systems (e.g.,

drug release) (Green et al., 2016; Martel, Mohammadi, Felfoul, Lu, & Pouponneau,

2009), and architectural design of environmentally-responsive buildings (e.g., self-

opening windows) (Holstov et al., 2015; Reichert et al., 2015). In terms of

implementation, computational models would help decipher the interaction between

morphological and compositional gradients, and how these features could be an asset for

the design of synthetic actuating systems (Hößl et al., 2007; Vergauwen, De Laet, & De

Temmerman, 2017; Zickler et al., 2012). This would be a starting point for the realization

of proof-of-concepts prototypes for novel bio-inspired materials with uses in a variety of applications (Kempaiah & Nie, 2014; Naleway et al., 2015; Stoychev et al., 2012; Vaia &

Baur, 2008).

138

139

MATERIALS AND METHODS

Plant Materials

S. lepidophylla were acquired and maintained as described in (Rafsanjani et al.,

2015).

Time-lapse Video Capture

Time-lapse video capture for Supplementary Video 1 (not shown in thesis) was adapted from the procedure described in (Rafsanjani et al., 2015). A wedge-shape portion of a representative, hydrated S. lepidophylla plant was isolated and allowed to air-dry to a fully dehydrated state. The wedge was then allowed to rehydrate over the course of six hours and changes in stem deformation were recorded as described in (Rafsanjani et al.,

2015).

Stem and Tissue Tensile Testing

20 S. lepidophylla plants were rehydrated to 100% relative water content. For whole stem tests, 75 stems were isolated randomly from these 20 plants: 25 inner stems,

25 outer stems with microphylls and 25 outer stems without microphylls. For adaxial/abaxial region tests, 50 inner stems were isolated randomly and cut lengthwise

(25 adaxial/abaxial, 25 left/right stem sides) and the vascular bundle removed. Stems were secured between clamps of an ADMET MicroEP machine with the base of the stem always clamped at the load cell end. A 10lb load cell was used for testing. Stems were tested in a hydrated state. Stems were pulled at a rate of (10mm/min) until failure. Stem thickness, width and length were measured prior to testing. Load and displacement were recorded using MTESTQuattro software.

140

Light Microscopy

Five, fully hydrated S. lepidophylla stems were isolated from three different

plants and embedded in polyethylene glycol (PEG) using the protocol from (Gierlinger,

Keplinger, & Harrington, 2012). Embedded samples were then sectioned (10µm) using a

Leica RM2245 semi-automated rotary microtome. Solidified PEG was then removed

using washes of ddH2O. One set of samples was mounted, unstained, and the set was

stained with Toluidine Blue O following the protocol (Peterson, Peterson, & Melville,

2008). Samples were mounted in ddH2O and slides were sealed with nail polish to

prevent water from evaporating. Samples were examined using a Leica DM6000B epifluorescence microscope with the brightfield setting (10x and 40x), and images were acquired using a Qimaging Retiga CCD camera operated through Openlab.

Transmission Electron Microscopy

Ten inner and ten outer stems were isolated from five hydrated plants for both inner and outer regions. 2mm-long sections corresponding to apical, middle, and basal regions of the stem were cut from the ten samples. Five replicates from each stem region were immediately placed in a fixing solution, and another five replicates were allowed to air-dry overnight to a completely dried state. Stem sections were fixed and embedded in

Spurr’s Resin following the protocol outlined in (Rafsanjani et al., 2015). Sample blocks were then thin-sectioned on a Leica EM UC6 Ultramicrotome using an Ultra 45

DiaTOME knife (clearance angle of 6 degrees, speed of sectioning 0.8mm/s, feed 70nm).

Sections were transferred to copper, formvar-coated grids (200mesh or 0.4 x 2mm slotted) and allowed to air dry. Imaging was performed at the Facility for Electron

141

Microscopy Research at McGill University using the FEI Tecnai 12 BioTWin 120kV

TEM, equipped with an AMT XR80C CCD camera system.

Micro-Computed X-Ray Tomography

The tridimensional anatomy and microstructural features of freshly cut inner and

outer stems of S. lepidophylla are acquired by performing a series of synchrotron

radiation-based phase contrast X-ray tomographic microscopy (srPCXTM) experiments,

a non-destructive and high spatial resolution imaging technique at the TOMCAT

beamline (Stampanoni et al., 2007) of the Swiss Light Source at the Paul Scherrer

Institute (Villigen, Switzerland). The TOMCAT beamline exploits a 2.9T magnetic

dipole with a critical energy of 11.1 keV. A double crystal multilayer monochromator is

used to select X-rays with central X-ray photon energy of 15 keV. The X-rays, after

interacting with the sample, are converted into visible lights using a LuAG:Ce 20 mm

scintillator. This light passes through a 10´ optical microscope, which is reflected in a

mirror and finally captured on a 2560x2160 pixels sCMOS camera (PCO.Edge 5.5). As

the woody tissue of the stems of S. lepidophylla is composed of elements with small

atomic numbers, i.e. carbon, oxygen and hydrogen, it has a low X-ray attenuation

coefficient. Therefore, phase contrast tomography method was used in which the samples

are subjected to a very low dosage of X-ray beam energy. Here, the source of phase contrast is the difference in the X-ray index of refraction caused by density gradient at the material-air interface. During each tomography run, 1701 projections were collected over

180° at an exposure time of 120ms per image, resulting in a total scanning time of

approximately four minutes. The simultaneous phase and amplitude retrieval algorithm

142

was used for tomographic reconstruction (Paganin et al., 2002). After reconstruction, a

tomographic dataset consists of 2160 slices stacked at one-pixel interval along the axial

direction and each cross-sectional slice has 2560x2560 pixels. With an isotropic voxel

size 0.65mm3, the approximate field of view is about 1664x1664x1404 mm3.

Atomic Force Microscopy

A JPK Atomic Force Microscope (JPK Nano-wizard@3 Bio Science, Berlin,

Germany) was used for imaging and force spectroscopy. To prepare the samples for

Atomic Force Microscopy (AFM), S. lepidophylla stems were cut transversally by hand

and placed on double-sided clear tape on a microscope slide. Cortical stem tissue in

adaxial and abaxial regions was located to perform force measurements. All the

measurements were performed on tissue in a dry state. Using the QI imaging mode of the

JPK AFM, a force map was created within the area of 30 µm2 on the sample. For consistency considerations, only the points located on the selected parts were indented. For contact mode imaging, super-sharp standard Force Modulation Mode

Reflex Coating (FMR) cantilevers with diamond-like carbon nano-tip of radius 2-3 nm were used. For indentation measurements, Non-Contact High Resonance

(NCHR) cantilevers (Nanotools USA LLC, Henderson, NV) with a nominal spring constant of 40 N/m and integrated spherical tip of radius 50 nm (+/-10%) and 100 nm

(+/-10%) were applied. The indentation frequency was 250 Hz. The deflection sensitivity of the piezo module was obtained by probing the surface of the glass substrate. A thermal tuning method was then used to calibrate the stiffness of the cantilever. The indentation was repeated at the same location for consistency as well to ensure that the sample was

143 not permanently deformed. The indentation depth depends on the applied load, as well as the stiffness of the tip and that of the sample. The elastic modulus of the sample was estimated from the approaching force-indentation depth curve, according to the Hertzian contact model. AFM data analysis was performed with the native JPK data processing software. Statistical significance was determined by a paired student's t-test, when applicable. Differences were considered significant at p<0.05.

Immunohistochemical Labelling

Ten inner and ten outer stems were randomly isolated from five hydrated S. lepidophylla plants, and 2mm sections were cut from the apical, middle and basal regions of each stem. Stems were fixed in 4% formaldehyde in 50mM PIPES at 4°C for one week and subsequently embedded in London-Resin (LR) White (using an ethanol dehydration series and gradual LR White infiltration) following the protocol from (Young et al.,

2008). LR White-embedded samples were then semi-thin sectioned (500nm, feed of

25mm/s) using a Leica EM UC6 Ultramicrotome. Sections were placed onto Teflon- coated slides (EMS #63424-06), and allowed to air-dry on a covered slide-heater at 40ºC overnight. Samples were immunolabelled following a protocol adapted from (Young et al., 2008). Antibodies used are described below in the section “Primary and Secondary

Antibodies”.

Samples were incubated at room temperature in a blocking solution [5% (w/v) normal goat serum (NGS) in 1x Tris-buffered saline/0.2% Tween (v/v) (TBST)] in a homemade humidity box for 40 minutes. Blocking solution was washed off using 1x

TBST. Slides were then incubated with primary antibodies at 1:10 dilution (v/v) in 1%

144

(w/v) NGS blocking solution in the humidity box for 1 hour. Slides were washed 2x 20 minutes in 1x TBST. Secondary antibodies were diluted at 1:100 (v/v) in 1% (w/v) NGS blocking solution in the dark for 45 minutes. Slides were washed 2 x 20 minutes in the dark and mounted in 90% (v/v) glycerol. Control slides were used to test the specificity of the secondary antibodies and also to test for autofluorescence. Blocked slides that were not incubated with primary or secondary antibodies were imaged, as well as slides blocked and incubated with only secondary antibody.

For examination of lignin, samples were prepared (with basic fuchsin) as described in (Rafsanjani et al., 2015).

Samples were examined using a Leica DM6000B epifluorescence microscope, and images were acquired using a Qimaging Retiga CCD camera operated through

Openlab. The following channels were used: YFP (immunofluorescence imaging of

LM10 and LM11), GFP and TX2 (for detection of basic fuchsin (Dharmawardhana et al.,

1992; Kapp, Barnes, Richard, & Anderson, 2015)).

Primary and Secondary Antibodies

Previously published LM10 (Rat IgG2c) and LM11 (Rat IgM) antibodies were obtained from Plant Probes, UK (McCartney et al., 2005). These anti-rat monoclonal antibodies were generated against (1→4)-D-xylans. LM10 binds to unsubstituted or low substituted xylan backbone chains, while LM11 is able to additionally bind to wheat arabinoxylan. Alexa-fluor 488 secondary antibody was obtained from Invitrogen (goat anti-Rat IgG (H+L) polyclonal, CAT# A-11006).

145

FIGURES AND TABLES

146

Figure 3.1.

147 Figure 3.1. Conformational changes in S. lepidophylla, and Organ and Tissue

Stiffness. (A) Hydrated (H) and dehydrated (D) S. lepidophylla plant showing open conformation with spirally arranged stems and closed conformation showing outer stems curled over inner stems. (B) Time-lapse stills showing reversible curling of inner and outer stem types in response to a change in hydration state. (C) Inner stem (average length range = 3-6cm); insets show adaxial (Ad.) and abaxial (Ab.) stem sides including microphylls. (D) Outer stem (average length range = 6-12cm); insets show adaxial (Ad.) and abaxial (Ab.) stem sides including microphylls. I: inner; O(M): outer with microphylls; O(NM): outer with no microphylls. (E) Boxplot comparing average stiffness with standard error among inner and outer stem types (further information is included in

Table 3.1; n=25 per stem type). (F) Boxplot comparing average stiffness with standard error between adaxial and abaxial cortical tissue of inner stems (further information is included in Table 3.2).

148

Figure 3.2.

149 Figure 3.2. Morphology and Tissue Structure in S. lepidophylla. All images are from the apical region of inner stems with the exception of (D) which shows inner basal cells.

(A) Unstained cross-section showing tissue morphology. E= epidermis; C= cortex (Ad- adaxial; Ab-abaxial); VB= vascular bundle; T= trabeculae; scale bar = 200µm. (B)

Toluidine Blue O (TBO)-stained cross-section showing changes in tissue thickness and cell density between adaxial and abaxial cortex. Scale bars: 200µm and 50µm (insets).

(C-D) Transmission electron microscopy (TEM) images of (C) apical cortex and (D) basal cortex cell shape. Scale bar: 2µm. (E-F) Three-dimensional micro-computed X-ray tomography reconstructions showing a colour map of (E) cortex cell orientation and (F) lumen volume between adaxial and abaxial, as well as left and right stem sides. (G-H)

Quantification of normalized cell orientation and lumen volume of cortical tissue are shown for (G) left and right and (H) adaxial and abaxial stem sides. 0 represents the centre of the stem and the relative distance (in µm) that individual cells are from the stem centre. Cell orientation and lumen volume do not significantly differ between left and right stem sides whereas they significantly differ for adaxial and abaxial stem sides.

150

Figure 3.3.

151 Figure 3.3. Cell Wall Properties in S. lepidophylla. (A) Atomic force microscopy

(AFM) images of apical, inner stem adaxial and abaxial cells. Scale bar: 5µm. (B)

Young’s Modulus distribution showing average stiffness values for adaxial and abaxial cells represented in (A). (C) Immunofluorescence results for inner stem apical and basal regions showing LM11 binding pattern. Scale bar: 200µm. (D) Boxplot showing average fluorescence intensity (LM11) and standard error between adaxial and abaxial tissue regions for apical, middle and basal stem cross-sections.

152

Figure 3.4.

153 Figure 3.4. Summary of Gradients Responsible for Directional Stem Bending and

the Degree of Stem Curling in S. lepidophylla. Inner stems curl tightly into a spiral

shape upon drying, whereas outer stems curl into an arc shape upon drying. The ability to

bend in a specific direction relies primarily upon structural gradients (tissue density, cell angle, and cell/lumen size, hemicellulose distribution) between adaxial and abaxial

cortex. The difference in the degree of stem curling between inner and outer stem types is

most likely driven by tip-to-base compositional gradients (lignification and hemicellulose

distribution) in the cell wall. Outer stems do not show tip-to-base gradients whereas inner

stems do. CA: cell angle relative to the stem axis; CS/CW: cell size/cell wall thickness;

HC: hemicellulose; L: lignin.

154

Table 3.1. Mechanical Properties of S. lepidophylla Stems Mechanical Properties* in MegaPascal (Mpa) Whole Stem Young's Modulus (E) Yield Upper Tensile Strength Failure Modulus of Resilience Inner 288.08 ± 28.39 281.45 ± 27.23 258.77 ± 23.49 255.74 ± 23.61 144.33 ± 16.75 Outer with Microphylls 483.40 ± 55.79 399.83 ± 45.17 396.71 ± 41.55 377.07 ± 40.39 173.94 ± 20.83 Outer without Microphylls 511.83 ± 49.17 430.72 ± 48.10 403.62 ± 40.51 377.29 ± 40.93 188.07 ± 25.94 Tukey's HSD** 1 vs 2 0.01+ 0.11 0.02+ 0.05 0.60 1 vs 3 0.00+ 0.03+ 0.02+ 0.05 0.33 2 vs 3 0.90 0.86 0.99 1.00 0.89 * Mean ± standard error ** 1= Inner | 2= Outer with Microphylls | 3= Outer without Microphylls Differences among stem types were tested using one-way ANOVA (analysis of variance) with a cut-off of P= 0.05. 25 stems of each stem type were tested. Differences among stem types were further tested with a Tukey's HSD (honestly significant difference) analysis. Significant results (i.e., p< 0.05) are marked by +.

155 Table 3.2. Mechanical Properties of S. lepidophylla Stem Tissue Mechanical Properties* in MegaPascal (MPa) Tissue Region Young's Modulus (E) Yield Upper Tensile Strength Failure Modulus of Resilience Adaxial+ 269.74 ± 28.44 194.45 ± 32.58 194.28 ± 32.03 209.72 ± 24.61 90.32 ± 31.73 Abaxial+ 876.80 ± 113.46 1031.59 ± 190.45 1045.89 ± 190.86 1006.23 ± 185.14 1037.78 ± 448.30 Control** 809.21 ± 223.51 502.51 ± 154.48 455.13 ± 119.70 406.32 ± 123.35 179.02 ± 68.09 * Mean ± standard error Differences between adaxial and abaxial stem regions for each mechanical property were tested using two-sided Wilcoxon sign- rank tests with a cut-off of P= 0.05. 15 pairs (adaxial and abaxial portion of an individual stem) were tested. Significant differences (denoted by +) were found between adaxial and abaxial regions for all properties tested. ** Control tissues (left and right stem sides) were also similarly tested. No significant difference existed between left and right sides, and therefore, both sides were grouped as a single control.

156

Table 3.3. Adaxial and Abaxial Cortical Tissue Cell Dimensions Stem Portion and Tissue Region* Cell Dimensions Apical Middle Basal Adaxial Abaxial Adaxial Abaxial Adaxial Abaxial Total Cell Area 290.60 ± 16.19+ 203.00 ± 10.81+ 274.86 ± 12.02+ 230.32 ± 11.49+ 247.33 ± 9.36+ 215.02 ± 9.52+ Cell Wall Area 201.84 ± 10.15 182.63 ± 9.57 245.45 ± 10.43 217.91 ± 11.02 226.24 ± 8.50+ 202.52 ± 9.09+ Lumen Area 88.76 ± 6.89+ 20.38 ± 2.19+ 29.41 ± 2.50+ 12.41 ± 0.80+ 21.09 ± 1.47+ 12.50 ± 0.73+ Cell Wall Thickness 3.95 ± 0.1+ 5.19 ± 0.15+ 5.68 ± 0.13 6.08 ± 0.12 5.88 ± 0.12 5.81 ± 0.14 * Mean ± standard error Differences between adaxial and abaxial cell dimensions were tested with paired Student T-Tests with a cut-off of P= 0.05. 100 cells were measured for each stem and tissue region. Significant results (i.e., p< 0.05) are marked by +.

157 SUPPLEMENTARY INFORMATION

158

Supplementary Figure 3.1.

159 Supplementary Figure 3.1. Tissue Lignification as Detected with Basic Fuchsin.

Apical cross-sections show a higher degree of tissue lignification in abaxial cortex, near the periphery of the stem. Middle cross-sections are lignified throughout the abaxial cortex. Uniform lignification is observed in both adaxial and abaxial cortex in basal cross-sections. Scale bar: 200µm.

160

Supplementary Figure 3.2.

161 Supplementary Figure 3.2. LM10 Binding Pattern. Control (first row) cross-sections incubated with secondary antibody show little to no fluorescence signal. LM10 (second row) bound in a pattern similar to LM11 (Figure 3.3C), and showed similar fluorescence intensity (boxplot) patterns between adaxial and abaxial stem sides, and from stem tip-to- base. Scale bars: 200µm.

162

Supplementary Figure 3.3.

163 Supplementary Figure 3.3. Antibody Binding Pattern in Outer Stem Cross-Sections.

Outer stems showed similar binding patterns between LM10 and LM11 antibodies, as well as in tip-to-base cross-sections (middle section not shown). Scale bars: 200µm.

164

Supplementary Table 3.1. S. Lepidophylla Stem Dimensions Dimensions*/** Cortical Tissue Width Thickness Length Adaxial 1.93 ± 0.08 0.77 ± 0.06+ 4.65 ± 0.22 Abaxial 1.78 ± 0.11 0.41 ± 0.06+ 4.65 ± 0.21 * Width and thickness measured in mm, length in cm ** Mean ± standard error Differences between adaxial and abaxial stem regions were tested using two-sided Wilcoxon sign-rank tests with a cut-off of P= 0.05. 15 pairs (adaxial and abaxial portion of an individual stem) were tested. Significant results (i.e., p< 0.05) are marked by +.

165 Supplementary Table 3.2. S. lepidophylla Tissue and Cell Wall Composition Antibodya-c Polysaccharide Epitope Present Gradient Gradient Location & Tissue Type** Apical to Adaxial to Basal Abaxial LM5a Pectin Linear (1-4)-β-D-galactosyl residues - - - - LM6a Pectin Linear (1-5)-α-L-arabinan - - - - LM7a Pectin Partially methyl-esterified homogalacturonan - - - - LM10b Hemicellulose Unsubstituted/low substituted xylan + + + SCW Cortical LM11b Hemicellulose Unsubstituted/low/highly substituted xylan + + + SCW Cortical Unesterified/partially methyl-esterified LM18c Pectin - - - - homogalacturonan LM20c Pectin Methyl-esterified homogalacturonan - - - - Unesterified/partially methyl-esterified JIM5a Pectin + - - ML Metaxylem homogalacturonan JIM7a Pectin Methyl-esterified homogalacturonan + - - ML Metaxylem | PCW Phloem JIM13a Hemicellulose Arabinogalactan, AGPs* + - - PCW Trabeculae JIM14a Hemicellulose Arabinogalactan, AGPs - - - - M7a Pectin Rhamnogalacturonan I - - - - M14a Pectin Unbranched rhamnogalacturonan I - - - - Arabinogalactan on sidechains of M15a Hemicellulose + - - ML Metaxylem | SCW Cortical rhamnogalacturonan I, AGPs M36a Pectin Unbranched rhamnogalacturonan I - - - - ML Metaxylem | PCW Phloem | M38a Pectin Fully de-esterified homogalacturonan + - - SCW Cortical M58a Hemicellulose Xyloglucan + - - ML Cortical | PCW Phloem ML Cortical | PCW Phloem | SCW M89a Hemicellulose Xyloglucan (except XXXG) + - - Cortical M100a Hemicellulose Xyloglucan subunit XXXG + - - ML Cortical M118a Hemicellulose Monocot xylans - - - - * AGP= arabinogalactan ** ML= middle lamella; PCW= primary cell wall; SCW= secondary cell wall Twenty antibodies specific to various hemicellulose and pectin epitopes, from three different antibody series (LM, JIM, and M) were tested on cross-sections from apical, middle, and basal regions of inner stems. The presence (+) or absence (-) of a binding signal was recorded, as well as the presence/absence of gradient binding patterns (tip-to-base, and adaxial-to-abaxial). The cell wall location (PCW, SCW, or ML) and tissue type (cortex or vasculature) where the signal was observed was also recorded. a-c: (a) Complex Carbohydrate Research Centre (Pattathil et al. 2010), (b) Plant Probes (McCartney et al. 2005) (c) Plant Probes (Hall et al. 2013).

166 LINK BETWEEN CHAPTER 3 AND 4

Chapter 3 characterizes the functional gradients responsible for Selaginella

lepidophylla stem deformation. Inner and outer stem types both exhibit a morphological

gradient between adaxial and abaxial cortical tissue. Abaxial cortex is denser than adaxial cortex, cells are arranged parallel to the longitudinal stem axis (compared to adaxial cortex where cells range from ~30-55° relative to the axis), and cell walls are thicker.

Differences in tissue density, cell angle and cell wall thickness are observed through all stem segments from tip to base. In addition, inner stems exhibit compositional gradients from stem tip to base, whereas outer stems do not. Compositional gradients are observed at the tissue level, and include changes in cortex lignification and xylan distribution.

Given that they are observed in both inner and outer stems, it seems most likely that morphological gradients contribute to directed stem curling, while compositional gradients, that are observed in inner stems and to a limited degree in outer stems, most likely contribute to the degree of stem curling. These morphological and compositional gradients feed into the changes in stiffness observed between adaxial and abaxial cortex of inner stems, as well as to the changes in stiffness among inner and outer stem types.

In Chapter 4, morphological, compositional, and stiffness gradients are explored at the cell wall level in more detail. Because inner and outer stems are comparable in terms of morphology and inner stems show stronger compositional gradients, we focused our cell wall investigation on inner stems only. Atomic force microscopy (AFM) is used to analyze the topology of adaxial and abaxial cortex cell walls along the stem length and to investigate cell wall stiffness and viscoelastic properties. Finally, cell wall composition is explored using various histochemical stains.

167

CHAPTER 4

Characterizing Cell Wall Stiffness in the Resurrection Plant Selaginella lepidophylla as it Relates to Directional Stem Curling

168

ABSTRACT

As a physiological response to water loss during drought conditions, inner S.

lepidophylla stems curl into a spiral shape to prevent photoirradiation damage to their

photosynthetic surfaces. Curling is reversible and involves hierarchical deformation,

making S. lepidophylla an attractive model in which to study water-responsive actuation.

Investigation at the organ and tissue level has led to the understanding that the direction

and extent of stem curling can be partially attributed to stiffness gradients between adaxial and abaxial stem sides. Here, we examine cell wall stiffness to understand how it contributes to overall stem curling. We compare stiffness along the stem length and

between adaxial and abaxial stem sides using atomic force microscopy We show that

changes in cortex secondary cell wall development lead to cell wall stiffness gradients

from stem tip to base, and also between adaxial and abaxial stem sides. Changes in

cortical cell wall morphology and secondary cell wall composition are suggested to

contribute to the observed stiffness gradients.

169

INTRODUCTION

Nature is a perpetual source of inspiration for biomimetic and actuating devices

(Burgert & Fratzl, 2009a). Current biomimetic research involves a multi-scale approach to investigate how structure and mechanical properties at various length-scales determine organism function (Naleway et al., 2015). As more hierarchical models are studied, and their micro and nano properties better understood, more complex mimetic and actuating devices can be designed. Ideally, this would involve multi-functional devices with longer working lifespans arising from improved structural integrity at small length scales (C. Lv et al., 2018; Naleway et al., 2015). Of the many organisms studied, plants are interesting because of the wide range of functions and structures produced from a limited set of starting materials (i.e., cell wall components) (Erb et al., 2013; Shtein, Bar-On, & Popper,

2018).

Selaginella lepidophylla, a desiccation tolerant spikemoss, is a new hierarchical model to study water-responsive actuation (Brulé et al., 2016; Rafsanjani et al., 2015). S. lepidophylla plants are composed of hundreds of stems that curl upon themselves when dehydrated (Figure 4.1A, C). Curling is a physiological mechanism to prevent thermal and photoirradiation damage to photosynthetic surfaces during prolonged periods of drought (Lebkuecher & Eickmeier, 1991, 1993). Stems can be divided into two sides based on the direction of stem curling: the stem tip curls toward the adaxial (upper) side, and curls away from the abaxial (under) side during desiccation (Figure 4.1B, C). Stem

(un)curling is driven by differential swelling and shrinking of adaxial and abaxial cortex in response to water gain or loss. Deformation resulting from differential swelling and

170 shrinking is a well-established form of actuation in various plant species (reviewed in

(Rivka Elbaum & Abraham, 2014)), and is driven by changes in stiffness resulting from bilayer-like variations in morphology and composition at tissue and/or cell wall levels.

Stiffness, defined as the ability of a material to resist deformation in response to load stress (described in Box 1 of Appendix 1), is a critical mechanical property to consider with respect to actuation because it determines the direction and extent of conformational change undertaken by a material. The stiffer a material is, the less it will deform under stress. By layering or otherwise joining materials with different stiffness, it is possible to control composite material deformation in a specific direction (Niklas, 1992; Vincent,

1992).

Stiffness profiles leading to deformation in S. lepidophylla have been studied at the organ ((Rafsanjani et al., 2015), see Chapter 2) and tissue (Chapter 3) level. Here, we characterize cell wall stiffness of S. lepidophylla to understand its contribution to the observed direction and degree of stem curling. To accomplish this, we take advantage of atomic force microscopy, a technique frequently used to study plant cell wall mechanics

(reviewed in (Cosgrove, 2016; Milani, Braybrook, & Boudaoud, 2013; Vogler, Felekis,

Nelson, & Grossniklaus, 2015)). We compare adaxial and abaxial cortex cell wall stiffness along the length of the stem to gain an idea of local stiffness profiles across stem regions and between stem sides.

171

RESULTS AND DISCUSSION

Cell Wall Stiffness Differs between Adaxial and Abaxial Stem Sides

To characterize S. lepidophylla cortex cell wall stiffness, we performed contact mode AFM on transverse sections from five different regions spanning from stem tip to base. Both adaxial and abaxial cortical cells were indented in each region. Since samples were air-dried prior to indentation, turgor was not an effect on cell wall stiffness that we needed to consider for testing or analysis. A representative topological scan of a cortical cell wall (Figure 4.2A and inset) shows the areas chosen for indentation. Vertical deflection and elastic modulus distribution (Figure 4.2B) graphs were generated for each indented spot in the cell wall. These distribution graphs were consolidated and an average elastic modulus (E) for adaxial and for abaxial cortical cell walls for each stem region was calculated. Results for the five tested stem regions are presented in Figure 4.2C. An overall trend of increasing stiffness from stem tip to base is observed for adaxial cortical cell walls. In contrast, abaxial cell wall stiffness remains more or less constant across stem regions, with an increase in stiffness seen only in the topmost stem segment. In addition, a second gradient is visible between stem sides. At the stem base, adaxial cortical cell walls are significantly stiffer (1.44E-03, p< 0.05; Supplementary Table 4.1) than abaxial cell walls, whereas the opposite is observed at the stem tip (5.02E-05, p<0.05). The middle region shows no significant difference between adaxial and abaxial cortical cell wall stiffness (Supplementary Table 4.1). This bilayer-like difference in stiffness between adaxial and abaxial cortical cell walls is similar to the adaxial/abaxial stiffness bilayer observed at the tissue level (Chapter 3). At the cell wall level, though there is the added complexity of a cell wall stiffness gradient from stem tip to base.

172

Interestingly, the gradient of increasing stiffness in adaxial cortical cell walls follows a secondary cell wall developmental (SCW) gradient in which walls become increasingly lignified and thicker moving from stem tip to base, as reported in monocots such as bamboo, eudicots such as Arabidopsis thaliana, and tree species (Barra-Jiménez

& Ragni, 2017; Tan et al., 2011; Ruiqin Zhong, Taylor, & Ye, 1997). This type of stiffness gradient allows the upper portions of stems/trunks/culms to be flexible and bend more readily in response to environmental factors such as wind, while the stiffer base resists movement and prevents the plant from being uprooted (Dixon & Gibson, 2014;

Gardiner, Berry, & Moulia, 2016; Ghavami, Rodrigues, & Paciornik, 2003). The difference between adaxial and abaxial SCW stiffness can probably also be explained by a developmental gradient. In this instance, however, the gradient involves distinct SCW differentiation programs, which, as reported for Arabidopsis thaliana, is essential for correct adaxial/abaxial leaf patterning and identity (reviewed in (Fukuda, 2004)). In S. lepidophylla, these SCW developmental gradients appear to contribute to longitudinal and adaxial-abaxial stiffness gradients that work together to allow the stem tip to curl tightly on itself toward the adaxial side (least stiff region of the stem), while restricting movement at the stem base and to direct movement away from the abaxial side.

Cell Wall Layering Differs Between Adaxial and Abaxial Stem Sides

Studies suggest a role for cell wall morphology in driving morphogenesis and fluid-driven tissue/organ deformation in plants (Amir J Bidhendi & Geitmann, 2018;

Geitmann & Ortega, 2009; Vogler et al., 2015). We therefore investigated S. lepidophylla cortex cell wall morphology. Air-dried, transverse sections from different segments along

173

the length of the stem were scanned with AFM to generate topographical images of

adaxial and abaxial cortical cell walls (Figure 4.4A-B; Supplementary Figure 4.1).

Overall, cortex cells appear round to oval in geometry, with no obvious change in cell

shape between adaxial and abaxial stem sides or between tip and basal stem segments.

Differences arise when comparing cell wall layering between adaxial and abaxial cortical

cells. Abaxial cell walls show very distinct secondary cell wall layers, while those in the

adaxial region appear relatively smooth. This pattern is observable in sections along the

length of the stem and is also visible in cortical cell walls imaged with transmission

electron microscopy (Supplementary Figure 4.2). Secondary cell walls usually contain

three layers (S1-S3, with S2 being the thickest layer), as observed in Arabidopsis

interfascicular fibers and in wood species (R. Zhong & Ye, 2015). In some species such

as bamboo, more layers are visible. Bamboo fibers can have up to six to eight distinct cell

wall layers (D. Liu, Song, Anderson, Chang, & Hua, 2012; Parameswaran & Liese, 1976;

R. Zhong & Ye, 2015; Zou, Jin, Lu, & Li, 2009). Qualitatively, S. lepidophylla cortical

cells walls, especially those on the abaxial stem side, appear to have more than three

SCW layers, and therefore resemble bamboo with respect to cell wall layer morphology.

Compared to adjacent cell types, fiber cells in Arabidopsis, wood, and bamboo are stiffer

(Habibi et al., 2015; Persson, 2000; Tan et al., 2011; Ruiqin Zhong et al., 1997). Thus,

given their similar SCW morphology, S. lepidophylla stem cortical cells most likely act

like fiber cells, providing structural and mechanical support to the stem. Further work is required to determine whether or not the difference in adaxial and abaxial SCW layer patterning affect stiffness in each stem side.

174

Cell Wall Composition Resembles Gelatinous Fibers of Tension Wood, Coiling

Vines, and Ice Plant Seed Capsules

It is well established that cell wall composition affects cell wall mechanical properties, including stiffness (A. J. Bidhendi & Geitmann, 2016; Cosgrove & Jarvis,

2012; Gibson, 2012). Therefore, we examined cortical cell wall composition along the length of inner S. lepidophylla stems on both adaxial and abaxial sides. We were particularly interested in identifying whether or not there was any relation between cell wall composition and the cell wall layering patterns observed with AFM (Figure 4.2) and

TEM (Supplementary Figure 4.2).

Cell walls were stained with safranin O to detect lignin and counter-stained with alcian blue (Figure 4.5A-B) (Bond, Donaldson, Hill, & Hitchcock, 2008; Marjamaa et al.,

2003; Ruzin, 1999). In all sections, the outermost SCW layers of S. lepidophylla cortex stained with safranin O, while the innermost SCW layer stained with alcian blue (Figure

4.5B). Increasing wall thickness from stem tip to base corresponded with increasing SCW layers stained with safranin O while alcian blue continued to exclusively stain the innermost SCW layer. Further investigation with other cell wall stains and antibodies revealed that the SCW inner layer stained by alcian blue appears to be predominantly enriched in cellulose (Supplementary Figure 4.3) and hemicellulose (Supplementary

Figure 4.4), but not pectin (Supplementary Figure 4.5; see also Supplementary Table 3.2 in Chapter 3).

While this compositional pattern does not appear to correspond to the SCW morphological layering observed with AFM and TEM, it does resemble the established bilayer-like cell wall composition of gelatinous (G) fiber and G-fiber-like cells in species

175 such as tension wood, coiling vines, and ice plant seed capsules (Bowling & Vaughn,

2008, 2009; Harrington et al., 2011; Meloche et al., 2007). G-fiber and G-fiber-like cells differ from normal fiber cells by the presence of a tertiary hydrophilic, gelatinous layer in their cell wall that is surrounded by a lignified SCW layer. In both tension wood and coiling vines, the gelatinous layer is pectin-rich (Bowling & Vaughn, 2008, 2009), while in ice plant seed capsules, this layer is predominantly composed of cellulose (Harrington et al., 2011). In all these species, G-fibers lead to directional organ movement, and in the case of ice plants, reversible deformation. Given the compositional bilayer similarity, and the (hemi)cellulose inner layer that resembles the cellulose-rich inner layer of ice plant seed capsules cells, S. lepidophylla cortical cell composition probably contributes to reversible stem deformation through swelling and shrinking of its (hemi)cellulose inner

SCW layer. The degree of swelling/shrinking is most likely determined by the proportion of lignin:(hemi)cellulose layers present in the wall, resulting from developmental gradients of SCW lignin deposition and cell wall thickness (i.e., increased SCW lignification and cell wall thickness as cells mature) (Ruiqin Zhong et al., 1997). In adaxial apical cortex where walls are thinnest and the proportion of lignin layers is qualitatively similar to that of the (hemi)cellulose, swelling probably is not as restricted as it might be in adaxial basal cortex where cell walls are thicker (see Table 3.3 in

Chapter 3) and the proportion of lignified layers is higher. Similarly, at least at the stem tip, abaxial cortical cell walls most likely do not swell to the same extent as adaxial ones.

This gives rise to differential swelling leading to stem curling toward the adaxial side, and a gradient of curling along the stem length.

176

CONCLUSION

S. lepidophylla is a desiccation tolerant plant used as a model to study hierarchical, water-responsive deformation. Previous studies at the organ (Chapter 2) and tissue level

(Chapter 3) have highlighted the mechanical, morphological and compositional properties leading to reversible stem curling. Here, we examine the cell wall level to understand its contribution to overall stem curling. Using atomic force microscopy, we identify cortical cell wall stiffness gradients from stem tip to base, and also between adaxial and abaxial stem sides (Figure 4.2, Supplementary Table 4.1). These gradients correspond to established secondary cell wall developmental gradients (i.e., increasing stem tip to base lignification [Figure 4.4, Chapters 2-3] and cell wall thickness [Chapter

3], and adaxial/abaxial differentiation programs) (Barra-Jiménez & Ragni, 2017; Tan et al., 2011; Ruiqin Zhong et al., 1997). Changes in SCW layer morphology (Figure 4.3) and composition (Figure 4.4.) between adaxial and abaxial stem sides, as well as composition along the stem length, are suggested to contribute to the observed stiffness gradients. As well, G-fiber-like bilayer composition of cortical cell walls, combined with the presence of developmental gradients, is suggested to result in differential swelling between adaxial and abaxial stem sides and along the stem length, resulting in directional stem curling.

S. lepidophylla has provided the opportunity to study properties at small length scales that lead to macro-scale actuation. Incorporating small-scale properties into synthetic actuating systems is critical for improving actuator structural integrity and functional lifespan (C. Lv et al., 2018; Naleway et al., 2015). The cell wall properties

177 identified here could be explored further with computational models to extract features that could lead to designing synthetic actuators with improved function and structure.

178

MATERIALS AND METHODS

Plant Materials. Mature Selaginella lepidophylla plants were purchased from Canadian

Air Plants (New Brunswick, Canada), and maintained in a desiccated state (at 25°C and

50% air humidity) until use. Prior to experimentation, S. lepidophylla plants were placed in a plate of water and allowed to rehydrate for three consecutive days (to achieve 100% relative water content).

Time-lapse Video Capture. Video capture was conducted following the protocol for inner stems as described in (Rafsanjani et al., 2015).

Staining and Microscopy. Inner stems were isolated and cut into five regions (tip, tip- middle, middle, middle-base, and base). Lignin Detection and Cell Wall Counter-

Staining. Samples were fixed with FAA (formaldehyde:acetic acid:alcohol) solution for one week, dehydrated through an ethanol series, and embedded in paraffin. Following paraffin removal, thick sections (1µm) were stained with Safranin O and Alcian Blue as outlined in (Ruzin, 1999). Pectin and Hemicellulose Detection. Samples were prepared following the method outlined in Chapter 3. Cellulose Detection. Samples were prepared following the method outlined in as outlined in (Ruzin, 1999) for Calcofluor

White staining. All samples were imaged using a Leica DM6000B epifluorescence microscope and Qimaging Retiga CCD camera with Openlab imaging software.

Scanning Electron Microscopy. Fresh, hand-cut sections (~1mm long) from apical the stem region were mounted on SEM stubs and allowed to air-dry for 24 hours prior to imaging. Dry sections were imaged using a Hitachi TM3030Plus SEM in backscatter electron mode.

179

Atomic Force Microscopy. A JPK Atomic Force Microscope (JPK Nano-wizard@3 Bio

Science, Berlin, Germany) was used for imaging and force spectroscopy. S. lepidophylla stem samples (three stems, five regions per stem, 3 sections per region) were cut transversally by hand and sections placed on double-sided clear tape on a microscope slide. Cortical stem tissue in adaxial and abaxial regions was located to perform force measurements. All the measurements were performed on tissue in a dry state. Using the

QI imaging mode of the JPK AFM, a force map was created within the area of 30 µm2 on the sample. For consistency considerations, only the points (~5-9 per cell) located on the selected parts were indented. For contact mode imaging, super-sharp standard Force

Modulation Mode Reflex Coating (FMR) cantilevers with diamond-like carbon nano-tip of radius 2-3 nm were used. For indentation measurements, Non-Contact High

Resonance (NCHR) cantilevers (Nanotools USA LLC, Henderson, NV) with a nominal spring constant of 40 N/m and integrated spherical tip of radius 50 nm (+/-10%) and 100 nm (+/-10%) were applied. The indentation frequency was 250 Hz. The deflection sensitivity of the piezo module was obtained by probing the surface of the glass substrate.

A thermal tuning method was then used to calibrate the stiffness of the cantilever. The indentation was repeated at the same location for consistency as well to ensure that the sample was not permanently deformed. The indentation depth depends on the applied load, as well as the stiffness of the tip and that of the sample. The elastic modulus of the sample was estimated from the approaching force-indentation depth curve, according to the Hertzian contact model. AFM data analysis was performed with the native JPK data processing software. Statistical significance was determined by a paired student's t-test, when applicable. Differences were considered significant at p<0.05.

180

FIGURES

181

Figure 4.1.

182 Figure 4.1. Water-Responsive Deformation in S. lepidophylla. (A) Mature S. lepidophylla plant showing dehydrated and hydrated conformations, as well as an isolated inner stem (inset). In a dehydrated state, S. lepidophylla appears as a ball with outer stems curled over inner ones. In a hydrated state, stems are uncurled and lie flat in a spiral rosette shape. (B) Transverse section from the apical stem region imaged with scanning electron microscopy showing adaxial (upper) and abaxial (under) stem sides. Selaginella stems appear as thick cylinders of cortex surrounding a centralized vascular bundle

(Brighigna et al., 2002; Weng et al., 2010). Scale bar: 200µm. (C) Time-lapse image captures showing pattern of reversible inner stem curling. As stems dry, the tip curls on itself toward the stem base. Upon rehydration with water, this process is reversed.

183

Figure 4.2.

* *

184 Figure 4.2. Stiffness of Cortical Cell Walls in the S. lepidophylla Stem. (A-C)

Representative images of AFM results from scanning and indenting an apical adaxial

cortical cell from an inner stem. (A) Topological scan showing regions of indentation

(inset). (B) Vertical deflection curve from AFM contact mode. (C) Histogram showing

the average distribution of Young’s Modulus (E) across cell wall indentations. (C) Graph showing differences in stiffness between adaxial and abaxial cell walls for five regions along the length of the stem. Significant differences are marked by * (p<0.05;

Supplementary Table 5.1). Adaxial cell walls show increased stiffness from stem tip to base, while abaxial cells across stem regions show relatively consistent stiffness with an increase visible only at the stem tip.

185

Figure 4.3.

186 Figure 4.3. Cell wall layering in S. lepidophylla Cortex. Topological scans of apical

(A) and (B) basal regions showing adaxial and abaxial cell shape and cell wall layering.

Abaxial cell walls in both apical and basal stem regions show more prominent layering than in adaxial cell walls. Scale bars: 5µm for large scan area, and 1µm for smaller scan area.

187

Figure 4.4.

188 Figure 4.4. S. lepidophylla Cortical Cell Wall Composition. (A) Transverse sections from four inner stem regions (apical, apical-middle, middle-basal, and basal) stained with alcian blue (blue) and with safranin O (red) to detect lignin. Stem cortex became increasingly stained for safranin O moving from stem tip to base. At the tip, abaxial cortex is visibly more stained with safranin O than adaxial cortex. At the base, both stem sides appear to stain strongly with safranin O. Apical-middle and middle-basal regions show intermediate staining between what is observed the stem tip and base. Scale bar:

200µm. (B) High magnification images of adaxial and abaxial cell walls in apical and basal stem regions. In the apical region in adaxial cells, safranin O stains the outer secondary wall layer, while the counterstain, alcian blue, stains the inner secondary wall layer. A similar pattern is seen in abaxial cortical cells near the center of the stem, but cells near the stem periphery show more layers with safranin O staining and a narrow inner layer stained with alcian blue. In contrast, adaxial and abaxial cell wall layers in the basal stem region almost exclusively stain with safranin O, with only a very small inner layer stained with alcian blue. Scale bar: 10µm.

189

SUPPLEMENTARY INFORMATION

190

Supplementary Figure 4.1.

191 Supplementary Figure 4.1. Topological Scans of Adaxial and Abaxial Cortical

Cells/Cell Walls in the Middle Region of Inner S. lepidophylla Stems. Similar to the rest of the stem, cells in the middle region are round to ovoid in shape. Cell walls resemble those of the basal region, and abaxial walls show more distinct layering than adaxial cell walls. Scale bars: 7.5µm for low magnification, and 2.5 µm for high magnification.

192

Supplementary Figure 4.2.

193 Supplementary Figure 4.2. Cell Wall Layering Visualized by Transmission Electron

Microscopy. Adaxial (left) and abaxial (right) cortical cells from tip, middle and basal stem regions appear round to oval in shape. Higher magnification images reveal cell wall layering. Layering is more prominent in abaxial cortical cell walls along the length of the stem as compared to adaxial cortex. Scale bars: 2µm for low magnification, 500nm for high magnification.

194

Supplementary Figure 4.3.

195 Supplementary Figure 4.3. Cellulose Cell Wall Distribution. (A-B) Arabidopsis thaliana (Landsberg) apical stem cross-section as a control for Calcofluor White (CFW) staining. Cellulose is indicated by a blue-white colour. CFW binds to most tissue types in

Arabidopsis stems, and binds most strongly to the phloem cap and xylem. (C-E) S. lepidophylla apical stem cross-section. (C) CFW binds to both (D) adaxial and (E) abaxial cortex, as well as to the phloem. CFW binds strongly to the innermost secondary cortical cell wall layer (observed in merged images).

196

Supplementary Figure 4.4.

197 Supplementary Figure 4.4. Hemicellulose Cell Wall Distribution. (A-E) LM10 binding pattern. (A-B) Arabidopsis (Columbia-0) apical stem cross-section as a control for LM10 binding pattern. Binding pattern is consistent with that in (Pattathil et al.,

2010). (C-E) S. lepidophylla apical stem cross-section. (C) LM10 binds to cortical tissue and xylem. (D) Adaxial and (E) abaxial cortex. LM10 binds most strongly to the innermost secondary cell wall layer in adaxial cortex, and binds throughout the rest of the secondary cell wall in both adaxial and abaxial cortex to a lesser degree. (F-J) LM11 binding pattern. (F-G) Arabidopsis (Columbia-0) apical stem cross-section as a control for LM11 binding pattern. Binding pattern is consistent with that in (Pattathil et al.,

2010). (C-E) S. lepidophylla apical stem cross-section. (C) LM11 binds to cortical tissue and xylem. (D) Adaxial and (E) abaxial cortex. LM11 binds throughout the secondary cell wall in both adaxial and abaxial cortex.

198

Supplementary Figure 4.5.

199 Supplementary Figure 4.5. Pectin Cell Wall Distribution. (A-C) JIM7 binding pattern.

(A-B) Arabidopsis (Columbia-0) apical stem cross-section as a control for JIM7 binding.

The binding pattern is consistent with that described in (Hall, Cheung, & Ellis, 2013). (C)

S. lepidophylla apical stem cross-section. JIM7 binds to phloem primary cell walls and the middle lamella of the xylem. (D-F) JIM13 Binding pattern. (D-E) Arabidopsis

(Columbia-0) apical stem cross-section as a control for JIM13 binding. The binding pattern is consistent with that described in (Hall et al., 2013). (F) S. lepidophylla apical stem cross-section. JIM13 binds in some places to the epidermis. Neither JIM7 nor

JIM13 bind to cortex cell walls or the middle lamella between cortical cells.

200

Supplementary Table 4.1. S. lepidophylla Cortical Cell Wall Stiffness Cell Wall Stiffness* Cortical Tissue Apical Middle Basal Adaxial 339.28 ± 13.29+ 699.38 ± 21.81 927.08 ± 21.12+ Abaxial 867.52 ± 31.50+ 718.96 ± 27.70 738.34 ± 23.84+ * Mean ± standard error Differences between adaxial and abaxial stem regions were tested using two-sided Wilcoxon sign-rank tests with a cut-off of P= 0.05. 5 pairs of cells (adaxial and abaxial portion of an individual stem) were tested for each stem region (apical, middle, basal). Significant results are marked by +

201

CONCLUSIONS AND FUTURE WORK

The plant kingdom is an excellent source of inspiration for actuating devices. As techniques evolve to study biological systems, plant actuators can be studied to greater depths to better understand and more precisely mimic their actuating functions.

Generating synthetic actuators that accurately replicate the function of biological models requires knowledge of the micro and nano-scale features contributing to macro-scale actuation. Thus, recent studies of biological actuators involve a hierarchical approach to investigate properties at different length scales leading to actuation. As well, new models displaying hierarchical actuation, reversible deformation, and/or long functional lifespans are sought after to improve the functional complexity synthetic actuators.

This thesis explores the hypothesis that the resurrection plant S. lepidophylla is a suitable model for studying reversible, hierarchical actuation. Based on the results presented in Chapters 2-4, it is indeed an excellent model for studying hierarchical actuation leading to the development of more complex synthetic actuators:

Chapter 2 presents a kinematic study of S. lepidophylla stems that explores the relationship between water gain/loss, stem stiffness, and organ deformation. Two stem types (inner and outer) are characterized based on their differing stiffness properties and their pattern and rate of deformation in response to water gain and loss. Preliminary investigation at the tissue level suggests that the curling profiles of inner and outer stem types arise from asymmetric tissue lignification and tissue density between adaxial and abaxial stem sides. Finally, based on the macro-level findings, curling profiles of inner and outer stem types can be computationally replicated using finite element modelling.

202

Chapter 3 presents an in-depth examination of the tissue-level morphological, compositional and mechanical properties leading to the specific curling profiles of inner and outer S. lepidophylla stem types. Morphological bilayers in tissue density and cell angle between adaxial and abaxial stem sides are suggested to affect the direction of curling of both inner and outer stems. Meanwhile, compositional gradients in tissue lignification and hemicellulose distribution along the length of the stem (tip to base) most likely contribute to the extent of stem curling observed in inner and outer stem types.

Chapter 4 builds on the tissue-level results presented in Chapter 3, and explores the morphological, compositional and stiffness properties of inner S. lepidophylla stem cell walls leading to organ-level deformation. Changes in stiffness between adaxial and abaxial cell walls are visible along the length of the stem. These changes arise from morphological (cell wall layering) and compositional (lignin/(hemi)cellulose layering) cell wall gradients that contribute to differential swelling between adaxial and abaxial stem tissue, and that govern the different degrees of stem curling visible from inner stem tip to base. Continued work using AFM to indent S. lepidophylla stem cell walls at different frequencies will reveal more information about cell wall mechanical behaviour in this plant, including viscoelasticity.

This thesis provides an experimental foundation for further work investigating S. lepidophylla actuation through computational modelling. The identified properties leading to hierarchical actuation could be explored through finite element (FE) modelling simulations to better understand the relationship between properties at different length scales, as well as the contribution of each property to stem deformation. This could be accomplished by building on the FE models presented in Chapter 2, by modifying

203 morphological, compositional and mechanical gradients and observing how these changes affect the direction and degree of stem deformation. This would also provide an opportunity to observe how the properties of S. lepidophylla could be scaled up for use in larger actuators, since S. lepidophylla stems are relatively small (~3-10cm in length).

Together with the outcomes of a computational modelling approach, it would be interesting to develop proof-of-concept prototypes based on the findings of Chapters 2-

4. In fact, some actuating prototypes have already been developed or inspired by the work presented in Chapter 2 ((Velders et al., 2017; Zhang, Desta, & Naumov, 2016; Zhang,

Qiu, Yuan, & Zhang, 2017). These prototypes are based on macro-scale observations of

S. lepidophylla movement. It would be interesting to see prototypes developed that incorporate the functional gradients observed at the tissue and cell wall levels in S. lepidophylla. For example, these functional gradients could be useful in generating hydrogels with more complex actuation. This could be accomplished through 4D printing, a recent technique that involves 3D printing materials that can change shape in response to a stimulus ((Bakarich, Gorkin, & Spinks, 2015; Ding et al., 2017; Ge et al.,

2016; Gladman, Matsumoto, Nuzzo, Mahadevan, & Lewis, 2016). Incorporating morphological or compositional bilayers or gradients into hydrogels could result in gels that are capable of more dynamic or complex ranges of movement/deformation, or that demonstrate multiple functions in response to a single stimulus.

Continued work on M. flabellifolius (Appendix 4.2) would involve a hierarchical experimental investigation similar to what was conducted with S. lepidophylla. Building on the preliminary kinematic work presented in Appendix 4.2, tensile testing of leaves

(along both x and y axes) would help to understand the organ level stiffness properties

204 involved in leaf deformation. Characterization of tissue and cell wall level mechanical, morphological and compositional properties would help to understand the mechanisms underlying hierarchical deformation in M. flabellifolius leaves. Unlike S. lepidophylla, I would expect to see more extensive cell wall folding in response to water loss. Previous work on M. flabellifolius has shown that parenchymatous cells lose up to 44% of their volume during drying, and that their cell walls dramatically fold in response to dehydration (Jill M Farrant, 2000; Moore et al., 2006). Interestingly, studies have mostly focused on parenchymatous tissue rather than sclerenchymatous ribs and their role in the pattern of M. flabellifolius leaf deformation. Given the structure of sclerenchyma as compared to parenchyma, I expect the ribs to play a rigidifying role, and most likely offer mechanical support to the leaf during deformation. However, it would be interesting to characterize their specific role in the leaf through investigation of their mechanical behaviour, as well as their morphology and composition. Like S. lepidophylla, M. flabellifolius would contribute to our understanding of hierarchical deformation.

However, S. lepidophylla is a relatively simple system with a simple mode of deformation (bending in x and y axes). In contrast, M. flabellifolius shows much more complex deformation (folding in x, y and z axes, and a shift from a flat to a concave conformation), and coordinated folding sequences (i.e., fan-like folding of parenchymatous tissue, and simultaneous folding of the whole leave against the stem). A better understanding of the mechanisms leading to this coordinated folding could lead to prototypes of actuating devices with stepwise changes in conformation, or that are capable of simultaneous deformation in different directions.

205

LITERATURE CITED

Abdalla, K. O., Baker, B., & Rafudeen, M. S. (2010). Proteomic analysis of nuclear proteins during dehydration of the resurrection plant Xerophyta viscosa. Plant Growth Regulation, 62(3), 279-292. Abdalla, K. O., & Rafudeen, M. S. (2012). Analysis of the nuclear proteome of the resurrection plant Xerophyta viscosa in response to dehydration stress using iTRAQ with 2DLC and tandem mass spectrometry. Journal of proteomics, 75(8), 2361-2374. Abraham, Y., Tamburu, C., Klein, E., Dunlop, J. W., Fratzl, P., Raviv, U., & Elbaum, R. (2012). Tilted cellulose arrangement as a novel mechanism for hygroscopic coiling in the stork's bill awn. J R Soc Interface, 9(69), 640-647. doi:10.1098/rsif.2011.0395 Acosta-Gamboa, L. M., Liu, S., Langley, E., Campbell, Z., Castro-Guerrero, N., Mendoza- Cozatl, D., & Lorence, A. (2017). Moderate to severe water limitation differentially affects the phenome and ionome of Arabidopsis. Functional Plant Biology, 44(1), 94-106. Agarwal, U. P. (2011). Lignin quantitation by FT-Raman spectroscopy. Paper presented at the Proceedings 16th International Symposium on Wood, Fiber and Pulping Chemistry. Agarwal, U. P. (2014). 1064 nm FT-Raman spectroscopy for investigations of plant cell walls and other biomass materials. Frontiers in plant science, 5. Agarwal, U. P., & Ralph, S. A. (1997). FT-Raman spectroscopy of wood: identifying contributions of lignin and carbohydrate polymers in the spectrum of black spruce (Picea mariana). Applied spectroscopy, 51(11), 1648-1655. Aharoni, H., Abraham, Y., Elbaum, R., Sharon, E., & Kupferman, R. (2012). Emergence of spontaneous twist and curvature in non-euclidean rods: application to Erodium plant cells. Phys Rev Lett, 108(23), 238106. Allen, R. D. (1969). Mechanism of the seismonastic reaction in Mimosa pudica. Plant Physiology, 44(8), 1101-1107. Alpert, P., Bone, E., & Holzapfel, C. (2000). Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in plant ecology, evolution and systematics, 3(1), 52-66. Antony, R., & Thomas, R. (2011). A mini review on medicinal properties of the resurrecting plant Selaginella bryopteris (Sanjeevani). International Journal of Pharmacy & Life Sciences, 2(7). Apostolova, E., Rashkova, M., Anachkov, N., Denev, I., Toneva, V., Minkov, I., & Yahubyan, G. (2012). Molecular cloning and characterization of cDNAs of the superoxide dismutase gene family in the resurrection plant Haberlea rhodopensis. Plant Physiology and Biochemistry, 55, 85-92. Armon, S., Efrati, E., Kupferman, R., & Sharon, E. (2011). Geometry and mechanics in the opening of chiral seed pods. Science, 333(6050), 1726-1730. Arnarson, P. Ö. (2011). Biomimicry. Reykjaví¿ k University. Askey, R., & Daalhuis, A. O. (2010). Generalized Hypergeometric Functions and Meijer G-Function, Chapter 16 in FWJ Olver, DW Lozier, RF Boisvert CW Clark

206

(Eds.) NIST Handbook of Mathematical Functions: Cambridge University Press. Aziz, M. S. (2016). Biomimicry as an approach for bio-inspired structure with the aid of computation. Alexandria Engineering Journal, 55(1), 707-714. Bai, R., Yang, J., Morelle, X. P., Yang, C., & Suo, Z. (2018). Fatigue Fracture of Self- Recovery Hydrogels. ACS Macro Letters, 7, 312-317. Bakarich, S. E., Gorkin, R., & Spinks, G. M. (2015). 4D printing with mechanically robust, thermally actuating hydrogels. Macromolecular rapid communications, 36(12), 1211-1217. Banks, J. A. (2009). Selaginella and 400 million years of separation. Annu Rev Plant Biol, 60, 223-238. doi:10.1146/annurev.arplant.59.032607.092851 Bar-Cohen, Y. (2006). Biomimetics—using nature to inspire human innovation. Bioinspiration & biomimetics, 1(1), P1. Barbacci, A., Lahaye, M., & Magnenet, V. (2013). Another brick in the cell wall: biosynthesis dependent growth model. PLoS One, 8(9), e74400. Barra-Jiménez, A., & Ragni, L. (2017). Secondary development in the stem: when Arabidopsis and trees are closer than it seems. Current Opinion in Plant Biology, 35, 145-151. Benyus, J. M. (1997). Biomimicry: Innovation inspired by nature: Morrow New York. Bergtrom, G., Schaller, M., & Eickmeier, W. G. (1982). Ultrastructural and biochemical bases of resurrection in the drought-tolerant vascular plant, Selaginella lepidophylla. J Ultrastruct Res, 78(3), 269-282. Bertinetti, L., Fischer, F., & Fratzl, P. (2013). Physicochemical basis for water- actuated movement and stress generation in nonliving plant tissues. Physical review letters, 111(23), 238001. Bidhendi, A. J., & Geitmann, A. (2016). Relating the mechanics of the primary plant cell wall to morphogenesis. J Exp Bot, 67(2), 449-461. doi:10.1093/jxb/erv535 Bidhendi, A. J., & Geitmann, A. (2018). Finite element modeling of shape changes in plant cells. Plant Physiology, 176(1), 41-56. Bond, J., Donaldson, L., Hill, S., & Hitchcock, K. (2008). Safranine fluorescent staining of wood cell walls. Biotechnic & Histochemistry, 83(3-4), 161-171. Bowling, A. J., & Vaughn, K. C. (2008). Immunocytochemical characterization of tension wood: gelatinous fibers contain more than just cellulose. American journal of botany, 95(6), 655-663. Bowling, A. J., & Vaughn, K. C. (2009). Gelatinous fibers are widespread in coiling tendrils and twining vines. American journal of botany, 96(4), 719-727. Braam, J. (2005). In touch: plant responses to mechanical stimuli. New Phytologist, 165(2), 373-389. Brighigna, L., Bennici, A., Tani, C., & Tani, G. (2002). Structural and ultrastructural characterization of Selaginella lepidophylla, a desiccation-tolerant plant, during the rehydration process. Flora-Morphology, Distribution, Functional Ecology of Plants, 197(2), 81-91. Brulé, V., Rafsanjani, A., Pasini, D., & Western, T. L. (2016). Hierarchies of plant stiffness. Plant Science.

207

Burgert, I., & Fratzl, P. (2009a). Actuation systems in plants as prototypes for bioinspired devices. Philos Trans A Math Phys Eng Sci, 367(1893), 1541-1557. doi:10.1098/rsta.2009.0003 Burgert, I., & Fratzl, P. (2009b). Plants control the properties and actuation of their organs through the orientation of cellulose fibrils in their cell walls. Integr Comp Biol, 49(1), 69-79. doi:10.1093/icb/icp026 Busse-Wicher, M., Grantham, N. J., Lyczakowski, J. J., Nikolovski, N., & Dupree, P. (2016). Xylan decoration patterns and the plant secondary cell wall molecular architecture. Biochemical Society Transactions, 44(1), 74-78. Butler, H. J., Ashton, L., Bird, B., Cinque, G., Curtis, K., Dorney, J., . . . Martin-Hirsch, P. L. (2016). Using Raman spectroscopy to characterize biological materials. nature protocols, 11(4), 664. Cai, Z.-L., Zeng, H., Chen, M., & Larkum, A. W. (2002). Raman spectroscopy of chlorophyll d from Acaryochloris marina. Biochimica et Biophysica Acta (BBA)-Bioenergetics, 1556(2-3), 89-91. Canny, M. (2000). Water transport at the extreme–restoring the hydraulic system in a resurrection plant. The New Phytologist, 148(2), 187-193. Chen, Q., Shi, Q., Gorb, S. N., & Li, Z. (2014). A multiscale study on the structural and mechanical properties of the luffa sponge from Luffa cylindrica plant. Journal of biomechanics, 47(6), 1332-1339. Chen, X., Zeng, Q., & Wood, A. J. (2002). The stress-responsive Tortula ruralis gene ALDH21A1 describes a novel eukaryotic aldehyde dehydrogenase protein family. Journal of plant physiology, 159(7), 677-684. Collett, H., Butowt, R., Smith, J., Farrant, J., & Illing, N. (2003). Photosynthetic genes are differentially transcribed during the dehydration-rehydration cycle in the resurrection plant, Xerophyta humilis. Journal of Experimental Botany, 54(392), 2593-2595. Collett, H., Shen, A., Gardner, M., Farrant, J. M., Denby, K. J., & Illing, N. (2004). Towards transcript profiling of desiccation tolerance in Xerophyta humilis: construction of a normalized 11 k X. humilis cDNA set and microarray expression analysis of 424 cDNAs in response to dehydration. Physiologia plantarum, 122(1), 39-53. Cooper, K., & Farrant, J. M. (2002). Recovery of the resurrection plant Craterostigma wilmsii from desiccation: protection versus repair. J Exp Bot, 53(375), 1805- 1813. Cosgrove, D. J. (2005). Growth of the plant cell wall. Nat Rev Mol Cell Biol, 6(11), 850- 861. doi:10.1038/nrm1746 Cosgrove, D. J. (2016). Plant cell wall extensibility: connecting plant cell growth with cell wall structure, mechanics, and the action of wall-modifying enzymes. J Exp Bot, 67(2), 463-476. doi:10.1093/jxb/erv511 Cosgrove, D. J., & Jarvis, M. C. (2012). Comparative structure and biomechanics of plant primary and secondary cell walls. Front Plant Sci, 3, 204. doi:10.3389/fpls.2012.00204 Costa, M.-C. D., Cooper, K., Hilhorst, H. W., & Farrant, J. M. (2017). Orthodox seeds and resurrection plants: two of a kind? Plant Physiology, 175(2), 589-599.

208

Dace, H., Sherwin, H. W., Illing, N., & Farrant, J. M. (1998). Use of metabolic inhibitors to elucidate mechanisms of recovery from desiccation stress in the resurrection plant Xerophyta humilis. Plant Growth Regulation, 24(3), 171- 177. Dawson, C., Vincent, J. F., & Rocca, A.-M. (1997). How pine cones open. Nature, 390(6661), 668. de Haan, L. T., Verjans, J. M., Broer, D. J., Bastiaansen, C. W., & Schenning, A. P. (2014). Humidity-responsive liquid crystalline polymer actuators with an asymmetry in the molecular trigger that bend, fold, and curl. Journal of the American Chemical Society, 136(30), 10585-10588. De, M. G. (1955). Velvet type fabric and method of producing same: Google Patents. Deeba, F., Pandey, A. K., & Pandey, V. (2016). Organ specific proteomic dissection of Selaginella bryopteris undergoing dehydration and rehydration. Frontiers in plant science, 7, 425. Deeba, F., Pandey, V., Pathre, U., & Kanojiya, S. (2009). Proteome analysis of detached fronds from a resurrection plant Selaginella bryopteris–response to dehydration and rehydration. J Proteomics Bioinform, 2(2), 108-116. Deravi, L. F., Magyar, A. P., Sheehy, S. P., Bell, G. R., Mäthger, L. M., Senft, S. L., . . . Hanlon, R. T. (2014). The structure–function relationships of a natural nanoscale photonic device in cuttlefish chromatophores. Journal of the Royal Society Interface, 11(93), 20130942. Dharmawardhana, D., Ellis, B., & Carlson, J. (1992). Characterization of vascular lignification in Arabidopsis thaliana. Canadian Journal of Botany, 70(11), 2238-2244. Di Luca, M., Mintchev, S., Heitz, G., Noca, F., & Floreano, D. (2017). Bioinspired morphing wings for extended flight envelope and roll control of small drones. Interface focus, 7(1), 20160092. Dieing, T., & Hollricher, O. (2008). High-resolution, high-speed confocal Raman imaging. Vibrational Spectroscopy, 48(1), 22-27. Dinakar, C., & Bartels, D. (2013). Desiccation tolerance in resurrection plants: new insights from transcriptome, proteome and metabolome analysis. Front Plant Sci, 4, 482. doi:10.3389/fpls.2013.00482 Dinakar, C., Djilianov, D., & Bartels, D. (2012). Photosynthesis in desiccation tolerant plants: energy metabolism and antioxidative stress defense. Plant Sci, 182, 29-41. doi:10.1016/j.plantsci.2011.01.018 Ding, Z., Yuan, C., Peng, X., Wang, T., Qi, H. J., & Dunn, M. L. (2017). Direct 4D printing via active composite materials. Science advances, 3(4), e1602890. Dixon, P. G., & Gibson, L. J. (2014). The structure and mechanics of Moso bamboo material. Journal of the Royal Society Interface, 11(99), 20140321. du Sablon, M. (1888). Sur la réviviscence du Selaginella lepidophylla. B. Soc. Bot. FR, 35, 109-112. Dumais, J., & Forterre, Y. (2012). “Vegetable Dynamicks”: the role of water in plant movements. Annual Review of Fluid Mechanics, 44, 453-478. Dunlop, J. W., Weinkamer, R., & Fratzl, P. (2011). Artful interfaces within biological materials. Materials Today, 14(3), 70-78.

209

Eickmeier, W. G. (1980). Photosynthetic recovery of resurrection spikemosses from different hydration regimes. Oecologia, 46(3), 380-385. Elbaum, R., & Abraham, Y. (2014). Insights into the microstructures of hygroscopic movement in plant seed dispersal. Plant Science, 223, 124-133. doi:http://dx.doi.org/10.1016/j.plantsci.2014.03.014 Elbaum, R., Gorb, S., & Fratzl, P. (2008). Structures in the cell wall that enable hygroscopic movement of wheat awns. J Struct Biol, 164(1), 101-107. doi:10.1016/j.jsb.2008.06.008 Elbaum, R., Zaltzman, L., Burgert, I., & Fratzl, P. (2007). The role of wheat awns in the seed dispersal unit. Science, 316(5826), 884-886. doi:10.1126/science.1140097 Erb, R. M., Sander, J. S., Grisch, R., & Studart, A. R. (2013). Self-shaping composites with programmable bioinspired microstructures. Nat Commun, 4, 1712. doi:10.1038/ncomms2666 Evangelista, D., Hotton, S., & Dumais, J. (2011). The mechanics of explosive dispersal and self-burial in the seeds of the filaree, Erodium cicutarium (Geraniaceae). Journal of Experimental Biology, 214(4), 521-529. Faisal, T. R., Rey, A. D., & Pasini, D. (2013). A multiscale mechanical model for plant tissue stiffness. Polymers, 5(2), 730-750. Farrant, J., Vander Willigen, C., Loffell, D., Bartsch, S., & Whittaker, A. (2003). An investigation into the role of light during desiccation of three angiosperm resurrection plants. Plant, cell & environment, 26(8), 1275-1286. Farrant, J. M. (2000). A comparison of mechanisms of desiccation tolerance among three angiosperm resurrection plant species. Plant Ecology, 151(1), 29-39. Farrant, J. M., & Kruger, L. (2001). Longevity of dry Myrothamnus flabellifolius in simulated field conditions. Plant Growth Regulation, 35(2), 109-120. Farrant, J. M., & Moore, J. P. (2011). Programming desiccation-tolerance: from plants to seeds to resurrection plants. Curr Opin Plant Biol, 14(3), 340-345. doi:10.1016/j.pbi.2011.03.018 Florey, E., & Kriebel, M. E. (1969). Electrical and mechanical responses of chromatophore muscle fibers of the squid, Loligo opalescens, to nerve stimulation and drugs. Zeitschrift für vergleichende Physiologie, 65(1), 98-130. Forterre, Y. (2013). Slow, fast and furious: understanding the physics of plant movements. J Exp Bot, 64(15), 4745-4760. doi:10.1093/jxb/ert230 Forterre, Y., Skotheim, J. M., Dumais, J., & Mahadevan, L. (2005). How the Venus flytrap snaps. Nature, 433(7024), 421-425. doi:10.1038/nature03185 Fratzl, P., & Barth, F. G. (2009). Biomaterial systems for mechanosensing and actuation. Nature, 462(7272), 442-448. doi:10.1038/nature08603 Friedman, J., Gunderman, N., & Ellis, M. (1978). Water response of the hygrochastic skeletons of the true rose of Jericho (Anastatica hierochuntica L.). Oecologia, 32(3), 289-301. Fukuda, H. (2004). Signals that control plant vascular cell differentiation. Nature reviews Molecular cell biology, 5(5), 379. Gardiner, B., Berry, P., & Moulia, B. (2016). Review: Wind impacts on plant growth, mechanics and damage. Plant Science, 245, 94-118.

210

Gasulla, F., Dorp, K., Dombrink, I., Zähringer, U., Gisch, N., Dörmann, P., & Bartels, D. (2013). The role of lipid metabolism in the acquisition of desiccation tolerance in Craterostigma plantagineum: a comparative approach. The Plant Journal, 75(5), 726-741. Ge, Q., Sakhaei, A. H., Lee, H., Dunn, C. K., Fang, N. X., & Dunn, M. L. (2016). Multimaterial 4D printing with tailorable shape memory polymers. Scientific reports, 6, 31110. Gechev, T. S., Dinakar, C., Benina, M., Toneva, V., & Bartels, D. (2012). Molecular mechanisms of desiccation tolerance in resurrection plants. Cell Mol Life Sci, 69(19), 3175-3186. doi:10.1007/s00018-012-1088-0 Geitmann, A., & Ortega, J. K. (2009). Mechanics and modeling of plant cell growth. Trends Plant Sci, 14(9), 467-478. doi:10.1016/j.tplants.2009.07.006 Georgieva, T., Christov, N. K., & Djilianov, D. (2012). Identification of desiccation- regulated genes by cDNA-AFLP in Haberlea rhodopensis: a resurrection plant. Acta physiologiae plantarum, 34(3), 1055-1066. Ghavami, K., Rodrigues, C. d. S., & Paciornik, S. (2003). Bamboo: functionally graded composite material. Giarola, V., & Bartels, D. (2015). What can we learn from the transcriptome of the resurrection plant Craterostigma plantagineum? Planta, 242(2), 427-434. Gibson, L. J. (2012). The hierarchical structure and mechanics of plant materials. J R Soc Interface, 9(76), 2749-2766. doi:10.1098/rsif.2012.0341 Gierlinger, N., Keplinger, T., & Harrington, M. (2012). Imaging of plant cell walls by confocal Raman microscopy. nature protocols, 7(9), 1694. Gladman, A. S., Matsumoto, E. A., Nuzzo, R. G., Mahadevan, L., & Lewis, J. A. (2016). Biomimetic 4D printing. Nature materials, 15(4), 413. Grabber, J. H. (2005). How do lignin composition, structure, and cross-linking affect degradability? A review of cell wall model studies. Crop Science, 45(3), 820- 831. Grantham, N. J., Wurman-Rodrich, J., Terrett, O. M., Lyczakowski, J. J., Stott, K., Iuga, D., . . . Dupree, R. (2017). An even pattern of xylan substitution is critical for interaction with cellulose in plant cell walls. Nature plants, 3(11), 859. Green, D., Ben-Nissan, B., Yoon, K.-S., Milthorpe, B., & Jung, H.-S. (2016). Bioinspired materials for regenerative medicine: going beyond the human archetypes. Journal of Materials Chemistry B, 4(14), 2396-2406. Guiducci, L. (2014). Passive biomimetic actuators: the role of material architecture. Universität Potsdam Potsdam. Guiducci, L., Fratzl, P., Bréchet, Y. J., & Dunlop, J. W. (2014). Pressurized honeycombs as soft-actuators: a theoretical study. Journal of the Royal Society Interface, 11(98), 20140458. Guo, J., Li, C., Ling, S., Huang, W., Chen, Y., & Kaplan, D. L. (2017). Multiscale design and synthesis of biomimetic gradient protein/biosilica composites for interfacial tissue engineering. Biomaterials, 145, 44-55. Gutterman, Y., & Ginott, S. (1994). Long-term protected'seed bank'in dry inflorescences of Asteriscus pygmaeus; achene dispersal mechanism and germination. Journal of Arid Environments, 26(2), 149-163.

211

Habibi, M. K., Samaei, A. T., Gheshlaghi, B., Lu, J., & Lu, Y. (2015). Asymmetric flexural behavior from bamboo’s functionally graded hierarchical structure: Underlying mechanisms. Acta biomaterialia, 16, 178-186. Hall, H. C., Cheung, J., & Ellis, B. E. (2013). Immunoprofiling reveals unique cell- specific patterns of wall epitopes in the expanding Arabidopsis stem. The Plant Journal, 74(1), 134-147. Harholt, J., Sorensen, I., Fangel, J., Roberts, A., Willats, W. G., Scheller, H. V., . . . Ulvskov, P. (2012). The glycosyltransferase repertoire of the spikemoss Selaginella moellendorffii and a comparative study of its cell wall. PLoS One, 7(5), e35846. doi:10.1371/journal.pone.0035846 Harrington, M. J., Razghandi, K., Ditsch, F., Guiducci, L., Rueggeberg, M., Dunlop, J. W., . . . Burgert, I. (2011). Origami-like unfolding of hydro-actuated ice plant seed capsules. Nat Commun, 2, 337. doi:10.1038/ncomms1336 Harten, J. B., & Eickmeier, W. G. (1986). Enzyme dynamics of the resurrection plant Selaginella lepidophylla (Hook. & Grev.) spring during rehydration. Plant Physiology, 82(1), 61-64. Hedderson, N., Balsamo, R., Cooper, K., & Farrant, J. (2009). Leaf tensile properties of resurrection plants differ among species in their response to drying. South African Journal of Botany, 75(1), 8-16. Hegazy, A. K., Barakat, H. N., & Kabiel, H. F. (2006). Anatomical significance of the hygrochastic movement in Anastatica hierochuntica. Ann Bot, 97(1), 47-55. doi:10.1093/aob/mcj011 Helseth, L. (2008). Elastic response of the resurrection fern Polypodium polypodioides during rehydration. Applied Physics Letters, 92(4), 043902. Helseth, L., & Fischer, T. (2005). Physical mechanisms of rehydration in Polypodium polypodioides, a resurrection plant. Physical review E, 71(6), 061903. Hodick, D., & Sievers, A. (1989). On the mechanism of trap closure of Venus flytrap (Dionaea muscipula Ellis). Planta, 179(1), 32-42. Höfte, H., & Voxeur, A. (2017). Plant cell walls. Current biology, 27(17), R865-R870. Holstov, A., Bridgens, B., & Farmer, G. (2015). Hygromorphic materials for sustainable responsive architecture. Construction and Building Materials, 98, 570-582. Hößl, B., Böhm, H. J., Rammerstorfer, F. G., & Barth, F. G. (2007). Finite element modeling of arachnid slit sensilla—I. The mechanical significance of different slit arrays. Journal of Comparative Physiology A, 193(4), 445-459. Hößl, B., Böhm, H. J., Rammerstorfer, F. G., Müllan, R., & Barth, F. G. (2006). Studying the deformation of arachnid slit sensilla by a fracture mechanical approach. Journal of biomechanics, 39(10), 1761-1768. Huang, W. M., Song, C. L., Fu, Y. Q., Wang, C. C., Zhao, Y., Purnawali, H., . . . Zhang, J. L. (2013). Shaping tissue with shape memory materials. Adv Drug Deliv Rev, 65(4), 515-535. doi:10.1016/j.addr.2012.06.004 Ingle, R. A., Schmidt, U. G., Farrant, J. M., Thomson, J. A., & Mundree, S. G. (2007). Proteomic analysis of leaf proteins during dehydration of the resurrection plant Xerophyta viscosa. Plant, cell & environment, 30(4), 435-446. Ionov, L. (2011). Soft microorigami: self-folding polymer films. Soft Matter, 7(15), 6786-6791. 212

Ionov, L. (2013). Biomimetic Hydrogel-Based Actuating Systems. Advanced Functional Materials, 23(36), 4555-4570. Ionov, L. (2014). Polymer origami: programming the folding with shape. e-Polymers, 14(2), 109-114. Ionov, L. (2017). Actuating Hydrogel Thin Films. Responsive Polymer Surfaces: Dynamics in Surface Topography. Iturriaga, G., Cushman, M. A. F., & Cushman, J. C. (2006). An EST catalogue from the resurrection plant Selaginella lepidophylla reveals abiotic stress-adaptive genes. Plant Science, 170(6), 1173-1184. Jensen, O. E., & Fozard, J. A. (2015). Multiscale Models in the Biomechanics of Plant Growth. Physiology, 30(2), 159-166. doi:10.1152/physiol.00030.2014 Jiang, G., Wang, Z., Shang, H., Yang, W., Hu, Z., Phillips, J., & Deng, X. (2007). Proteome analysis of leaves from the resurrection plant Boea hygrometrica in response to dehydration and rehydration. Planta, 225(6), 1405. Johnson, E., Bonser, R., & Jeronimidis, G. (2009). Recent advances in biomimetic sensing technologies. Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences, 367(1893), 1559- 1569. Julius, A. A., Sakar, M. S., Steager, E., Cheang, U. K., Kim, M., Kumar, V., & Pappas, G. J. (2009). Harnessing bacterial power in microscale actuation. Paper presented at the Robotics and Automation, 2009. ICRA'09. IEEE International Conference on. Kapp, N., Barnes, W. J., Richard, T. L., & Anderson, C. T. (2015). Imaging with the fluorogenic dye Basic Fuchsin reveals subcellular patterning and ecotype variation of lignification in Brachypodium distachyon. Journal of Experimental Botany, 66(14), 4295-4304. Karshalev, E., Kumar, R., Jeerapan, I., Castillo, R., Campos, I., & Wang, J. (2018). Multi- Stimuli Responsive Camouflage Swimmers. Chemistry of Materials. Kempaiah, R., & Nie, Z. (2014). From nature to synthetic systems: shape transformation in soft materials. Journal of Materials Chemistry B, 2(17), 2357-2368. Kennedy, E., Fecheyr-Lippens, D., Hsiung, B.-K., Niewiarowski, P. H., & Kolodziej, M. (2015). Biomimicry: A path to sustainable innovation. Design Issues, 31(3), 66-73. Kim, D. H., Cheang, U. K., Kőhidai, L., Byun, D., & Kim, M. J. (2010). Artificial magnetotactic motion control of Tetrahymena pyriformis using ferromagnetic nanoparticles: A tool for fabrication of microbiorobots. Applied Physics Letters, 97(17), 173702. Kim, K., Yi, H., Zamil, M. S., Haque, M. A., & Puri, V. M. (2015). Multiscale stress-strain characterization of onion outer epidermal tissue in wet and dry states. Am J Bot, 102(1), 12-20. doi:10.3732/ajb.1400273 Kim, S., Laschi, C., & Trimmer, B. (2013). Soft robotics: a bioinspired evolution in robotics. Trends in biotechnology, 31(5), 287-294. Koetting, M. C., Peters, J. T., Steichen, S. D., & Peppas, N. A. (2015). Stimulus- responsive hydrogels: Theory, modern advances, and applications. Materials Science and Engineering: R: Reports, 93, 1-49. 213

Korall, P., & Kenrick, P. (2002). Phylogenetic relationships in Selaginellaceae based on RBCL sequences. Am J Bot, 89(3), 506-517. doi:10.3732/ajb.89.3.506 Korte, N., & Porembski, S. (2012). A morpho-anatomical characterisation of Myrothamnus moschatus (Myrothamnaceae) under the aspect of desiccation tolerance. Plant Biology, 14(3), 537-541. Kothera, C. S., Woods, B. K., Sirohi, J., Wereley, N. M., & Chen, P. C. (2010). Fluid- driven artificial muscles as mechanisms for controlled actuation: Google Patents. Kranner, I., Beckett, R. P., Wornik, S., Zorn, M., & Pfeifhofer, H. W. (2002). Revival of a resurrection plant correlates with its antioxidant status. The Plant Journal, 31(1), 13-24. Lafarguette, F., Leplé, J. C., Déjardin, A., Laurans, F., Costa, G., Lesage-Descauses, M. C., & Pilate, G. (2004). Poplar genes encoding fasciclin-like arabinogalactan proteins are highly expressed in tension wood. New Phytologist, 164(1), 107- 121. Langer, T., Belyaev, A. G., & Seidel, H.-P. (2005). Asymptotic analysis of discrete normals and curvatures of polylines. Paper presented at the Proceedings of the 21st spring conference on Computer graphics. Layton, B. E., Boyd, M. B., Tripepi, M. S., Bitonti, B. M., Dollahon, M. N. R., & Balsamo, R. A. (2010). Dehydration-induced expression of a 31-kDa dehydrin in Polypodium polypodioides (Polypodiaceae) may enable large, reversible deformation of cell walls. American journal of botany, 97(4), 535-544. Lebkuecher, J. G., & Eickmeier, W. G. (1991). Reduced photoinhibition with stem curling in the resurrection plant Selaginella lepidophylla. Oecologia, 88(4), 597-604. Lebkuecher, J. G., & Eickmeier, W. G. (1993). Physiological benefits of stem curling for resurrection plants in the field. Ecology, 1073-1080. Leprince, O., & Buitink, J. (2010). Desiccation tolerance: from genomics to the field. Plant Science, 179(6), 554-564. Li, D., Choi, H., Cho, S., Jeong, S., Jin, Z., Lee, C., . . . Park, S. (2015). A hybrid actuated microrobot using an electromagnetic field and flagellated bacteria for tumor- targeting therapy. Biotechnology and bioengineering, 112(8), 1623-1631. Li, S., Vogt, D. M., Rus, D., & Wood, R. J. (2017). Fluid-driven origami-inspired artificial muscles. Proceedings of the National Academy of Sciences, 201713450. Liu, D., Song, J., Anderson, D. P., Chang, P. R., & Hua, Y. (2012). Bamboo fiber and its reinforced composites: structure and properties. Cellulose, 19(5), 1449-1480. Liu, M. S., Chien, C. T., & Lin, T. P. (2008). Constitutive components and induced gene expression are involved in the desiccation tolerance of Selaginella tamariscina. Plant Cell Physiol, 49(4), 653-663. doi:10.1093/pcp/pcn040 Lüttge, U., Beck, E., & Bartels, D. (2011). Plant desiccation tolerance (Vol. 215): Springer Science & Business Media. Lv, C., Sun, X.-C., Xia, H., Yu, Y.-H., Wang, G., Cao, X.-W., . . . Yu, Y.-D. (2018). Humidity- responsive actuation of programmable hydrogel microstructures based on 3D printing. Sensors and Actuators B: Chemical, 259, 736-744.

214

Lv, H., Leng, J., Liu, Y., & Du, S. (2008). Shape-memory polymer in response to solution. Advanced Engineering Materials, 10(6), 592-595. Ma, C., Lu, W., Yang, X., He, J., Le, X., Wang, L., . . . Chen, T. (2018). Bioinspired Anisotropic Hydrogel Actuators with On–Off Switchable and Color-Tunable Fluorescence Behaviors. Advanced Functional Materials. Ma, C., Wang, H., Macnish, A. J., Estrada-Melo, A. C., Lin, J., Chang, Y., . . . Jiang, C.-Z. (2015). Transcriptomic analysis reveals numerous diverse protein kinases and transcription factors involved in desiccation tolerance in the resurrection plant Myrothamnus flabellifolia. Horticulture research, 2, 15034. Ma, M., Guo, L., Anderson, D. G., & Langer, R. (2013). Bio-inspired polymer composite actuator and generator driven by water gradients. Science, 339(6116), 186- 189. Magna, R. L., Gabler, M., Reichert, S., Schwinn, T., Waimer, F., Menges, A., & Knippers, J. (2013). From nature to fabrication: biomimetic design principles for the production of complex spatial structures. International Journal of Space Structures, 28(1), 27-39. Mao, Y., Ding, Z., Yuan, C., Ai, S., Isakov, M., Wu, J., . . . Qi, H. J. (2016). 3D printed reversible shape changing components with stimuli responsive materials. Scientific reports, 6, 24761. Marjamaa, K., Lehtonen, M., Lundell, T., Toikka, M., Saranpää, P., & Fagerstedt, K. V. (2003). Developmental lignification and seasonal variation in β-glucosidase and peroxidase activities in xylem of Scots pine, Norway spruce and silver birch. Tree Physiology, 23(14), 977-986. Marmottant, P., Ponomarenko, A., & Bienaimé, D. (2013). The walk and jump of Equisetum spores. Proc. R. Soc. B, 280(1770), 20131465. Martel, S., Mohammadi, M., Felfoul, O., Lu, Z., & Pouponneau, P. (2009). Flagellated magnetotactic bacteria as controlled MRI-trackable propulsion and steering systems for medical nanorobots operating in the human microvasculature. The International Journal of Robotics Research, 28(4), 571-582. Martel, S., Tremblay, C. C., Ngakeng, S., & Langlois, G. (2006). Controlled manipulation and actuation of micro-objects with magnetotactic bacteria. Applied Physics Letters, 89(23), 233904. McCartney, L., Marcus, S. E., & Knox, J. P. (2005). Monoclonal antibodies to plant cell wall xylans and arabinoxylans. Journal of Histochemistry & Cytochemistry, 53(4), 543-546. McFarlane, H. E., Döring, A., & Persson, S. (2014). The cell biology of cellulose synthesis. Annual Review of Plant Biology, 65, 69-94. Meloche, C. G., Knox, J. P., & Vaughn, K. C. (2007). A cortical band of gelatinous fibers causes the coiling of redvine tendrils: a model based upon cytochemical and immunocytochemical studies. Planta, 225(2), 485-498. Meng, H., & Li, G. (2013). A review of stimuli-responsive shape memory polymer composites. Polymer, 54(9), 2199-2221. Mikkonen, K. S. (2013). Recent studies on hemicellulose-based blends, composites and nanocomposites Advances in Natural Polymers (pp. 313-336): Springer.

215

Milani, P., Braybrook, S. A., & Boudaoud, A. (2013). Shrinking the hammer: micromechanical approaches to morphogenesis. Journal of Experimental Botany, ert169. Milwich, M., Speck, T., Speck, O., Stegmaier, T., & Planck, H. (2006). Biomimetics and technical textiles: solving engineering problems with the help of nature's wisdom. Am J Bot, 93(10), 1455-1465. doi:10.3732/ajb.93.10.1455 Mirabet, V., Das, P., Boudaoud, A., & Hamant, O. (2011). The role of mechanical forces in plant morphogenesis. Annu Rev Plant Biol, 62, 365-385. doi:10.1146/annurev-arplant-042110-103852 Mitra, P. P., & Loqué, D. (2014). Histochemical staining of Arabidopsis thaliana secondary cell wall elements. Journal of visualized experiments: JoVE(87). Mladenov, P., Finazzi, G., Bligny, R., Moyankova, D., Zasheva, D., Boisson, A.-M., . . . Simova, S. (2015). In vivo spectroscopy and NMR metabolite fingerprinting approaches to connect the dynamics of photosynthetic and metabolic phenotypes in resurrection plant Haberlea rhodopensis during desiccation and recovery. Frontiers in plant science, 6, 564. Mladenov, P., Zasheva, D., Peshev, D., Christov, N., Rolland, N., Djilianov, D., & Tchorbadjieva, M. (2013). Sub-cellular fractionation and gel-based proteomics of Haberlea rhodopensis: a promising approach to open the black box of resurrection plants. Bulg. J. Agric. Sci, 19(2), 22-25. Moore, J. P., Farrant, J. M., & Driouich, A. (2008). A role for pectin-associated arabinans in maintaining the flexibility of the plant cell wall during water deficit stress. Plant Signal Behav, 3(2), 102-104. Moore, J. P., Hearshaw, M., Ravenscroft, N., Lindsey, G. G., Farrant, J. M., & Brandt, W. F. (2007). Desiccation-induced ultrastructural and biochemical changes in the leaves of the resurrection plant Myrothamnus flabellifolia. Australian Journal of Botany, 55(4), 482-491. Moore, J. P., Lindsey, G. G., Farrant, J. M., & Brandt, W. F. (2007). An overview of the biology of the desiccation-tolerant resurrection plant Myrothamnus flabellifolia. Ann Bot, 99(2), 211-217. doi:10.1093/aob/mcl269 Moore, J. P., Nguema-Ona, E., Chevalier, L., Lindsey, G. G., Brandt, W. F., Lerouge, P., . . . Driouich, A. (2006). Response of the leaf cell wall to desiccation in the resurrection plant Myrothamnus flabellifolius. Plant Physiol, 141(2), 651-662. doi:10.1104/pp.106.077701 Moore, J. P., Nguema-Ona, E. E., Vicre-Gibouin, M., Sorensen, I., Willats, W. G., Driouich, A., & Farrant, J. M. (2013). Arabinose-rich polymers as an evolutionary strategy to plasticize resurrection plant cell walls against desiccation. Planta, 237(3), 739-754. doi:10.1007/s00425-012-1785-9 Moore, J. P., Vicre-Gibouin, M., Farrant, J. M., & Driouich, A. (2008). Adaptations of higher plant cell walls to water loss: drought vs desiccation. Physiol Plant, 134(2), 237-245. doi:10.1111/j.1399-3054.2008.01134.x Morse, M., Rafudeen, M. S., & Farrant, J. M. (2011). An overview of the current understanding of desiccation tolerance in the vegetative tissues of higher plants Advances in Botanical Research (Vol. 57, pp. 319-347): Elsevier.

216

Moulia, B. (2013). Plant biomechanics and mechanobiology are convergent paths to flourishing interdisciplinary research. Journal of Experimental Botany, 64(15), 4617-4633. Moyankova, D., Mladenov, P., Berkov, S., Peshev, D., Georgieva, D., & Djilianov, D. (2014). Metabolic profiling of the resurrection plant Haberlea rhodopensis during desiccation and recovery. Physiologia plantarum, 152(4), 675-687. Mutlu, R., Alici, G., & Li, W. (2011). Kinematic modeling for artificial flagellum of a robotic bacterium based on electroactive polymer actuators. Paper presented at the Advanced Intelligent Mechatronics (AIM), 2011 IEEE/ASME International Conference on. Naleway, S. E., Porter, M. M., McKittrick, J., & Meyers, M. A. (2015). Structural design elements in biological materials: application to bioinspiration. Advanced materials, 27(37), 5455-5476. Neale, A., Blomstedt, C., Bronson, P., Le, T. N., Guthridge, K., Evans, J., . . . Hamill, J. (2000). The isolation of genes from the resurrection grass Sporobolus stapfianus which are induced during severe drought stress. Plant, cell & environment, 23(3), 265-277. Niklas, K. J. (1992). Plant biomechanics: an engineering approach to plant form and function: University of Chicago press. Nothnagel, A. L., & Nothnagel, E. A. (2007). Primary cell wall structure in the evolution of land plants. Journal of Integrative Plant Biology, 49(8), 1271- 1278. Oliver, M. J., Dowd, S. E., Zaragoza, J., Mauget, S. A., & Payton, P. R. (2004). The rehydration transcriptome of the desiccation-tolerant bryophyte Tortula ruralis: transcript classification and analysis. Bmc Genomics, 5(1), 89. Oliver, M. J., Jain, R., Balbuena, T. S., Agrawal, G., Gasulla, F., & Thelen, J. J. (2011). Proteome analysis of leaves of the desiccation-tolerant grass, Sporobolus stapfianus, in response to dehydration. Phytochemistry, 72(10), 1273-1284. Oliver, M. J., Murdock, A. G., Mishler, B. D., Kuehl, J. V., Boore, J. L., Mandoli, D. F., . . . Karol, K. G. (2010). Chloroplast genome sequence of the moss Tortula ruralis: gene content, polymorphism, and structural arrangement relative to other green plant chloroplast genomes. Bmc Genomics, 11(1), 143. Oliver, M. J., Tuba, Z., & Mishler, B. D. (2000). The evolution of vegetative desiccation tolerance in land plants. Plant Ecology, 151(1), 85-100. Pandey, V., Ranjan, S., Deeba, F., Pandey, A. K., Singh, R., Shirke, P. A., & Pathre, U. V. (2010). Desiccation-induced physiological and biochemical changes in resurrection plant, Selaginella bryopteris. Journal of plant physiology, 167(16), 1351-1359. Parameswaran, N., & Liese, W. (1976). On the fine structure of bamboo fibres. Wood Science and Technology, 10(4), 231-246. Paris, O., Burgert, I., & Fratzl, P. (2010). Biomimetics and biotemplating of natural materials. Mrs Bulletin, 35(3), 219-225. Passino, K. M. (2002). Biomimicry of bacterial foraging for distributed optimization and control. IEEE control systems, 22(3), 52-67.

217

Pattathil, S., Avci, U., Baldwin, D., Swennes, A. G., McGill, J. A., Popper, Z., . . . Chennareddy, C. (2010). A comprehensive toolkit of plant cell wall glycan- directed monoclonal antibodies. Plant Physiology, 153(2), 514-525. Perera, P. N., Schmidt, M., Chiang, V. L., Schuck, P. J., & Adams, P. D. (2012). Raman- spectroscopy-based noninvasive microanalysis of native lignin structure. Analytical and bioanalytical chemistry, 402(2), 983-987. Persson, K. (2000). Micromechanical modelling of wood and fibre properties (Vol. 1013): Division of Structural Mechanics, Lund Institute of Technology. Peterson, R. L., Peterson, C. A., & Melville, L. H. (2008). Teaching plant anatomy through creative laboratory exercises: NRC Research Press. Pilate, G., Chabbert, B., Cathala, B., Yoshinaga, A., Leplé, J.-C., Laurans, F., . . . Ruel, K. (2004). Lignification and tension wood. Comptes rendus biologies, 327(9-10), 889-901. Platt, K., Oliver, M., & Thomson, W. (1997). Importance of the Fixative for Reliable Ultrastructural Preservation of Poikilohydric Plant Tissues. Observations on Dry, Partially, and Fully Hydrated Tissues ofSelaginella lepidophylla. Annals of Botany, 80(5), 599-610. Potts, R., & Penfound, W. T. (1948). Water relations of the polypody fern, Polypodium polypodioides (L.) AS Hitchcock. Ecology, 29(1), 43-53. Rafsanjani, A., Brulé, V., Western, T. L., & Pasini, D. (2015). Hydro-Responsive Curling of the Resurrection Plant Selaginella lepidophylla. Scientific reports, 5. Rascio, N., & Rocca, N. L. (2005). Resurrection plants: the puzzle of surviving extreme vegetative desiccation. Critical Reviews in Plant Sciences, 24(3), 209- 225. Razghandi, K., Bertinetti, L., Guiducci, L., Dunlop, J. W., Fratzl, P., Neinhuis, C., & Burgert, I. (2014). Hydro-actuation of ice plant seed capsules powered by water uptake. Bioinspired, Biomimetic and Nanobiomaterials, 3(3), 169-182. Reichert, S., Menges, A., & Correa, D. (2015). Meteorosensitive architecture: Biomimetic building skins based on materially embedded and hygroscopically enabled responsiveness. Computer-Aided Design, 60, 50-69. Reyssat, E., & Mahadevan, L. (2009). Hygromorphs: from pine cones to biomimetic bilayers. Journal of the Royal Society Interface, rsif20090184. Richter, S., Müssig, J., & Gierlinger, N. (2011). Functional plant cell wall design revealed by the Raman imaging approach. Planta, 233(4), 763-772. Rodriguez, M. C. S., Edsgärd, D., Hussain, S. S., Alquezar, D., Rasmussen, M., Gilbert, T., . . . Mundy, J. (2010). Transcriptomes of the desiccation-tolerant resurrection plant Craterostigma plantagineum. The Plant Journal, 63(2), 212-228. Röhrig, H., Colby, T., Schmidt, J., Harzen, A., Facchinelli, F., & Bartels, D. (2008). Analysis of desiccation-induced candidate phosphoproteins from Craterostigma plantagineum isolated with a modified metal oxide affinity chromatography procedure. Proteomics, 8(17), 3548-3560. Rohrig, H., Schmidt, J., Colby, T., Brautigam, A., Hufnagel, P., & Bartels, D. (2006). Desiccation of the resurrection plant Craterostigma plantagineum induces dynamic changes in protein phosphorylation. Plant, cell & environment, 29(8), 1606-1617.

218

Rossiter, J., Yap, B., & Conn, A. (2012). Biomimetic chromatophores for camouflage and soft active surfaces. Bioinspiration & biomimetics, 7(3), 036009. Ruzin, S. E. (1999). Plant microtechnique and microscopy (Vol. 198): Oxford University Press New York. Sakar, M. S., Steager, E. B., Kim, D. H., Agung Julius, A., Kim, M., Kumar, V., & Pappas, G. J. (2011). Modeling, control and experimental characterization of microbiorobots. The International Journal of Robotics Research, 30(6), 647- 658. Sarikaya, M. (1994). An introduction to biomimetics: a structural viewpoint. Microscopy research and technique, 27(5), 360-375. Sarkar, P., Bosneaga, E., & Auer, M. (2009). Plant cell walls throughout evolution: towards a molecular understanding of their design principles. Journal of Experimental Botany, 60(13), 3615-3635. Schleicher, S., Lienhard, J., Poppinga, S., Speck, T., & Knippers, J. (2015). A methodology for transferring principles of plant movements to elastic systems in architecture. Computer-Aided Design, 60, 105-117. Schmidt, M., Schwartzberg, A. M., Carroll, A., Chaibang, A., Adams, P. D., & Schuck, P. J. (2010). Raman imaging of cell wall polymers in Arabidopsis thaliana. Biochemical and biophysical research communications, 395(4), 521-523. Schneider, H., Manz, B., Westhoff, M., Mimietz, S., Szimtenings, M., Neuberger, T., . . . Volke, F. (2003). The impact of lipid distribution, composition and mobility on xylem water refilling of the resurrection plant Myrothamnus flabellifolia. New Phytologist, 159(2), 487-505. Schneider, H., Thürmer, F., Zhu, J., Wistuba, N., Gessner, P., Lindner, K., . . . Bentrup, F.-W. (1999). Diurnal changes in xylem pressure of the hydrated resurrection plant Myrothamnus flabellifolia: evidence for lipid bodies in conducting xylem vessels. The New Phytologist, 143(3), 471-484. Scott, P. (2000). Resurrection plants and the secrets of eternal leaf. Annals of Botany, 85(2), 159-166. Sehaqui, H., Zhou, Q., & Berglund, L. A. (2011). Nanostructured biocomposites of high toughness—a wood cellulose nanofiber network in ductile hydroxyethylcellulose matrix. Soft Matter, 7(16), 7342-7350. Sherwin, H. W., & Farrant, J. M. (1996). Differences in rehydration of three desiccation-tolerant angiosperm species. Annals of Botany, 78(6), 703-710. Sherwin, H. W., Pammenter, N., February, E., Vander Willigen, C., & Farrant, J. M. (1998). Xylem hydraulic characteristics, water relations and wood anatomy of the resurrection plant Myrothamnus flabellifolius Welw. Annals of Botany, 81(4), 567-575. Shtein, I., Bar-On, B., & Popper, Z. A. (2018). Plant and algal structure: from cell walls to biomechanical function. Physiologia plantarum. Siau, J. F. (1984). Wood Structure and Chemical Composition Transport Processes in Wood (pp. 35-72): Springer. Sinibaldi, E., Argiolas, A., Puleo, G. L., & Mazzolai, B. (2014). Another lesson from plants: the forward osmosis-based actuator. PLoS One, 9(7), e102461.

219

Skotheim, J. M., & Mahadevan, L. (2005). Physical limits and design principles for plant and fungal movements. Science, 308(5726), 1308-1310. doi:10.1126/science.1107976 Smith-Espinoza, C., Bartels, D., & Phillips, J. (2007). Analysis of a LEA gene promoter via Agrobacterium-mediated transformation of the desiccation tolerant plant Lindernia brevidens. Plant cell reports, 26(9), 1681-1688. Soni, D. K., Ranjan, S., Singh, R., Khare, P. B., Pathre, U. V., & Shirke, P. A. (2012). Photosynthetic characteristics and the response of stomata to environmental determinants and ABA in Selaginella bryopteris, a resurrection spike moss species. Plant Science, 191, 43-52. Speck, T., & Burgert, I. (2011). Plant stems: functional design and mechanics. Annual Review of Materials Research, 41, 169-193. Srinivasan, A., Haritos, G., & Hedberg, F. (1991). Biomimetics: Advancing man-made materials through guidance from nature. Applied Mechanics Reviews, 44(11), 463-482. Stahlberg, R. (2009). The phytomimetic potential of three types of hydration motors that drive nastic plant movements. Mechanics of Materials, 41(10), 1162- 1171. Stoychev, G., Zakharchenko, S., Turcaud, S. b., Dunlop, J. W., & Ionov, L. (2012). Shape-programmed folding of stimuli-responsive polymer bilayers. ACS nano, 6(5), 3925-3934. Studart, A. R. (2015). Biologically inspired dynamic material systems. Angewandte Chemie International Edition, 54(11), 3400-3416. Studart, A. R., & Erb, R. M. (2014). Bioinspired materials that self-shape through programmed microstructures. Soft Matter, 10(9), 1284-1294. Sun, L., Simmons, B. A., & Singh, S. (2011). Understanding tissue specific compositions of bioenergy feedstocks through hyperspectral Raman imaging. Biotechnology and bioengineering, 108(2), 286-295. Sun, R.-Z., Lin, C.-T., Zhang, X.-F., Duan, L.-X., Qi, X.-Q., Gong, Y.-H., & Deng, X. (2018). Acclimation-induced metabolic reprogramming contributes to rapid desiccation tolerance acquisition in Boea hygrometrica. Environmental and Experimental Botany, 148, 70-84. Synytsya, A., Čopı́ková, J., Matějka, P., & Machovič, V. (2003). Fourier transform Raman and infrared spectroscopy of pectins. Carbohydrate Polymers, 54(1), 97-106. Tan, T., Rahbar, N., Allameh, S. M., Kwofie, S., Dissmore, D., Ghavami, K., & Soboyejo, W. O. (2011). Mechanical properties of functionally graded hierarchical bamboo structures. Acta Biomater, 7(10), 3796-3803. doi:10.1016/j.actbio.2011.06.008 Temmei, Y., Uchida, S., Hoshino, D., Kanzawa, N., Kuwahara, M., Sasaki, S., & Tsuchiya, T. (2005). Water channel activities of Mimosa pudica plasma membrane intrinsic proteins are regulated by direct interaction and phosphorylation. FEBS letters, 579(20), 4417-4422. Thomson, W. W., & Platt, K. A. (1997). Conservation of cell order in desiccated mesophyll of Selaginella lepidophylla ([Hook and Grev.] Spring). Annals of Botany, 79(4), 439-447. 220

Timoshenko, S. (1925). Analysis of bi-metal thermostats. JOSA, 11(3), 233-255. Toldi, O., Tuba, Z., & Scott, P. (2009). Vegetative desiccation tolerance: is it a goldmine for bioengineering crops? Plant Science, 176(2), 187-199. Uphof, J. (1920). PHYSIOLOGICAL ANATOMY OF XEROPHYTIC SELAGINELLAS. New Phytologist, 19(5-6), 101-131. Vaia, R., & Baur, J. (2008). Adaptive composites. Science, 319(5862), 420-421. Vaughn, K. C., & Bowling, A. J. (2010). Biology and Physiology of Vines Horticultural Reviews (pp. 1-21): John Wiley & Sons, Inc. Velders, A. H., Dijksman, J. A., & Saggiomo, V. (2017). Hydrogel Actuators as Responsive Instruments for Cheap Open Technology (HARICOT). Applied Materials Today, 9, 271-275. Vergauwen, A., De Laet, L., & De Temmerman, N. (2017). Computational modelling methods for pliable structures based on curved-line folding. Computer-Aided Design, 83, 51-63. Vicre, M., Lerouxel, O., Farrant, J., Lerouge, P., & Driouich, A. (2004). Composition and desiccation-induced alterations of the cell wall in the resurrection plant Craterostigma wilmsii. Physiol Plant, 120(2), 229-239. doi:10.1111/j.0031- 9317.2004.0234.x Vicré, M., Sherwin, H., Driouich, A., Jaffer, M., & Farrant, J. (1999). Cell wall characteristics and structure of hydrated and dry leaves of the resurrection plant Craterostigma wilmsii, a microscopical study. Journal of plant physiology, 155(6), 719-726. Vincent, J. F. (1992). Biomechanics--materials: IRL Press at Oxford University Press. Vogler, H., Felekis, D., Nelson, B., & Grossniklaus, U. (2015). Measuring the Mechanical Properties of Plant Cell Walls. Plants, 4(2), 167. Volkov, A. G., Adesina, T., Markin, V. S., & Jovanov, E. (2008). Kinetics and mechanism of Dionaea muscipula trap closing. Plant Physiology, 146(2), 694-702. Volkov, A. G., Foster, J. C., Ashby, T. A., Walker, R. K., Johnson, J. A., & Markin, V. S. (2010). Mimosa pudica: electrical and mechanical stimulation of plant movements. Plant, cell & environment, 33(2), 163-173. Volkov, A. G., Foster, J. C., & Markin, V. S. (2010). Signal transduction in Mimosa pudica: biologically closed electrical circuits. Plant, cell & environment, 33(5), 816-827. Wang, M., Tian, X., Ras, R. H., & Ikkala, O. (2015). Sensitive Humidity-Driven Reversible and Bidirectional Bending of Nanocellulose Thin Films as Bio- Inspired Actuation. Advanced Materials Interfaces, 2(7). Wang, X., Chen, S., Zhang, H., Shi, L., Cao, F., Guo, L., . . . Dai, S. (2010). Desiccation tolerance mechanism in resurrection fern-ally Selaginella tamariscina revealed by physiological and proteomic analysis. Journal of proteome research, 9(12), 6561-6577. Wang, Y., Chantreau, M., Sibout, R., & Hawkins, S. (2013). Plant cell wall lignification and monolignol metabolism. Front Plant Sci, 4, 220. doi:10.3389/fpls.2013.00220 Wanntorp, L., Wanntorp, H.-E., Oxelman, B., & Källersjö, M. (2001). Phylogeny of Gunnera. Plant systematics and evolution, 226(1-2), 85-107.

221

Weng, J.-K., Akiyama, T., Bonawitz, N. D., Li, X., Ralph, J., & Chapple, C. (2010). Convergent evolution of syringyl lignin biosynthesis via distinct pathways in the lycophyte Selaginella and flowering plants. The Plant Cell, 22(4), 1033- 1045. Weng, J.-K., Li, X., Stout, J., & Chapple, C. (2008). Independent origins of syringyl lignin in vascular plants. Proceedings of the National Academy of Sciences, 105(22), 7887-7892. Willie, B. M., Petersen, A., Schmidt-Bleek, K., Cipitria, A., Mehta, M., Strube, P., . . . Duda, G. (2010). Designing biomimetic scaffolds for bone regeneration: why aim for a copy of mature tissue properties if nature uses a different approach? Soft Matter, 6(20), 4976-4987. Wood, A. J., Duff, R. J., & Oliver, M. J. (1999). Expressed sequence tags (ESTs) from desiccated Tortula ruralis identify a large number of novel plant genes. Plant and Cell Physiology, 40(4), 361-368. Xiao, L., Yang, G., Zhang, L., Yang, X., Zhao, S., Ji, Z., . . . Chen, M. (2015). The resurrection genome of Boea hygrometrica: A blueprint for survival of dehydration. Proceedings of the National Academy of Sciences, 112(18), 5833- 5837. Xu, L., Chen, H.-Q., Zou, J., Dong, W.-T., Gu, G.-Y., Zhu, L.-M., & Zhu, X.-Y. (2017). Bio- inspired annelid robot: a dielectric elastomer actuated soft robot. Bioinspiration & biomimetics, 12(2), 025003. Yobi, A., Schlauch, K. A., Tillett, R. L., Yim, W. C., Espinoza, C., Wone, B. W., . . . Oliver, M. J. (2017). Sporobolus stapfianus: Insights into desiccation tolerance in the resurrection grasses from linking transcriptomics to metabolomics. BMC plant biology, 17(1), 67. Yobi, A., Wone, B. W., Xu, W., Alexander, D. C., Guo, L., Ryals, J. A., . . . Cushman, J. C. (2012). Comparative metabolic profiling between desiccation-sensitive and desiccation-tolerant species of Selaginella reveals insights into the resurrection trait. Plant J, 72(6), 983-999. doi:10.1111/tpj.12008 Yobi, A., Wone, B. W., Xu, W., Alexander, D. C., Guo, L., Ryals, J. A., . . . Cushman, J. C. (2013). Metabolomic profiling in Selaginella lepidophylla at various hydration states provides new insights into the mechanistic basis of desiccation tolerance. Mol Plant, 6(2), 369-385. doi:10.1093/mp/sss155 Young, R. E., McFarlane, H. E., Hahn, M. G., Western, T. L., Haughn, G. W., & Samuels, A. L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. The Plant Cell, 20(6), 1623-1638. Yuk, H., Lin, S., Ma, C., Takaffoli, M., Fang, N. X., & Zhao, X. (2017). Hydraulic hydrogel actuators and robots optically and sonically camouflaged in water. Nature communications, 8, 14230. Zhang, L., Desta, I., & Naumov, P. (2016). Synergistic action of thermoresponsive and hygroresponsive elements elicits rapid and directional response of a bilayer actuator. Chemical Communications, 52(35), 5920-5923. Zhang, L., Qiu, X., Yuan, Y., & Zhang, T. (2017). Humidity-and Sunlight-Driven Motion of a Chemically Bonded Polymer Bilayer with Programmable Surface Patterns. ACS applied materials & interfaces, 9(47), 41599-41606. 222

Zhao, X. (2014). Multi-scale multi-mechanism design of tough hydrogels: building dissipation into stretchy networks. Soft Matter, 10(5), 672-687. Zhao, X., Wu, H., Guo, B., Dong, R., Qiu, Y., & Ma, P. X. (2017). Antibacterial antioxidant electroactive injectable hydrogel as self-healing wound dressing with hemostasis and adhesiveness for cutaneous wound healing. Biomaterials, 122, 34-47. Zhao, Y., Xu, T., Shen, C.-Y., Xu, G.-H., Chen, S.-X., Song, L.-Z., . . . Lv, W.-T. (2014). Identification of a retroelement from the resurrection plant Boea hygrometrica that confers osmotic and alkaline tolerance in Arabidopsis thaliana. PLoS One, 9(5), e98098. Zheng, J., Xiao, P., Le, X., Lu, W., Théato, P., Ma, C., . . . Chen, T. (2018). Mimosa inspired bilayer hydrogel actuator functioning in multi-environments. Journal of Materials Chemistry C. Zhong, R., Taylor, J. J., & Ye, Z.-H. (1997). Disruption of interfascicular fiber differentiation in an Arabidopsis mutant. The Plant Cell, 9(12), 2159-2170. Zhong, R., & Ye, Z. H. (2015). Secondary cell walls: biosynthesis, patterned deposition and transcriptional regulation. Plant Cell Physiol, 56(2), 195-214. doi:10.1093/pcp/pcu140 Zhou, D., Zhao, C.-Y., & Tian, Y. (2012). Review on thermal energy storage with phase change materials (PCMs) in building applications. Applied energy, 92, 593- 605. Zickler, G. A., Ruffoni, D., Dunlop, J. W., Elbaum, R., Weinkamer, R., Fratzl, P., & Antretter, T. (2012). Finite element modeling of the cyclic wetting mechanism in the active part of wheat awns. Biointerphases, 7(1-4), 42. doi:10.1007/s13758-012-0042-x Zolfagharian, A., Kouzani, A. Z., Khoo, S. Y., Moghadam, A. A. A., Gibson, I., & Kaynak, A. (2016). Evolution of 3D printed soft actuators. Sensors and Actuators A: Physical, 250, 258-272. Zolfagharian, A., Kouzani, A. Z., Khoo, S. Y., Nasri-Nasrabadi, B., & Kaynak, A. (2017). Development and analysis of a 3D printed hydrogel soft actuator. Sensors and Actuators A: Physical, 265, 94-101. Zonneveld, B., Leitch, I., & Bennett, M. (2005). First nuclear DNA amounts in more than 300 angiosperms. Annals of Botany, 96(2), 229-244. Zou, L., Jin, H., Lu, W.-Y., & Li, X. (2009). Nanoscale structural and mechanical characterization of the cell wall of bamboo fibers. Materials Science and Engineering: C, 29(4), 1375-1379.

223

APPENDICES

APPENDIX 1. HIERARCHIES OF STIFFNESS

Appendix 1 is a published review of stiffness in plants (Brulé et al., 2016).

Because this review primarily focuses on hierarchical stiffness in the model species

Arabidopsis thaliana, it was not used as a literature review for this thesis. However, the mechanical concepts discussed in this review are relevant, and Boxes 1 and 2 provide useful information on the mechanical properties (Box 1) and the frequently used experimental techniques (Box 2) related to investigations of plant mechanics.

224

ABSTRACT

Plants must meet mechanical as well as physiological and reproductive requirements for survival. Management of internal and external stresses is achieved through their unique hierarchical architecture. Stiffness is determined by a combination of morphological (geometrical) and compositional variables that vary across multiple length scales ranging from the whole plant to organ, tissue, cell and cell wall levels.

These parameters include, among others, organ diameter, tissue organization, cell size, density and turgor pressure, and the thickness and composition of cell walls. These structural parameters and their consequences on plant stiffness are reviewed in the context of work on stems of the genetic reference plant Arabidopsis thaliana

(Arabidopsis), and the suitability of Arabidopsis as a model system for consistent investigation of factors controlling plant stiffness is put forward. Moving beyond

Arabidopsis, the presence of morphological parameters causing stiffness gradients across length-scales leads to beneficial emergent properties such as increased load-bearing capacity and reversible actuation. Tailoring of plant stiffness for old and new purposes in agriculture and forestry can be achieved through bioengineering based on the knowledge of the morphological and compositional parameters of plant stiffness in combination with gene identification through the use of genetics.

225

INTRODUCTION

Physiological and mechanical requirements, as well as the physical environment, are among the most important factors that contribute to shaping plant organs and anatomy during growth, and hence the stiffness of plants and their organs. It is essential to first recall the physical and chemical laws to which a plant is subjected to understand the different morphological features that a plant develops throughout its body. Physiological functions dictated by growth, survival and reproduction, as well as mechanical demands, are subject to habitat conditions that determine plant morphology, anatomy and each of its constitutive tissues.

The primary functions a plant must perform include photosynthesis, fluid movement, reproduction, and mechanics that withstand the static and variable forces encountered during the plant’s life span. Thus, internal (supporting self) and external forces (e.g., gravity, wind) must be balanced with the metabolic needs of life (e.g., acquisition of sufficient sunlight, water and nutrients; prevention of water loss), postembryonic growth and environmental responses within the ecological context of a particular plant species. Their ability to do this is linked to their unique, hierarchical architecture, in which the morphological and compositional parameters governing stiffness are developed across multiple length scales from organ to tissue to cells to cell walls (Figure 5.1) [1-4]. To clarify our use of the term parameters (variables, properties) influencing stiffness, as well as to define various mechanical terms used throughout the review, we have provided further description in Box 1 and Figure 5.1. In addition,

226

techniques commonly used to mechanically test the stiffness parameters described

throughout the review have been outlined in Box 2 and Figure 5.2.

In this review, we will focus on mature plant organs (stems) and their structural

and mechanical properties, rather than on the biomechanics of cell and tissue initiation,

growth and development, as these recently have been extensively reviewed (e.g., [1, 5,

6]). We investigate parameters affecting plant stiffness from a top-down perspective,

considering the work that has been performed on plant biomechanics in the genetic

reference plant Arabidopsis thaliana (Arabidopsis), and address the question of whether

Arabidopsis could be used as an appropriate model system in which to elucidate the roles

of various structural parameters that affect plant stiffness. Having discussed a series of

controlling factors across multiple length scales, we next consider the emergent

functional properties observed in plant species where gradients in morphological

parameters are present. These properties are often found in plant species that are used as

reference plants for engineered, bio-inspired actuation prototypes. Finally, we briefly address the need for and basic techniques for the bioengineering of plant stiffness to

improve agricultural success and develop better functional products.

227

TOP-DOWN INVESTIGATION OF PLANT STEM STIFFNESS HIERARCHIES

IN THE REFERENCE PLANT ARABIDOPSIS THALIANA

Plant stiffness is a mechanical property whose governing parameters span across multiple levels. The interaction between these parameters at different length scales makes it difficult to distinguish in what proportion individual parameters contribute to overall plant stiffness. In this section, we examine the data available with regard to the hierarchical levels of plant structural organization and their biomechanical properties as studied in Arabidopsis, and consider the merit of using Arabidopsis as a system in which to investigate the geometrical and compositional parameters at different length scales to contribute to an integrated, multiscale model of plant stiffness.

Whole Plant

A discussion of plant biomechanics, in particular plant stiffness, cannot avoid examining plant structure as a whole. Key factors that control plant stiffness include the force of gravity upon the stem, the weight of branches and leaf canopy, and the flow- induced stresses upon the whole plant due to wind acting upon the canopy [7, 8]. The effect of these forces is particularly strong in large, long-lived plants such as trees, and the ability of plants to withstand these stresses is dependent on properties of the stem (see section below) as well as the distribution of weight and formation of overall leaf canopy shape by the pattern of branching [3, 8]. However, in a small herbaceous annual such as

Arabidopsis, these factors are less important, and have not been studied. For some background on stiffness at the whole plant level, we refer you to [8-12].

228

Organ Level

As primary load-bearing structures for the plant, stems and branches must be able to support both themselves and other organs (e.g., fruit, flowers and leaves) [3, 13].

Through the alteration of geometric variables such as stem shape, height, cross sectional diameter or a combination thereof, it is possible for plant growth to be tailored, resulting in the final morphology accommodating mechanical requirements, including stiffness, imposed by both static and dynamic environmental factors (e.g., gravity and wind, respectively) [3, 7, 13].

Environmental History

Variation in mechanical performance naturally exists among accessions of

Arabidopsis. An investigation of the tensile stiffness of 12 accessions of Arabidopsis identified correlations with stem diameter and cell wall composition [14]. Natural variation in stiffness most likely exists due to the different habitats in which the accessions have evolved. This includes factors such as nutrient availability, environmental conditions and mechanical perturbation, the latter of which directly impacts stiffness in Arabidopsis. Stems of plants grown under mechanical perturbation mimicking wind disturbance were 50% shorter than control plants (Figure 5.3A) and demonstrated alterations in tissue shape and proportions (Figure 5.3E), as well as in cell wall thickness. Further, mechanically perturbed stems were 70% less stiff than unperturbed stems (Table 5.1), demonstrating that external mechanical stimulation can result in adaptive growth that modulates the final stiffness properties of a plant [13].

229

Height Diameter and Cross-sectional Stem Shape Organ-level variables that modulate stem stiffness are interrelated and include stem diameter and stem shape as predetermined through adaptive growth and evolution.

Depending upon the species, both stem diameter and shape can in part affect stem height

[7, 9]. In larger species, for example trees, height becomes a more critical parameter that can affect stiffness due to self-weight from large branching canopies and increased tissue mass [4, 9]. Height has little impact on plant stiffness in Arabidopsis, however, since this species is quite small. Thus, at the organ level for Arabidopsis, stem diameter and cross- sectional shape are the main parameters to consider in relation to plant stiffness as they alter the ability of the stem to bend and hence its flexural rigidity.

The Arabidopsis dominant gain-of-function mutant STURDY has a qualitative increase in stiffness with 40% thicker stems than wild type plants and a concomitant increase in cell number and reinforced/lignified tissue (Table 5.1). The results seen with

STURDY mutants contrast with those observed from mechanical testing of dried stems of natural accessions of Arabidopsis, where a negative correlation was determined between tensile stiffness and diameter. Instead, diameter was correlated with length of vegetative growth: i.e., time to flowering [14]. These results cannot be directly compared - quantitative mechanical testing was not performed on STURDY stems. However, this contrast highlights the point that stem diameter cannot be separated from sub-organ level morphological variables such as turgor, cell number (tissue mass), cell type and cell size.

This is also exemplified in the transgenic Arabidopsis line MYB87-SRDX, which has thicker stems accompanied both by changes in cell number and cell size (Figure 5.3B and

Table 5.1) [15].

230

Stem shape has been studied for its contributing role to overall plant stiffness in multiple species [8, 16], but has not been mechanically tested yet in Arabidopsis. Several

Arabidopsis meristem mutants have been identified with fasciated (flattened, broader) stems (i.e., clavata1, fasciata1 and 2, meristem enlargement1) (Figure 5.3C and Table

5.1) [17, 18]. These mutants could be interesting to mechanically test to determine if stem shape contributes to overall plant stiffness in Arabidopsis.

Tissue Level

Tissues are not distinct entities that operate independently within a plant organ; their physical adhesion to one another facilitates biochemical crosstalk, among other processes, and influences the manner in which tissues respond both individually and as a unit to local and global mechanical stresses [3]. Therefore, the spatial arrangement, density and proportion of tissue types relative to each other within the plant stem must be tightly controlled, as modifications in these parameters have mechanical ramifications, including changes to stem stiffness.

Arabidopsis stems have a radially symmetric pattern typical of many dicots in which the vascular bundles are arranged in a collateral manner with interfascicular fibres forming in the areas between bundles. This creates a ring of vascular and vascular- supporting tissue that separates the central pith from the cortex and epidermal tissues

(Figure 5.3D-i/iii) [19]. The role of interfascicular fibres as a key load bearing tissue has been demonstrated by the ~80% decrease in tensile strength of the basal stem of interfascicular fiberless1/revoluta (ifl1/rev) mutants that lack interfascicular fibres compared to wild type stems (Table 5.1 and 5.2) [20]. Though not tested for ifl1 stems, it

231 would be expected that stiffness would also be decreased as seen for cellulose mutants that affect the thickness of interfascicular fibre cell walls (Tables 5.1 and 5.2).

Tissue Organization and Relative Tissue Proportions

While mutations in the IFL1/REV gene affect the presence of interfascicular fibre tissue, other mutants have been identified that affect the organization of vascular and other tissues in Arabidopsis stems. This includes, coincidentally, the dominant amphivasal vascular bundle1 (avb1) allele of REV/IFL1, whose mutants have amphivasal vascular bundles more typical of those that are found in monocots (i.e., xylem tissue completely surrounds the phloem tissue) (Figure 5.3D-ii, Table 5.1 and 5.2) [19].

Breaking force tests demonstrated a 60% decrease in avb1 stem strength compared with wild type stems. The reduction in cell wall thickness observed for avb1 interfascicular fibres (Figure 5.3I) suggests that this mechanical defect is most likely due to lack of stem reinforcement rather than altered distribution of vascular bundles as a load-bearing tissue

[19]. However, the potential contribution of tissue organization to stem stiffness in

Arabidopsis cannot be excluded until definitively proven to have no effect. This would be best studied in mutants in which stem reinforcement remained constant, but tissue organization was altered.

Other tissue organization mutants include high cambial activity (hca) 1 and hca2 that lack (or have very few) interfascicular fibres and, instead, have a continuous ring of vascular tissue (Figure 5.3D-iv and Table 5.1) [21, 22]. These mutants are also affected in the relative proportions of vascular versus other tissues, e.g., hca1 mutants have an increase in vascular versus pith tissue, and a reduction in cell size that correlates with

232

decreased stem diameter despite the overproliferation of vascular tissue (Figure 5.3D-iv)

[21]. While neither of these stem organization mutants has been mechanically tested,

changes in the proportion of stem tissues has been linked to the mechanical properties of

Arabidopsis. The 70% reduction in stem stiffness observed in the mechanical

perturbation experiments mentioned previously was accompanied by reductions in the

proportion of pith and interfascicular fibres tissue, while the area of cortex tissue

increased. While thinner interfascicular fibre cell walls were also observed in the

perturbed plants, it was suggested that altered tissue proportions were responsible for

~20% of the decrease in stem rigidity [13]. Obviously, it would be of interest to compare

the effects on plant stiffness of various tissue organization mutants with those resulting

from mechanical perturbation.

Tissue Density

Tissue density is affected by cell number, size, degree of cell-to-cell adhesion, cell wall thickness (i.e., thicker walls can increase density), and cell wall composition, making it difficult to discern density’s direct contribution to plant stiffness. However, tissue density is predicted to alter stiffness in stems, and its contribution can be approximated if variables such as wall composition are assumed to remain unchanged.

For example, using the assumption that cell wall composition did not change, the reduced density of interfascicular fibre tissue of mechanically wind-perturbed Arabidopsis plants resulting from thinner cell walls was calculated to decrease bending stiffness by ~60% in perturbed plant stems (Figure 5.3E) [13]. The increased cell density seen in the thinner

233

stems of the hca1 mutant noted above also makes it an attractive candidate for future

mechanical investigation [21].

Cell Level

The base unit of plant structure is the cell; usually a living protoplast performing metabolic functions surrounded by the cell wall, an extracellular matrix that forms the boundary of the cell and mediates adhesion and interaction with other cells, as well as

with other biotic and abiotic factors (exceptions would include, for example, woody

tissues where cells have lost their protoplast following secondary cell wall reinforcement)

[23, 24]. As mechanical entities in their own rights, plant cells have balanced external

forces (typically tensile and compressive) that result from interaction with other cells

with internal forces such as turgor to maintain plant structure in response to forces

exerted on the organism [24]. This balance is mediated by modulating cell shape and size,

as well as cell wall production and cell-to-cell adhesion, allowing cells to be sufficiently

stiff that they withstand mechanical stress while simultaneously permitting growth [23,

25]. In this section, we will focus on cell geometry and cell-to-cell adhesion. Properties

of plant cell walls beyond those relating cell-to-cell adhesion will be discussed in detail in

the section of the same name.

Cell Geometry

Plant tissues are cohesive units formed from specialized cell types whose

individual physical attributes combine and impart unique properties, including stiffness,

that dictate the manner in which a given tissue type responds to mechanical stress [26].

234

Pith, vascular, cortex, and interfascicular stem tissues are composed of cells with

different shapes that modify the mechanical properties of the stem, including stiffness, in

a species-dependent manner [2, 24]. Different conformational shapes along the transverse

and longitudinal axes of the plant cell can change its bending stiffness [4, 27]. Shape also

affects how cells fit together both within a single tissue and between different tissue

types. Cells that share large contact surfaces are packed tightly together with little

airspace between them, allowing them to form a cohesive unit that can better distribute

mechanical stress. For cells with the same elasticity, this would generate stiffer tissues

than those in which cells are more loosely packed [2, 28].

In Arabidopsis stems most tissues are typically made up of the same polyhedral- shaped cells when observed in cross-section. Thus, there is little to no change in shape in the transverse plane among the different tissues. However, several biochemically and structurally perturbed Arabidopsis mutants have a dwarfed plant phenotype with shortening in the length of load bearing and non-load bearing cell types along the longitudinal axis of the plant [13, 19, 21, 29]. These mutants could be used to help elucidate the contribution of cell shape to stiffness if other parameters (e.g., changes in cell wall thickness) that may also change are taken into consideration.

The other facet of cell geometry is size, which takes into account the volume of a cell. Changes in cell size alter the amount of surface area available for contact with adjoining cells and also changes the total volume through which a mechanical force can be dissipated in individual cells [28]. If other variables (e.g., cell number) remain constant, increases or decreases in cell size can have consequences on both tissue size and overall stem diameter, thereby altering tissue and organ stiffness. To date there is no

235 literature that directly investigates the contribution of cell size in relation to cell mechanical stiffness in Arabidopsis.

Cell-cell Adhesion

Plant cells generally adhere to one another by the middle lamella that creates a physical connection between neighbouring cells through which biochemical and mechanical crosstalk can occur [30]. The middle lamella is initially composed primarily of pectins, but in tissues having secondary cell wall deposition may become lignified [31,

32]. The level of adhesion between cells can be altered by modifying the biochemical composition of the middle lamella, as seen in various Arabidopsis mutants [30]. Cell adhesion is dependent upon other properties that affect stiffness (e.g., cell shape and cell wall composition). The role of adhesion as a mechanical property has not yet been specifically tested in Arabidopsis. It may be possible to approximate the contribution of cell adhesion to an observed value of stiffness by calculating how much other variables contribute to total stiffness, as has been performed with tissue density. For example, mutants in both the QUASIMODO1 (QUA1) gene and the QUA2 gene of Arabidopsis have less pectic homogalacturonan in the primary cell wall and are defective in cell adhesion (Figure 5.3F) [33, 34]. While qua1 plants have not yet been tested, hypocotyls

(embryonic stem) mutant in the qua2 gene have been mechanically tested and demonstrate reduced tensile stiffness. It is likely that cell separation may be partly responsible for the observed reduction in stiffness [35].

236

Turgor Pressure

Turgor pressure, the hydrostatic force imposed by the plant cell protoplast upon the cell wall, has been extensively studied and reviewed with regard to its significant mechanical role in driving cell expansion and imparting stiffness to cells, tissues and plant organs that lack secondary cell walls (Figure 5.3G) [5, 36, 37]. Turgor varies among plant organs, tissues, cells and the developmental stages of the plant, as well as with the water status of the plant [38]. As a non-directional force, turgor acts upon regions of differing mechanical strength and stiffness within plant cell walls, establishing cell shape, size and direction of expansion [37, 38].

Although turgor pressure is responsible for much of the stiffness observed in plant cells that lack a secondary cell wall, its total contribution has been challenging to quantify, as previous efforts lacked the resolving power to distinguish between the input of cell wall properties and turgor on measured stiffness values in living cells [36-38].

Plasmolysis prior to testing via treatment with solutions of high osmolarity eliminates the stiffening effects of turgor pressure. Comparison of mechanical tests of Arabidopsis wild type and cell wall mutant hypocotyls revealed reduced stiffness and increased differential of mechanical properties after plasmolysis [29]. However, potential alterations to cell wall mechanical properties due to the plasmolysis treatment makes it preferable to do comparisons with hydrated tissues [39]. Emerging experimental methods and novel approaches using modified micro and nanoindentation methods such as atomic and cellular force microscopy combined with modelling have helped to separate the effects of cell wall composition from turgor pressure and their impact on cell stiffness (reviewed in

[24, 36, 38]).

237

Cell Wall Level

As mechanical entities, living plant cells must balance external forces with the internal osmotic force of turgor pressure. The ability to withstand such forces is largely conferred by the cell wall. Similarly, the mechanical properties of whole plants have a significant origin in the presence and specific compositional make-up of both primary and secondary cell walls [6, 39].

Primary Cell Walls

Young and growing primary cell walls require both strength to maintain the structural integrity of their cells and compliance to allow for changes in cell size and shape. Many cell types undergo secondary cell wall thickening once cell expansion ceases is completed. This provides structural support (further stiffness and strength) to the cell and the organism (e.g., vascular tissue and interfascicular fibres). In both cases, walls are comprised of stiff, semi-crystalline microfibrils of the β-1,4-D-glucose polymer cellulose that wrap around the cell like cables and interact with a matrix of other polysaccharides (hemicelluloses and pectins) and a small amount of proteins [39, 40].

This structure has led to the simplistic comparison of plant cell walls to fibre-reinforced composites (e.g., [6, 39]). In the primary cell walls of many dicots, such as Arabidopsis, the most significant hemicellulose is xyloglucan, which shares the β-1,4-D-glucose backbone of cellulose, but is substituted with short side chains containing xylose, galactose and fucose. The type and degree of interaction between xyloglucan and cellulose is currently debated, as is the degree of its role as a load-bearing polymer in the primary cell wall [37, 41]. Major pectins include homogalacturonan, xylogalacturonan,

238

and rhamnogalacturonan II, as well as rhamnogalacturonan I. The first three polymers

share a backbone of α-1,4-D-galacturonic acid residues that differ in their degree of

substitution: homogalacturonan lacks side chains, xylogalacturonan is substituted with

xylose and rhamnogalacturonan II has a set of four complex, evolutionarily-conserved

side chains. Rhamnogalacturonan I has a backbone of repeating units of α-1,4-D-

galacturonic acid - α-1,2-L-rhamnose that can be substituted on the rhamnose residues with arabinan, galactan and arabinogalactan side chains [42]. The structural properties of pectins within the wall, specifically gel stiffness and porosity, result from the proportion and degree of branching, as well as the amount of interaction between pectin molecules through the formation of linkages such as boron diesters between rhamnogalacturonan II side chains and calcium bridges between the free galacturonic acid residues of homogalacturonan molecules. The latter is modulated through the in-wall activity of pectin methylesterases, as homogalacturonan is thought to be synthesized and deposited into the wall in a relatively neutral, highly-esterified state [6, 42, 43]. As pectin is hydrophilic in nature, its structure is also correlated with the hydration state of the wall.

For example, homogalacturonan demethylation (leading to the presence of free acid groups) and the presence of certain rhamnogalacturonan I side chains such as arabinans have been correlated with increased cell wall hydration and decreased stiffness (reviewed in [6]; [44]). The porosity of the wall regulated by pectin structure also directly affects the water-holding capacity of cell walls and the ability of water to move within the wall, both of which affect biomechanical responses (reviewed in [6]; [43]). It is unclear whether the different types of pectins represent individual polymers, linear structural domains or branches of larger heterogeneous polymers. Pectins can also be complexed

239

with structural cell wall proteins such as the recently described ARABINOXYLAN

PECTIN ARABINOGALACTAN PROTEIN1 (APAP1) of Arabidopsis [37, 42]. An

increasing number of studies, including ones using solid-state nuclear magnetic

resonance [45, 46], have demonstrated the significant degree of physical interaction

between pectins and cellulose (reviewed in [6, 47]).

Arabidopsis is increasingly employed as an easily manipulated genetic model system to determine and dissect details of cell wall polymer synthesis and deposition. It is also being used to elucidate the interactions and mechanical roles of the various components during different aspects of plant development. This has included the selection of mutants affecting polymer synthesis, growth in the presence of chemical modulators, and bioengineering of plants with altered polymer synthesis and enzymatic modifications. With respect to the mechanics of the primary cell wall, research largely has focused on uniaxial testing of elongated hypocotyls of dark-grown seedlings. This early seedling stem-like tissue is preferred due to its cylindrical shape and the simplicity of its anatomical structure (Tables 5.1 and 5.2) [29, 48]. Since hypocotyls can dry out rapidly, testing has been performed with samples bathed in liquid or water vapour. While this ensures consistency between samples with respect to having undamaged, hydrated primary cell walls, it means that absolute values obtained include the stiffening effect of turgor pressure [39]. In the discussion below and in Table 5.2, the focus is comparative, based on the relative mechanical properties for cell wall and structural mutants tested under conditions that allowed for reasonable comparison. For a more comprehensive list of Arabidopsis mutants that have had mechanical investigations and their basic results, see Table 5.1.

240

Hypocotyl tensile tests have demonstrated that overall reductions in cellulose, xyloglucan and pectins all lead to decreases in both stiffness and strength (cellulose synthase inhibitor [49] herbicide 2,6-dichlorobenzonitrile-treated seedlings, cesa6/prc,

Atktn/fra2/bot, xxt1 xxt2 and qua2 mutants [29, 35, 50, 51]; Tables 5.1 and 5.2). From the mutants tested in a comparable manner (Table 5.2; [29, 39]), it is unsurprising to see that a significant reduction in cellulose content (~40%) due to inhibitor treatment led to the most drastic decreases in hypocotyl stiffness and strength [29]. Further, it appears that pectins, both in terms of quantity of homogalacturonan (qua2) and ability to form rhamnogalacturonan II cross-links (mur1), have a moderate effect on stiffness and limited effect on strength [29, 35]. The effects of altering xyloglucan side chains are more complex and can be greater than the effects of diminished pectins, with the loss of galactose substitution leading to the most severe reduction in stiffness and strength of any of the matrix polymer mutants, including one completely lacking xyloglucan. This result suggests a significant mechanical role for the galactose side chains of xyloglucan.

However, the limited effect of the complete loss of xyloglucan on plant growth and mechanical properties was surprising and has led to debate on the exact role of xyloglucan as a load-bearing polymer within primary cell walls. This result is further complicated by changes in proportions and organization of other cell wall components that appear to allow their assumption of a greater load-bearing role within the wall in these mutants [23, 50, 51]. Side chains of pectins also can affect primary cell wall mechanical properties. Stems of arabinan deficient mutants (arad 1 arad2) were stronger under compression and had decreased compliance in indentation tests, consistent with predictions that arabinans act as cell wall plasticizers [6, 52]. The degree of

241 homogalacturonan methylesterification modulates both hypocotyl and meristem cell wall stiffness, as seen with atomic force microscopy analysis of surface properties. Plants overexpressing PECTIN METHYLESTERASE5 have decreased homogalacturonan esterification and reduced cell wall stiffness, while those overexpressing the PECTIN

METHYLESTERASE INHIBITOR3 have increased homogalacturonan esterification and increased stiffness. There is a correlation between the degree of homogalacturonan methylesterification and consequent stiffness in both hypocotyl and meristematic tissues.

In both cases, these changes are under developmental regulation [53, 54]. The effect of the degree of methylesterification on pectin gel stiffness is complicated by the pattern of esterification, such that pectin methylesterase activity can increase or decrease the degree of calcium-crosslinking of homogalacturonan molecules [6, 43].

Secondary Cell Walls

Plant secondary cell walls are cellulose-reinforced primarily polysaccharide-based composites, similar to primary cell walls. However, secondary cell walls are stiffer at least in part due to their higher cellulose content, longer cellulose chains, increased diameter and/or bundling of microfibrils and a greater proportion of crystalline (versus amorphous) cellulose [55]. Secondary wall material is deposited after growth interior to the primary cell wall. In dicots and gymnosperms, secondary cell walls are generally observed to have three layers specified as S1, S2 and S3 moving from earlier to later stages of secondary cell wall production. Interestingly, these layers can differ significantly in thickness, with the S2 layer generally being substantially the thickest and thus responsible for the majority of secondary cell wall mechanical properties [2, 39, 56].

242

Secondary cell walls also differ from primary cell walls in having a parallel arrangement

of microfibrils that can be described by their angle relative to the longitudinal axis of the

cell, which is different among the three layers. In fibre cells such as those of stems, the

microfibril angle is correlated with cell mechanics such that cells with a lower angle

(more longitudinal to the cell axis) are more stiff in axial loading, while those with a

higher angle (more transverse to the cell axis) are less stiff [39, 57]). Secondary cell walls

also differ from primary cell walls in their matrix composition: there is little pectin or

protein, different hemicelluloses predominate and walls are impregnated to varying

degrees (depending upon the species) with the polyphenolic compound lignin. In many

dicots, including Arabidopsis, the main secondary cell wall hemicelluloses are xylans,

which have a β-1,4-D-xylose backbone that tends to be decorated with glucuronic acid

(unmodified or methylated) and smaller quantities of arabinose. Lesser quantities of glucomannans are also found in dicotylenous secondary cell walls, polymers with a backbone of both β-1,4-D-mannose and β-1,4-D-glucose [31, 56]. As seen in pectins, the

backbone residues of xylans and glucomannans can be acetylated [31, 43, 56]. After the

production of the polysaccharide secondary cell walls, lignification can occur through the

deposition and polymerization of monolignols within the wall. For example, dicot

secondary cell walls are rich in guaiacyl (G) and syringyl (S) lignin, formed from

coniferyl alcohol and sinapyl alcohol, respectively [31, 32, 56]. While the monomers

present are consistent in dicots, the proportions and degree of lignification can vary

among species. Lignin acts to strengthen cell walls, and, as a hydrophobic polymer, leads

to the exclusion of water ([31, 32, 56]).

243

A number of genes involved in secondary cell wall polymer synthesis have been

identified in Arabidopsis by screening for plants with thinner-walled, collapsed xylem

cells (irregular xylem [irx] mutants) and reduced stem strength (fragile fiber [fra] mutants), gene co-expression with previously identified secondary cell wall cellulose synthase genes (added to irx series) and protein homology to other secondary cell wall biosynthetic enzymes (e.g., IRX10-like) [58-61]. In most cases, secondary cell wall mutants have thinner or unevenly-deposited secondary cell walls in both xylem and interfascicular fibres. This correlates with their reduced polysaccharide content [31, 62], and weaker/less stiff stems under external testing conditions (Tables 5.1 and 5.2). While testing has not been done on isolated fibres or xylem cells from Arabidopsis due to their small dimensions, the collapsing of xylem cells seen in these mutants is thought to reflect the loss of strength in these cell walls that makes them unable to withstand the normal compressive forces from the negative pressure exerted by transpiration [58, 63]. The organ-level phenotype of these mutants is shortened plant stature, probably resulting from inefficient water transport through these collapsed cells (e.g., [31, 58, 62]). For a comprehensive overview of the irx mutants discussing the complexity of their roles and phenotypes, as well as xylan and lignin synthesis in general, see [31]. As with primary cell walls, the discussion below is comparative. See Table 5.2 for a gross comparison of results that were obtained in a similar manner, and Table 5.1 for a full listing of secondary cell wall experiments.

Differing levels of cellulose (cesa8/irx1/fra6, korrigan/irx2, cesa7/irx3/fra5) and lignin (irx4/ccr1, plus transgenic plants with a range of lignin production) were positively correlated with both bending stiffness and strength in three-point bending tests on stems

244 of four of the original irx mutants (Figure 5.3H and Table 5.2). Comparison between cellulose and lignin results suggests that the amount of lignin present in the cell wall has a greater impact on stem bending strength than does the cellulose content of the wall. irx3

(cesa7/irx3/fra5) mutants that have an 82% reduction in cellulose content (compared to wild type) have stronger stems than irx4 (irx4/ccr1) mutants with a 50% reduction in lignin [58, 63]. The majority of available mechanical testing results for secondary cell wall mutants consist only of uniaxial tension breaking-strength tests. However, since these were performed by the same research group using the same instrument, it is broadly possible to compare the results across a number of mutants affecting cellulose and xylan synthesis (Table 5.2). Unsurprisingly, a similar positive correlation between cellulose content and strength is seen with these tests as for three-point bending, however, the severity of the strength decreases appear to be greater. For example, stems of the fra5 allele (65% reduction in cellulose) of CESA7/IRX3/FRA5 have 10% of wild type tensile strength, while the irx3 allele (82% reduction in cellulose) has only a 50% decrease in bending strength, presumably due to the fact that bending involves both tensile and compressive forces [58, 64]). The breaking force strength of the loss-of-function ifl1 allele of IFL1/REV that lacks interfascicular fibres is similar to that of fra5 mutants that have a 65% reduction in stem cellulose. This hints at the considerable amount of cellulose found in interfascicular fibre secondary cell walls as a proportion of the total found in stems [20, 64]. While it has been possible to study changes in cellulose quantity versus mechanical properties, no mutants affecting only microfibril angle have yet been identified, and it seems like it may be difficult to separate cellulose, and perhaps other polysaccharide synthesis, from microfibril angle [50, 65, 66].

245

A large number of genes affecting different aspects of xylan synthesis have been tested for breaking force (Tables 5.1 and 5.2). Disruption of genes involved in xylan backbone synthesis or the production of the xylan reducing end tetrasacchride (irx9, fra8/irx7, irx8/gaut12 and parvus) led to plants containing 25-50% of wild type xylan levels, all of which have very severe decreases in tensile strength (~85% reduction).

While it appears that changes in xylan have a similar effect on strength as the loss of cellulose, this is unlikely, as lesser reductions in both cellulose and lignin in addition to xylan have been proposed for at least the irx9, fra8/irx7 and irx8/gaut12 mutants, suggesting that the decrease in xylan is only responsible for part of what is seen [67-69].

The xylan backbone can be O-acetylated in a number of positions, and recently a number of genes from the REDUCED WALL ACETYLATION (RWA) and TRICHOME

BIREFRINGENCE LIKE (TBL) gene families have been identified for their overlapping roles in that process [62, 70-73]. Comparison among mutants that specifically affect xylan acetylation (eskimo1(esk1)/tbl29 single, esk1/tbl29 tbl3 tbl31 triple and rwa1 rwa2 rwa3 rwa4 quadruple mutants) demonstrates a correlation between reduced stem strength and loss of acetylation. However, the reductions in strength do not necessarily track with the absolute level of acetylation, reflecting the contribution of differences in the pattern of acetylation (i.e., position on the xylose residue and presence of glucuronic acid substitution on the same monosaccharide) in these different mutants [62, 70-73]. While the loss of strength seen with a significant decrease in acetylation (56% of wild type acetylation in esk1/tbl29 tbl3 tbl31 triple mutants) is not as severe as that seen for loss of cellulose or xylan noted above (4-fold versus 5-fold, respectively), it is still quite significant [64, 67, 68, 73]. This demonstrates the important mechanical role of xylan

246 acetylation in the secondary cell walls. Acetylation increases the hydrophobicity of xylan and has been suggested to increase interaction between xylan, cellulose and lignin [62,

72-74]. Unlike acetylation, the presence of glucuronic acid side chains seems to have a minor effect on xylan interactions in secondary cell wall, as glucuronic acid substitution of xylan1 (gux1) gux2 double mutants have only a small decrease in stem strength as assessed via four-point bending (Table 5.1) [75]. However, since even strongly cellulose deficient mutants demonstrate much less significant changes in strength in bending tests than tensile tests (Table 5.2), it is hard to compare this result to those of other xylan mutants. Reductions of xylan quantity and/or acetylation probably lower stiffness as well as strength as seen for cellulose. Still, this needs to be confirmed and quantified.

Glucomannans are also found in Arabidopsis stems, and triple mutants defective for the backbone synthases CSLA2 CSLA3 and CSLA9 that lack detectable mannans had no change in stem stiffness or strength in four-point bending tests (Table 5.1) [76]. This suggests that, unlike xylans, mannans do not have a significant role in the mechanical properties of Arabidopsis stems.

247

CAN ARABIDOPSIS BE USED AS A MODEL TO UNDERSTAND THE ROLE

OF PARAMETERS THAT GOVERN PLANT STIFFNESS?

In the sections above, we have outlined examples of structural and mechanical properties of the genetic reference plant Arabidopsis with regard to parameters that affect plant stiffness across multiple length scales (organ, tissue, cell and cell wall). Now we return to the question that we posed at the beginning of this review paper: can

Arabidopsis be used as a model to understand the role of parameters that govern plant stiffness? In our opinion, the answer is yes, with some obvious caveats that will be discussed below.

First, we would like to highlight the need for such a model system in which consistent and systematic investigations of the different morphological and compositional parameters affecting plant stiffness and biomechanics can be performed. While a significant amount of work has been done and much learned about the factors that modulate plant biomechanics, there is a great deal of heterogeneity to the research that has been completed in this field. This heterogeneity ranges from the organs, tissues and cells investigated (from isolated wood fibres, to stems, to growing pollen tubes) to the species employed (from the herbaceous dicot Arabidopsis, to gymnosperm trees, to monocotyledonous grasses) to the techniques employed for mechanical testing (uniaxial tensile, to three-point bending, to nanoindentation) (for general reviews on plant mechanics see [2, 4, 39, 40]). Background structural differences, including organ size, tissue organization, cell number, cell type and cell wall composition, render it difficult to compare these studies and to develop a correct and cohesive model of the interaction of

248 controlling factors across length scales that give rise to particular plant stiffness properties. It also makes it difficult to discern how they could be modulated for human benefit.

With this need for a model system in which to investigate structural parameters contributing to plant stiffness, we believe that Arabidopsis is a useful candidate because it is a very well studied and easy to manipulate plant system (reviewed in [77] and references therein). In particular, its widespread use as the predominant plant genetic and genomic model for all aspects of plant growth and development has led to the continuing identification of mutants and genes modulating key factors expected to affect stiffness such as stem size, tissue organization and cell wall biosynthesis, as well as the ability to manipulate these factors through genetic engineering (Table 5.1).

Obviously, not all aspects of plant stiffness and biomechanics can be studied in

Arabidopsis, and information is needed on other species for economic purposes, among others. To that end, Arabidopsis, though a promising system, cannot be the sole model or the only species studied. Arabidopsis is a small, herbaceous, annual dicotyledon in which the hierarchies of plant stiffness are somewhat simplified. For example, bulk features of wood or whole plant issues faced by large species and trees such as the mechanical forces imposed by large size and leaf canopies cannot be addressed in Arabidopsis, though properties of vascular tissues, fibre cells and secondary cell walls that make up wood can be studied [78, 79]. Arabidopsis is neither a monocot nor a gymnosperm, so specific aspects of tissue organization and cell wall composition that are unique to these plant groups cannot be analyzed. However, once again, some aspects of these groups potentially can be addressed in Arabidopsis: monocot-type tissue organization can be

249 mimicked through mutants (e.g., avb1 – Section 2.3.1.) [19] and, while the specifics of interactions between different cell wall polymers do need to be addressed in their primary systems, it has been suggested that the matrix polymers that vary between plant groups mainly have the same physicochemical role to form a gel-like component in the cell wall and/or to reduce aggregation of cellulose microfibrils [50, 80, 81].

Caveats must also be raised for the use of genetic dissection as a tool to analyze the contribution of specific parameters. Changes to individual gene function can lead to complex and unpredictable consequences on the biology of organisms, making it difficult and perhaps unlikely to be able to target changes to a particular component or process by making selective changes to particular genes. This is due to a number of causes, including redundancy in gene function, the multiple functions of certain gene products (pleiotropy), and organismal stress-sensing systems that lead to compensatory changes in plant structure and physiological responses at multiple levels [82, 83]. Examples of unexpected consequences of genetic manipulation that can have biomechanical ramifications, and thus complicate studies, include: (1) Multiple tissue and/or cell level defects in mutants identified for their structural changes at the tissue level. For example, avb1 mutants have altered organization of their vascular bundles, but their interfascicular fibres also have thinner secondary cell walls [19], while hca1 mutants not only lack interfascicular fibres, but also have changes in tissue proportion, cell size and stem diameter [21]. (2)

Compensatory changes in composition and cell wall polymer organization in cell wall biosynthesis mutants. For example, while xxt1 xxt2 mutants lacking detectable xyloglucan have a subtle increase in other matrix polymers that allow greater load- bearing in the wall, it has recently been shown that there also is a decreased amount of

250 cellulose that is more aggregated and aligned than found in wild type cell walls (Figure

5.3J) [50, 51]. Many of the irx mutants affecting secondary cell wall production have effects on multiple polymers, pattern of secondary cell wall deposition and plant height

(reviewed in [31]). (3) Stress-sensing systems (e.g., the cell wall integrity signalling pathway) can be activated leading to unpredictable physiological responses including hormone responses, activation of plant defence mechanisms and ectopic deposition of cell wall material (reviewed in [84-86]). Such changes have been observed for mutants of multiple cellulose synthase subunits, among others (reviewed in [84]). This caveat and multiple examples highlight the necessity to ensure that lines chosen for study are well- characterized biologically at multiple levels and across all tissues to be studied so that all morphological and compositional parameters with biomechanical consequences can be accounted for in the production of future models. It may also suggest that a search for well-characterized mutants with structural changes of interest may be more practical than trying to target particular genes or pathways.

While we raise these issues related to genetic dissection, it must be pointed out that similar caveats of the complexity of differences exist when comparing samples from widely differing species and tissues. Considerations include scale, organ geometry, tissue organization, cell structure and cell wall composition. Physical dissection and chemical treatments (e.g., to remove certain cell wall components) also need to be considered carefully. With these in mind, an advantage of the use of Arabidopsis as a model system is that one is still working in the same species, with tissues and organs of largely the same structure, organization and composition. That said, in all cases, it is important to perform a detailed characterization of the system from tissue organization through to cell wall

251 composition and organization prior to mechanical testing and interpretation. Only then can all morphological and compositional parameters be incorporated into models. For an interesting discussion of the many requirements for analysis required to put together a comprehensive and multiscale model of plant cell walls alone, see reference [87].

252

FUNCTIONAL GRADIENTS IN PLANT STIFFNESS

While Arabidopsis is an useful system in which to generate data to add to an integrated model of plant stiffness parameters, as a small, short life-spanned, dicotyledonous, herbaceous annual, it cannot address all levels of plant stiffness or all structural or compositional peculiarities of different types of plants (e.g., trees, grasses, gymnosperms). In addition to the above subsets of basic factors that cannot be studied directly using Arabidopsis are functional gradients of plant structural properties across length-scales. These functional gradients enhance plant mechanical performance and are a source of inspiration for engineers (e.g., architecture, smart synthetic materials). These gradients will be discussed below along with their further functionalization in certain species that allow emergent properties such as hydration-based actuation.

Stiffness Gradients and Their Effects on Load-Bearing Capacity

The hierarchical architecture of plants is optimized in a manner that tackles environmental challenges and realizes complex functions. Spatial gradients have been evolutionarily incorporated into compositional and structural features of plant organs, allowing notably lower stress concentrations and greater adhesion to be achieved at the interfaces of different tissue types. This evolutionary adaptation strategy is well suited for resisting severe mechanical forces and accommodating larger deformations.

From a structural point of view, the stems of monocotyledonous plants can be regarded as fibre-reinforced composites where stiff vascular bundles with fibrous caps are embedded in a soft parenchymatous tissue [88]. The discontinuous transitions in

253 stiffness at the interfaces between these tissue types generate a highly non-uniform distribution of stresses, which can cause critical shear stresses at cell ends, and eventually lead to stem breaking. In stems of the Mexican fan palm Washingtonia robusta (Figure

5.4A), a gradual transition in stiffness from the central part of fibre caps to the surrounding parenchymatous tissue has been hypothesized to be an evolutionary adaptation strategy to mechanical forces. These changes in stiffness are correlated with a gradient in the degree of lignification in the cell walls of the fibre caps. This gradient of cell wall composition is predicted to alter both shear modulus and shear strength of the cell wall matrix, and eventually the axial tensile stiffness of the fibres [89]. The profiles of stiffness in the individual tissue types in the culms of the giant reed Arundo donax

(Figure 5.4B), indicate that gradual changes in stiffness exist within parenchymatous and sclerenchymatous tissues that are attributed to alterations at the tissue level. The absorbance pattern of the ultra-violet microspectrophotometry scans on toluidine blue cross-sections (Figure 5.4B i) and the intensity of stained lignin (Figure 5.4B ii) revealed that the fibre rings enclosing the vascular bundles had lignification gradually decreasing from the outer side towards the pith parenchyma. Thus, a gradual transition between the stiffness of the sclerenchymatous fibres and the parenchyma is expected; however, the profile of microfibril angle exhibits an abrupt change and thus it does not contribute to the stiffness gradient at the interface [88]. In Moso bamboo, Phyllostachys pubescens

(Figure 5.4C), a stiffness gradient across the fibre cap is developed by differential cell wall thickening which affects tissue density and thereby axial tissue stiffness in the different regions of the cap [90]. Radial gradients in tissue parameters (tissue density, vascular bundle volume and solid fractions) of bamboo give rise to an increase in the

254 axial stiffness, module of rupture and axial compressive strength from inside toward outside of the tissue [91, 92].

Plant Stiffness Gradients and Plant Actuation

The juxtaposition of cells having walls of differing hydration capacities allows for the creation of plant actuators. Those that act as bilayers whereby differential swelling

(extent or direction) of neighbouring tissues with dissimilar microfibril angle and/or degree of matrix swelling (type of matrix present and degree of lignification) leads to reversible shape changes are of particular interest. While differential swelling is the main mechanism commonly used to describe moisture-responsive actuation in plants, stiffness and compliance also come into play. In mechanical terms, even in the simplest case of a bilayer that is extensively used for analysis of plant movement, the net bending of the structure is a function of both the swelling ratio and the stiffness ratio of two constituents.

Heavily reinforced tissues are less able to bend, while those that swell are usually less stiff. Below we focus on an example from our own laboratories and discuss how stiffness gradients shape the stem deformation patterns we have observed in the species

Selaginella lepidophylla.

The spirally arranged stems of S. lepidophylla (Figure 5.4D i) compactly curl into a nest-ball shape upon dehydration (Figure 5.4D ii), limiting the photo-inhibitory and thermal damages the plant might experience in arid environments [93]. There is a dichotomy between the curling patterns of the stems located in the center of the plant and those at the exterior of the spiral phyllotaxy of the plant. Upon dehydration, the outer dead stems act as classical bilayers [94, 95] and bend into arcs after a relatively short

255

period of desiccation, whereas the axially graded inner green stems curl slowly into

spirals due to a hydro-actuated strain gradient along their length (Figure 5.4D iii). The

reversible curling/uncurling of stems is attributed to both compositional and structural

gradients giving rise to functionally graded stiffness within the transverse and

longitudinal sections of the stems (Figure 5.4D iv-vii) [96]. In inner stems, secondary cell

wall lignification occurs throughout the whole cortex in the basal sections, whereas the

middle of the stem has lignified tissues only in the abaxial cortex, and in a narrow abaxial

strip at the apical tip (Figure 5.4D iv) [96]. There is an abundance of xylan mostly in the

adaxial cortex of the stem (Figure 5.4D v), giving rise to a second compositional gradient

along the length of the stem, as elucidated using antibodies (Chapter 3). The

morphology of cortical cells also exhibits a smooth gradient in tissue density, which is

reflected in cell size and cell wall thickness (Figure 5.4D vi). Finally, the orientation of

cortical cells varies gradually from axially oriented cells in abaxial side of the stem to

almost 45o inclination in adaxial side (Figure 5.4D vii) (Chapter 3). The presence of multiple gradients in the coiling inner stems of S. lepidophylla suggests that it is a useful model for studying the contribution of structural and compositional gradients to plant stiffness, as well as to investigate how functional gradients predetermine the direction and magnitude of deformation observed in this species.

256

HARNESSING OF PLANT STIFFNESS: BIOMECHANICAL

TAILORING OF PLANTS

Agricultural and forestry engineering has historically focused on product quality,

yield, and resistance to biotic and abiotic challenges (e.g., pests, pathogens, soil quality,

drought). More recently, attention has been paid to the development of new products and

usages for current crops and crop residues, such as fibres for functional materials and

optimized degradation for use in biofuels. Explicit discussion of the manipulation of plant

stiffness is not necessarily present, but there are many new and old agricultural and

product needs that require or would benefit from modulation of plant stiffness properties.

As much recently has been published regarding the details of plant cell wall and tissue

structure optimization for biofuel and forestry purposes [97, 98], we will restrict our

discussion to examples of these needs and basic strategies for their manipulation, and

refer readers to the following reviews for the details of bioengineering [98, 99].

Agricultural issues related to plant stiffness include growth fitness (e.g., lodging)

and food quality (e.g., texture, shape maintenance). Stem and root lodging cause

irreversible damage leading to breakage in response wind and rain forces. Lodging is a

global problem that causes significant yearly food production losses [7]. While conventional methods of lodging resistance have included creating dwarfed crop species that are small and subsequently less prone to bending and breakage, not all crop species are amenable to dwarfing [100]. Here knowledge of stiffness parameters could be useful to breed crop species (dwarfed or not) with stronger stems to improve both lodging resistance and crop yield [7, 24, 101].

257

On the crop quality side, texture is critical: crops must obtain a desired state of

firmness, which can vary upon the end use (e.g., food crops used raw or cooked), as well

as the need to store and/or transport them without damage [102, 103]. An important area

of study thus includes fruit ripening, which includes the developmentally regulated

softening of plant tissues. This is achieved primarily through the loss of cell adhesion and

cell wall integrity via the activity of polysaccharide active enzymes on the middle lamella

and cell wall proper including pectin methylesterases that can loosen pectin gels and

allow the entrance of degradation enzymes such as pectate lyases and polygalacturonases

[102, 103]. Texture is also affected by the rheological properties of the cell wall

components present, particularly of matrix polymers such as pectins and arabinoxylans,

and highly glycosylated cell wall proteins. These matrix polymers are of particular

interest when extracted from cell walls or obtained as extruded mucilages or gums, as

they are used in for food, industrial and medical purposes as emulsifiers, glues and

coatings [104, 105]. Thus, texture properties can be modulated by manipulation of the

composition of cell walls, or the timing and degree of cell wall degradation via wall

active enzymes [103-105].

The stiffness of plant fibres is also of significant interest, ranging from individual

textile fibres such as cotton and linen (flax), to those that make up wood, to the use of

fibres derived from plant residues for the creation of novel composite materials. Unlike

the fruit ripening and pectins mentioned above that relate to primary cell walls, fibres are

generally derived from cells that have undergone secondary cell wall thickening. For example, cotton fibres elongate as hairs from the seed coat and, at maturity, are composed of a secondary cell wall that is mostly crystalline cellulose, while flax bast

258 fibres are derived from reinforced phloem cells of flax stems that have a more complex, gelatinous secondary cell wall [106-108]. The length, strength and stiffness of the mature fibres are critical for the production of textiles and for composite materials. Fibres with different biomechanical properties are required for different uses [106]. Wood is made up of many physically connected files of reinforced vascular and/or interfascicular fibre cells, depending on the type of tree, and the required structural parameters depend on the end use of the wood for construction or furniture versus providing fibres for products such as paper [2, 106]. The length and stiffness properties of individual fibres (or wood) can largely be modulated via alteration of cell wall properties (primary cell wall compliance for growth; degree and type of secondary cell wall reinforcement for stiffness and strength), either directly through manipulation of their biosynthesis or through the regulation of the timing and location of cells/cell types with desired primary cell wall or secondary cell wall properties [106, 109]. One could also envision for wood, and/or other plant products composed of more than individual fibres, the ability to create gradients of tissue, cell or cell wall properties to create actuator-type “smart” materials (i.e., that are capable of reacting to stimuli in their environment and responding in an adaptive manner). By studying how specific plant species take advantage of compositional and structural gradients to affect changes in their conformation through manipulating of tissue stiffness, it is possible to learn how to integrate these properties into the design of engineered plants with functional properties for advanced materials or sustainable construction (reviewed in [97, 110]) [96].

259

CONCLUDING REMARKS AND FUTURE DIRECTIONS

In this paper, we have laid out and discussed a number of morphological and compositional parameters affecting plant stiffness across multiple length scales ranging from organs to tissues to cells and cell walls. It is clear from this consideration that these properties are complex, intermingled and interdependent, making it difficult to determine the exact contribution of each. What is needed is a comprehensive, systematic and consistent multiscale mechanical analysis of structural parameters across length scales to feed into an integrated model of the development of plant stiffness.

In an attempt to look at these factors in a consistent background and avoid complications from considerations across organ type and species and plant groups, we have reviewed the literature of stem stiffness within the reference plant Arabidopsis, where the use of genetic dissection and transgenic plants has allowed some consideration of individual or related groups of variables. It is clear from this review that there are a number of useful mutant lines affecting key parameters. However, most either have not been mechanically tested, or tested with a very limited set of techniques (e.g., only tensile breaking force), or not in a comparable manner in terms of tissue and plant stage (see

Tables 5.1 and 5.2). While Arabidopsis is not a perfect model for examining all morphological and compositional variables affecting plant stiffness, it has the advantage of ease of use and the availability of mutant lines affecting multiple stiffness parameters.

As such, comprehensive analysis in Arabidopsis could be used in combination with current and new data from other plant systems to develop an integrated model of plant stiffness.

260

The next step is a systematic investigation of stiffness parameters across length scales in Arabidopsis using a combination of relevant tissues (e.g., starting with mature stems) and mutants such as those described earlier in this paper (listed in Tables 5.1 and

5.2). It is imperative, as discussed above, to have a comprehensive analysis of mutant plant morphological, structural and biochemical phenotypes to understand the full biological consequences of the genetic alterations in question in addition to mechanical testing to inform the interpretation of the results. Further, for mechanical experiments to be comparable, it is critical to ensure that there is consistency not only in the tissue used, but also in such things as the Arabidopsis accession studied, plant developmental stage, dimensions/location of sample within that tissue, and hydration state (e.g., for stems, it would be useful to consider their mechanical properties both in hydrated and dry states to allow consideration of the mechanics in living plants versus materials derived from dry, harvested plants). The mechanical experiments performed also need to be comprehensive and include different loading conditions (e.g., cyclic tension and compression, strength and fracture characterization, micro- and nanoindentation, and atomic force microscopy

[Box 2]). Molecule-level monitoring tools such as X-ray scattering or Raman microspectroscopy in combination with mechanical testing would also inform structure- function analysis (see [39, 40], and Box 2). Tests should move beyond one or two dimensions, take turgor pressure into consideration, combine in-situ mechanical testing with imaging techniques such as X-ray computed tomography (Box 2), and/or be performed in a non-destructive fashion to ensure that the properties measured are true to native plant conditions [24, 36].

261

Finally, there is a good deal of scope for addressing agricultural and new biodegradable and biocompatible material needs through the tailoring of stiffness of plants and products derived from them such as fibres and wood. The creation and harnessing of gradients of stiffness parameters to create adaptive or ‘smart’ materials is of particular interest. In addition to the use of genetic engineering of genes related to tissue organization, cellular differentiation and cell wall synthesis indicated above, it is possible to adapt plant structure through other means. These include taking advantage of plant responses to plant spacing, mechanical perturbation, and growth or treatment under certain nutrient or hormone conditions [13, 99, 106].

262

Figure 5.1.

263 Figure 5.1. A Schematic of the Hierarchy of Mechanical Parameters Influencing

Stiffness Across Different Length Scales in Plants. Examples of the different parameters are depicted and each length scale is described. The whole plant and organ level (meter-millimeter) includes examining the mechanical properties of both the entire plant as a whole structure, as well as individually studying its various organs (e.g., stem, petiole, leaf, root, etc.). The tissue level (centimeter- micrometer) involves examining individual tissues for their mechanical properties as well as how the organization of different tissues as a collective whole can influence plant/organ stiffness. The cell level

(millimeter-micrometer) includes studying isolated cells for their mechanical properties as well as how cells interact with neighbouring cells to influence the mechanical parameters observed at the tissue level. Finally, the cell wall level (micrometer- nanometer) involves studying the mechanical properties of the cell wall in terms of its composition, structure and interaction between its various components in order to understand how this building block of plant material ultimately affects the mechanical parameters observed at higher length scales (cell, tissue, organ and whole plant). Adapted from [3, 111].

264

Figure 5.2.

265 Figure 5.2. Examples of Uniaxial Mechanical Forces That Act on Objects, and

Commonly Used Mechanical Tests Used to Investigate the Mechanical Properties of

Objects. (A) Mechanical forces include: tension that pulls and can lengthen or break a material/object; compression that pushes and can shorten or crush/buckle a material/object; and shear that comprises two misaligned tensile or compressive forces that causes a material/object to slide against itself in opposite directions, usually leading to tearing. (B) Mechanical tests commonly used to analyze materials include: tensile tests in which a specimen is secured at both ends and pulled until failure or until a specific amount of force is reached (e.g., in the case of cyclic testing), and compression tests in which a specimen is held in place between two plates and is pressed until failure or until a specific amount of force is reached. If the holders or plates keeping the specimen in place are not perfectly aligned, shear tensile and compressive tests can be performed.

Three-point and four-point bending tests hold a sample in place between three and four fixtures, respectively, and the specimen is bent until failure or a specific amount of force is reached. Bending involves tensile and compressive forces acting on a specimen, and can be measured by calculating flexural rigidity from both three and four-point bending tests. In the schematic above, the arrows show the direction that the forces act in (A) and the direction in which each force type is applied (B). The grey rectangles and circles represent the different fixtures and holders that are used to keep samples in place during testing. The brown rectangle represents a load cell that reads the change in the amount of force applied to a specimen during the course of a mechanical test. Schematic adapted from [23, 40, 78, 113].

266

Figure 5.3.

267 Figure 5.3. Examples of Arabidopsis Inflorescence Stem Treatments and Mutants

that Affect Hierarchical Parameters Governing Stiffness. Parameters that have been

mechanically tested and shown to affect stiffness are marked with an asterisk (*). (A*)

Mechanically unperturbed (i) and mechanically perturbed (ii) mature WT Col-0 stems; a

reduction in stem height often accompanies changes in stiffness (scale: each square is

1cm) (reproduced with permission from [13]). (B) Variation in stem diameter is

correlated with altered stiffness [14]; mutants such as MYB87-SRDX (ii) that alter

diameter compared to WT (i Col-0) could possibly be used to investigate this parameter

further (scale bar: 200µm, reproduced with permission from [15]). (C) Mutants that could

potentially be used to examine the influence of cross-sectional shape on stiffness include

the oblong-shaped fasciata1 (i) and men1/+ (iii) stems when compared with rounded WT

(ii Col-0) stems (reproduced with permission from [17] (i) and [18] (ii, iii)). (D*) The

organizational conversion of collateral vascular bundles in WT Ler (i) to amphivasal

bundles and the change in interfascicular fibre tissue proportion in the amphivasal

vascular bundle1 overexpression mutant (ii) accompanied a reduction in stem stiffness

(reproduced with permission from [19]). Mutants such as high cambial activity1 (iii-

WT/iv- mutant) that also display organizational and tissue proportion phenotypes could be used to further explore the influence of these parameters on stiffness (scale bar:

100µm) (reproduced with permission from [21]). (E*) Changes in tissue density were found to reduce bending stiffness in mechanically perturbed Col-0 plants (ii) by 50% as compared to unperturbed WT plants (i) (scale bar: 0.01mm) (reproduced with permission from [13]). (F) Mutants, such as quasimodo1 that displays reduced adhesion (ii) compared to WT Ws (i) hypocotyls, could be used to investigate the mechanical

268 implications of adhesion in relation to stem stiffness (scale bar: 50µm) (reproduced with permission from [33]). (G*) Turgor pressure has been previously demonstrated to impart stiffness to plant tissue. (H*) Changes in cell wall composition affect stiffness; mutants, such as irregular xylem4 (ii), show reduced bending stiffness compared to WT (i Ler) as a result of decreased lignin content (visualized with phloroglucinol staining) (reproduced with permission from [63]). (I*) Cell wall thickness is correlated with stiffness; amphivasal vascular bundle1 (ii) whose walls were thinner than WT Col-0 (i) exhibited a reduction in stiffness (scale bar: 13µm) (reproduced with permission from [19]). (J)

Microfibril angle (MFA) is known to affect stiffness in other plant species; mutants such as xyloglucan xylosyltransferase (xxt)1/xxt2 could potentially provide the opportunity to study the effect of MFA on stiffness in Arabidopsis (scale bar: 500nm) (reproduced with permission from [50]). IF= interfascicular fiber, VB= vascular bundle, P= pith, Ph= phloem, Xy= xylem, C/Co= cortex, Ep= epidermis.

269

Figure 5.4.

270 Figure 5.4. The Role of Stiffness Gradients in Plant Function. (A) Cross sectional

gradients in petiole and sheath of palm branch (reproduced with permission from [16]).

(B) (i) Composite image of the giant reed Arundo donax cross section stained with toluidine blue and visualized with two-dimensional UV absorbance scans; and (ii) cross section stained with phloroglucinol/hydrochloric acid that stains lignified tissues in red.

The gradient in absorbance indicates variation in the degree of lignification (reproduced with permission from [88]). (C) The functionally graded structure of moso bamboo

(reproduced with permission from [92]). (D) The spikemoss Selaginella lepidophylla in

(i) hydrated and (ii) dried states and (iii) the curling pattern of its dead and living stems

(reproduced with permission from [96]). Compositional gradients in the apical cross- section of a living stem in (iv) the degree of lignification and (v) xylan distribution and the structural gradients reflected in (vi) the density gradients in the cross section and (vii) the variation of cells orientations observed in the longitudinal section (Chapter 3).

271

BOX 1. TERMINOLOGY RELATED TO MECHANICS OF BIOLOGICAL

SYSTEMS IN THE CONTEXT OF THIS PAPER

Stiffness parameters: these are inherent morphological and compositional features that

predetermine the extent to which an object will deform in response to an applied force. In

the case of plants, the contribution of different parameters to overall stiffness depends on

the species type, developmental stage and environmental history of the plant.

Length scale: this describes the different orders of magnitude at which parameters of

stiffness appear (Figure 5.1) [1, 111].

Load: an applied force (internal or external) that generates mechanical stress in an object

[4, 23, 27].

Stress: the amount of force (F) applied across a given area (A) of an object (stress = F/A,

N·m2) [4, 23, 27].

Strain: the change in length (l) of an object in response to an applied stress (strain =

∆l/lo, dimensionless) [4, 23, 27].

Stiffness: measures the resistance offered by an elastic object to deformation. A stiff object resists deformation in response to large stresses while a compliant (i.e., less stiff) object deforms in response to relatively small stresses [4, 23, 27].

Flexural rigidity: the ability of an object to resist bending in response to an applied force. Rigid objects resist bending deformation in response to large stresses and flexible objects bend in response to small stresses [4, 27].

Strength: the amount of load required to cause an object to break (e.g., tear, snap, lose cohesiveness, etc.) [4, 23, 27].

272

Failure: the point at which an object breaks (e.g., tears, snaps, buckles, etc.) [4, 23, 27].

Tension, compression and shear (Figure 5.2A): types of forces that can act upon an object. Tensile forces pull and cause lengthening; compressive forces push and cause shortening; shear forces are misaligned tensile or compressive forces that can cause tearing [4, 27].

Actuation: the generation of mechanical energy and a subsequent change in an object’s shape in response to an external, non-mechanical stimulus [57, 112].

Reversible actuation: the release of energy that allows an object to return to its original shape when a stimulus is removed [96, 97].

Functional gradients: transitions in morphology or composition along the length or cross-section of an object that predetermine the manner in which the object will deform in response to a stimulus [92].

Smart materials and structures: this refers to synthetic materials or structures that have built-in actuating properties based on concepts including those derived from studying biological actuators (e.g., pinecones) [97].

273

BOX 2. MECHANICAL TEST TYPES COMMONLY USED

FOR TESTING PLANTS

This is not intended to be an exhaustive list of techniques, but rather a general overview of techniques commonly used to study mechanical properties in plants. For a comprehensive overview of the available techniques along with their advantages and caveats, please refer to [24, 36, 40].

Types of Mechanical Testing

Uniaxial testing (Figure 5.2B): mechanical testing performed along a single axis of an object [4, 23].

Tensile test (Figure 5.2B): this involves securing a sample at both ends (usually along the longitudinal axis in the case of plants) and pulling with an applied amount of force to cause the object to lengthen. Tensile tests can measure tensile stiffness and strength [4,

23].

Compression test (Figure 5.2B): the opposite of a tensile test, this involves holding a sample in place between two plates and pushing with an applied force to cause an object to shorten. Compression tests can measure compressive stiffness and strength [4, 23].

Three-point and four-point bending tests (Figure 5.2B): these test types involve placing a sample between three (3-point) or four (4-point) fixtures and bending it. This test type applies tensile and compressive forces to an object and can be used to measure flexural stiffness and strength [78, 113].

274

Monotonic and cyclic testing: monotonic testing refers to a test in which an increasing amount of force is applied to an object and is usually continued until it fails. In contrast, cyclic testing refers to a test in which a force (below the amount needed to fail an object) is applied and subsequently removed over multiple cycles. Unlike monotonic testing, cyclic testing can measure structural fatigue, providing an idea of the behaviour of an object under repeatedly applied amounts of force [4, 114].

Micro- and nano-indentation: indentation assays utilize indenter probes with varying tip geometries in order to indent tissues and cells. The amount of deformation of cell surfaces as a result of indentation can then be used to measure the apparent stiffness of the sample [23, 24, 40, 115].

Microscopy and Spectroscopy Techniques Used in Combination with

Mechanical Testing Techniques

Mechanical tests can be combined with microscopy and spectroscopy techniques in order to provide structure-function relationships between observed mechanical behaviours and morphological features of a given plant, organ, tissue, cell or cell wall.

Below are a few examples commonly used to simultaneously study structure and function in plants:

Atomic and cellular force microscopy (AFM and CFM): both AFM and CFM use a probe to scan the surface of specimens, and can isolate cell wall stiffness from combined cell wall and turgor stiffness. These two indentation techniques combined with microscopy allow simultaneous collection of morphological and mechanical information that can be overlaid to create topographical maps of cell and tissue surface stiffness,

275 providing insight into the relationship between structure and mechanical function [24,

115, 116].

Small angle X-ray scattering (SAXS) and wide-angle X-ray scattering (WAXS): samples are exposed to X-rays and X-ray photons bounce off the electrons in the specimen. The resulting photon scatter is a reflection of the electron density of the sample, and is used as a measurement to provide information on the structure of the specimen. SAXS measures small scattering angles (near 0°) whereas WAXS measures large scattering angles (5°>). Cell wall polymers can be differentiated from each other based on their electron density (SAXS), or by atomic changes in polymer structure

(WAXS). SAXS and WAXS can both be used to measure cellulose microfibril angle in plant cell walls [89, 117, 118].

X-ray computed tomography: this non-invasive technique utilizes X-ray beams to scan samples from different angles. This two-dimensional information can be compiled using computer software to generate three-dimensional reconstructions of the sample. Thus X- ray computed tomography provides structural information including sample density, as well as and transverse and longitudinal morphology [119, 120].

Raman confocal spectroscopy: this non-destructive technique is able to chemically image the cell wall, identifying cell wall polymers based on their specific chemical signature. Samples are exposed to a laser beam and photons are deflected off the sample surface. Scattered photons can then be detected and can be used to measure changes in the vibration of chemical functional groups of different polymers in the sample. When

Raman spectroscopy is combined with confocal microscopy, it is possible to generate chemical image maps that combine compositional and structural information in order to

276 spatially identify the relative intensity of specific cell wall polymers across a cell or tissue. Mechanical testing stages can be used in combination with the Raman confocal spectroscope in order to map real time changes in chemical spectra during mechanical testing [23, 121, 122].

Solid-state nuclear magnetic resonance: nuclear magnetic resonance is a non- destructive, spectroscopic technique that examines electron distribution within a sample in order to identify specific chemical components. The nucleus of an atom has a small magnetic field that is generated by electrons orbiting the nucleus. Differences in the orientation of this magnetic field and dipolar interactions between atoms give rise to unique signatures for different chemicals, and hence unique, detectable signatures for specific plant cell wall polymers. Solid-state nuclear magnetic resonance can provide information on the structure of individual cell wall polymers, similar to Raman spectroscopy, as well as the overall in vivo architecture of the cell wall. Changes in electron distribution can also be used to infer mechanical properties of the cell wall based on its overall structure and polymer makeup [45, 123, 124].

277

Table 5.1. Mechanical Properties of Arabidopsis Stem Structure and Cell Wall Mutants

Basic Resultc

Gene Function/Predicted Function Mutant Phenotype Tissue Testeda Test Typeb Stiffnessd Strengthe Referencesf

Structural mutants

STURDY Patatin-like protein Increased cell number and stem Stem Qualitative observation ↑ [101] diameter

MYB87 Transcription factor OE of MYB87 fused to SRDX Stem Not tested [15] transcriptional repressor domain

CLAVATA1 Receptor kinase with leucine- Fasciated stem Stem Not tested [17] rich repeat

FASCIATA1 Chromatin Assembly Factor- Fasciated stem Stem Not tested [17] 1 subunit

FASCIATA2 Chromatin Assembly Factor- Fasciated stem Stem Not tested [17] 1 subunit

MERISTEM MicroRNA miR166a Fasciated stem Stem Not tested [18] ENLARGEME NT1

IFL1/REVg Transcription factor Lack IF Stem Tensile ↓ [20]

avb1 OE allele: Amphivasal Stem Tensile ↓ [19] vascular bundles; reduced IF cell wall thickness

HCA1 Unknown Altered vascular bundle Stem Not tested [21] organization; reduction in IF tissue

278

HCA2 Transcription factor Altered vascular bundle Stem Not tested [22] organization; reduction in IF tissue

(Natural n/a Natural variation in morphology Stem Tensile ~↑↓j ~↑↓j [14] accessions and composition x12)

(WT after n/a Altered stem morphology Stem 3-Point Bending ↓ [13] mechanical perturbation)

Primary Cell Walls

CESA6/PROC PCW cellulose synthase Reduced cellulose Hypocotyl Tensile ↓ [50] USTE subunit

AtKTN1/FRA2/ Katanin MT severing Altered MF organization; reduced Hypocotyl Tensile ↓ ↓ [29] BOT protein cellulose

MUR2 XyG fucosyltransferase Reduced XyG Fuc Hypocotyl Tensile ↓ ↓ [23, 29, 35, 125]

Pedicel Tensile ~ ~ [29]

MUR3 XyG galactosyltransferase Reduced XyG Gal Hypocotyl Tensile ↓ ↓ [23, 29, 125]

Pedicel Tensile ~ ~ [29]

XXT1 XXT2h XyG xylosyltransferases Lack detectable XyG Hypocotyl Tensile ↓ ↓ [51]

Pedicel Tensile ↓ [126]

MUR1 GDP-L-fucose synthase Reduced Fuc in XyG and RG II Hypocotyl Tensile ↓ ↓/~ [29, 35]

Pedicel Tensile ~ ↓ [29]

QUA2 HG methyltransferase Reduced HG Hypocotyl Tensile ↓ ~ [35]

279

PME5 Pectin methylesterase OE: decreased HG esterification Hypocotyl AFM ↓ [54]

Meristem AFM ↓ [53]

PMEI3 Pectin methylesterase OE: increased HG esterification Hypocotyl AFM ↑ [54] inhibitor

Meristem AFM ↑ [53]

ARAD1 Arabinan Decreased arabinans Stem Compression ↑ [52] ARAD2 arabinosyltransferases

Stem Indentation ~ [52]

AtKINESIN- Kinesin MT motor protein Altered matrix polysaccharide Leaf AFM ↑ [127] 4A/FRA1 secretion (MF organization?) AtKINESIN-4C

Secondary Cell Walls

CESA7/IRX3/FRA SCW cellulose Reduced cellulose Stem 3-Point Bending ↓ ↓ [58] 5 synthase subunit

Stem Tensile ↓ [64]

CESA8/IRX1/FRA SCW cellulose Reduced cellulose Stem 3-Point Bending ↓ ~ [58] 6 synthase subunit

KORRIGAN/IRX2 β-glucanase in irx2 SCW-specific allele: reduced Stem 3-Point Bending ↓ ↓ [58] cellulose synthase cellulose complex

AtKTN1/FRA2/B Katanin MT severing Altered MF organization; reduced Stem Tensile ↓ ↓ [59, 66, 128] OT protein cellulose

Stem 3-Point Bending ↓ ↓ [129]

280

AtKINESIN- Kinesin MT motor Altered matrix polysaccharide Stem Tensile ↓ [130] 4A/FRA1 protein secretion (MF organization?)

Stem 3-Point Bending ↓ ↓ [131]

IRX9 Xylosyltransferase in Severely reduced xylan; reduced Stem Tensile ↓ [68] xylan backbone cellulose & lignin elongation

GUX1 GUX2 Xylan Lack substitution of xylan backbone Stem 4-Point Bending ↓ [75] glucuronyltransferases

FRA8/IRX7 Glycosyltransferase in Severely reduced xylan; reduced Stem Tensile ↓ [67] the synthesis of xylan cellulose & lignin reducing end tetrasaccharide

IRX8/GAUT12 Glycosyltransferase in Severely reduced xylan; reduced Stem Tensile ↓ [68] the synthesis of xylan cellulose & lignin reducing end tetrasaccharide

PARVUS/GATL1 Glycosyltransferase in Reduced xylan Stem Tensile ↓ [132] the synthesis of xylan reducing end tetrasaccharide

RWA1 RWA2 Acetyl Coenzyme A Reduced acetylation of xylan; altered Stem Tensile ↓ [70] RWA3 RWA4 transporters proportion of GlcA vs Me-GlcA substitution

ESK1/TBL29 Xylan acetyltransferase Reduced acetylation of xylan Stem Tensile ↓ [62, 71-73]

[ESK1/TBL29] Xylan Reduced acetylation of xylan Stem Tensile ↓ [73] TBL3 TBL31i acetyltransferases

281

[ESK1/TBL29] Xylan Reduced acetylation of xylan; Stem Tensile ↓ [72] TBL34 acetyltransferases reduced cellulose

[ESK1/TBL29] Xylan Reduced acetylation of xylan; Stem Tensile ↓ [72] TBL34 TBL35 acetyltransferases reduced cellulose

[ESK1/TBL29] Xylan Reduced acetylation of xylan; Stem Tensile ↓ [62] TBL33 acetyltransferases reduced xylan & cellulose

[ESK1/TBL29] Xylan Reduced acetylation of xylan; Stem Tensile ↓ [62] TBL32 TBL33 acetyltransferases reduced xylan & cellulose

CSLA2 CSLA3 Glycosyltransferases Lack detectable glucomannan Stem 4-Point Bending ~ ~ [76] CSLA9 for glucomannan backbone synthesis

IRX4/CCR1 Cinnamoyl- Reduced lignin Stem Tensile ↓ ↓ [128] CoenzymeA reductase for lignin monomer synthesis

Stem 3-Point Bending ↓ ↓ [63]

FLA11/IRX13 Fasciclin-like Reduced cellulose, slightly increased Stem Tensile ↓ ↓ [133] FLA12 arabinogalactan lignin, slightly higher MFA, reduced proteins (AGPs) with AGPs possible roles in cellulose synthesis

Stem 3-Point Bending ~ ~ [133]

PME35 Pectin methylesterase Increased methyl-esterification of HG Stem Compression ↓ [134] aSimplified to basic type, see reference for details of region and age; bSimplified to basic type, see references for details of testing regime as well as Box 2 and Figure 2 for a description of the mechanical tests; cResults noted as ↑ = increased compared to wild type, ↓ = decreased compared to wild type, ~ = similar to wild type, blank = not reported; note that since most testing

282 experiments are not comparable, not showing degree of increase or decrease; dStiffness = Modulus of Elasticity (stress/strain); eStrength = Ultimate Strength (stress or force at failure) fReferences for the mechanical testing of these mutants- for recent reviews on the molecular function of PCW & SCW genes see [42]; [31]; [55]; [41]; [56]; [135] for cloning of CLAVATA1 and FASCIATED1 and 2 see [136]; [137] gGene names separated by backslash (/) indicate alternate names hResults either showed no variation, or an increase or decrease in stiffness and strength between natural accessions iGene names separated with space represent independent genes; results described regard double (triple, quadruple…) mutants jSquare brackets indicate that the esk1/tbl29 phenotype is enhanced (worsened) by the creation of double and/or triple mutants with the other listed TBL genes

PCW = primary cell wall; SCW = secondary cell wall; XyG = xyloglucan; HG = homogalacturonan; RG II = rhamnogalacturonan II; Fuc = fucose; Gal = galactose; GlcA = glucuronic acid; Me-GlcA = methylated glucuronic acid; AGP = arabinogalactan protein; MT = microtubules; MF = cellulose microfibrils; MFA = microfibril angle; IF = interfascicular fiber; OE = overexpression; AFM = atomic force microscopy; n/a = not applicable; WT = wild type

283

Table 5.2. Comparison of Mechanical Properties of Primary Cell Wall, Secondary Cell Wall and Stem Structural Mutants of Arabidopsis

Gene Phenotype Stiffnessa Strengtha References

Primary Cell Wall Mutants: Hypocotyl Tensile Tests - Ryden et al. 2003b

Atktn1/fra2/botc Reduced cellulose +++ +++ [29] WT with 0.25µM DCBd ~60% cellulose ++ ++ [29] mur2 Reduced XyG Fuc WT ++++ [29] mur3 Reduced XyG Gal ++ +++ [29] mur1 Reduced XyG & RGII Fuc +++ +++ [29]

Primary Cell Wall Mutants: Hypocotyl Tensile Tests - Burgert & Dunlop 2011e mur2 Reduced XyG Fuc ++++ +++++ [23, 35, 39] mur3 Reduced XyG Gal ++ ++++ [23, 39] xxt1 xxt2f No XyG +++ ++++ [39, 51] mur1 Reduced XyG & RGII Fuc ++++ +++++ [35, 39] qua2 Reduced HG (50%) ++++ +++++ [35, 39]

Secondary Cell Wall Mutants: Basal Stem Three-Point Bend Testsg cesa8/irx1/fra6g 40% cellulose ++ +++++ [58]

284 korrigan/irx2 36% cellulose ++ ++++ [58] cesa7/irx3/fra5 18% cellulose + +++ [58] irx4/ccr1 50% lignin ++ ++ [63]

Secondary Cell Wall and Structural Mutants: Basal Stem Breaking Force Testsh cesa7/irx3/fra5 45% cellulose nd + [64] Atktn1/fra2/bot 80% cellulose nd +++ [59] Atkinesin-4a/fra1 MFA, reduced matrix nd +++ [130] irx9 45% xylan, reduced cellulose & lignin nd + [68] fra8/irx7 42% xylan, reduced cellulose & lignin nd + [67] irx8/gaut12 25% xylan, reduced cellulose & lignin nd + [68] parvus/gatl1 ~50% xylan (reduced cellulose & lignin?) nd + [69] rwa1 rwa2 rwa3 rwa4 60% acetylation of xylan nd ++++ [70] esk1/tbl29 70% acetylation of xylan nd +++ [62, 71-73] esk1/tbl29 tbl3 tbl31 56% acetylation of xylan nd ++ [73] esk1/tbl29 tbl34 63% acetylation of xylan, reduced cellulose nd +++ [72] esk1/tbl29 tbl34 tbl35 63% acetylation of xylan, reduced cellulose nd ++ [72] esk1/tbl29 tbl33 37% acetylation, reduced cellulose & xylan nd + [62] esk1/tbl29 tbl32 tbl33 20% acetylation, reduced cellulose & xylan nd + [62]

285 ifl1/rev Missing IF nd + [20] avb1/ifl1/rev Amphivasal vascular bundles, reduced IF cell nd +++ [19] wall thickness (Fig. 3I) aAbsolute stiffness and strength measurements from the referenced papers were converted to percent of wild type, then noted on a 5+ scale where each + represents 20% of the total WT stiffness or strength value; note that all mutants listed are significantly different than wild type, thus a rating of +++++ means within a 20% range of the WT value. If the mutant is the same as WT, “WT” is noted. bAll data estimated from Ryden et al. 2003 from different figures cGene names separated by backslash (/) indicate alternate names dWild type hypocotyls grown in the presence of cellulose synthesis inhibitor 2,6-dichlorobenzonitrile (DCB; 0.25µM) eNote that tests were performed on 4 and 6 day old hypocotyls and are not absolutely comparable, comparisons listed here are based on Burgert & Dunlop 2011, see Burgert & Dunlop 2011 plus listed primary papers for details fGene names separated with space represent separate genes, results described regard double mutant gNote that tests were performed by the same group with the same apparatus, but in two different publications; 48d results taken hNote that tests were performed by the same group with the same apparatus, but in multiple publications, and the age of the stem can differ, see references for details WT = wild type; DCB = 2,6-dichlorobenzonitrile; XyG = xyloglucan; HG = homogalacturonan; RG II = rhamnogalacturonan II; Fuc = fucose; Gal = galactose; MFA = microfibril angle; IF = interfascicular fiber

286

REFERENCES

[1] O.E. Jensen, J.A. Fozard, Multiscale Models in the Biomechanics of Plant Growth, Physiology, 30 (2015) 159-166. [2] L.J. Gibson, The hierarchical structure and mechanics of plant materials, J R Soc Interface, 9 (2012) 2749-2766. [3] T. Speck, I. Burgert, Plant stems: functional design and mechanics, Annual Review of Materials Research, 41 (2011) 169-193. [4] K.J. Niklas, Plant biomechanics: an engineering approach to plant form and function, University of Chicago press, 1992. [5] V. Mirabet, P. Das, A. Boudaoud, O. Hamant, The role of mechanical forces in plant morphogenesis, Annu Rev Plant Biol, 62 (2011) 365-385. [6] A.J. Bidhendi, A. Geitmann, Relating the mechanics of the primary plant cell wall to morphogenesis, J Exp Bot, 67 (2016) 449-461. [7] B. Gardiner, P. Berry, B. Moulia, Review: Wind impacts on plant growth, mechanics and damage, Plant Science, 245 (2016) 94-118. [8] S. Burgess, D. Pasini, Analysis of the structural efficiency of trees, Journal of Engineering Design, 15 (2004) 177-193. [9] K.J. Niklas, Size-dependent Allometry of Tree Height, Diameter and Trunk-taper, Annals of Botany, 75 (1995) 217-227. [10] A. Bejan, S. Lorente, J. Lee, Unifying constructal theory of tree roots, canopies and forests, Journal of theoretical biology, 254 (2008) 529-540. [11] D. Lopez, E. De Langre, S. Michelin, A space-averaged model of branched structures, Computers & Structures, 146 (2015) 12-19. [12] D. Lopez, S. Michelin, E. De Langre, Flow-induced pruning of branched systems and brittle reconfiguration, Journal of Theoretical Biology, 284 (2011) 117-124. [13] C. Paul-Victor, N. Rowe, Effect of mechanical perturbation on the biomechanics, primary growth and secondary tissue development of inflorescence stems of Arabidopsis thaliana, Annals of Botany, 107 (2011) 209-218. [14] C.P. MacMillan, P.J. O’Donnell, A.-M. Smit, R. Evans, Z.H. Stachurski, K. Torr, M. West, J. Baltunis, T.J. Strabala, A survey of the natural variation in biomechanical and cell wall properties in inflorescence stems reveals new insights into the utility of Arabidopsis as a wood model, Functional Plant Biology, 40 (2013) 662-676. [15] S. Fujiwara, N. Mitsuda, Y. Nakai, K. Kigoshi, K. Suzuki, M. Ohme-Takagi, Chimeric repressor analysis identifies MYB87 as a possible regulator of morphogenesis via cell wall organization and remodeling in Arabidopsis, Biotechnology letters, 36 (2014) 1049-1057. [16] W. Liu, N. Wang, X. Jiang, Y. Peng, The mechanical influences of the graded s ratio of palm ׳distribution in the cross-sectional shape, the stiffness and Poisson branches, Journal of the Mechanical Behavior of Biomedical Materials, 60 (2016) 203- 211. [17] H.M. Ottoline Leyser, I.J. Furner, Characterisation of three shoot apical meristem mutants of Arabidopsis thaliana, Development, 116 (1992) 397-403. [18] J. Kim, J.-H. Jung, J.L. Reyes, Y.-S. Kim, S.-Y. Kim, K.-S. Chung, J.A. Kim, M. Lee, Y. Lee, V. Narry Kim, N.-H. Chua, C.-M. Park, microRNA-directed cleavage of

287 ATHB15 mRNA regulates vascular development in Arabidopsis inflorescence stems, The Plant Journal, 42 (2005) 84-94. [19] R. Zhong, Z.-H. Ye, amphivasal vascular bundle 1, a Gain-of-Function Mutation of the IFL1/REV Gene, Is Associated with Alterations in the Polarity of Leaves, Stems and Carpels, Plant and Cell Physiology, 45 (2004) 369-385. [20] R. Zhong, J.J. Taylor, Z.-H. Ye, Disruption of interfascicular fiber differentiation in an Arabidopsis mutant, The Plant Cell, 9 (1997) 2159-2170. [21] C. Pineau, A. Freydier, P. Ranocha, A. Jauneau, S. Turner, G. Lemonnier, J.-P. Renou, P. Tarkowski, G. Sandberg, L. Jouanin, B. Sundberg, A.-M. Boudet, D. Goffner, M. Pichon, hca: an Arabidopsis mutant exhibiting unusual cambial activity and altered vascular patterning, The Plant Journal, 44 (2005) 271-289. [22] Y. Guo, G. Qin, H. Gu, L.-J. Qu, Dof5. 6/HCA2, a Dof transcription factor gene, regulates interfascicular cambium formation and vascular tissue development in Arabidopsis, The Plant Cell, 21 (2009) 3518-3534. [23] I. Burgert, Exploring the micromechanical design of plant cell walls, American journal of botany, 93 (2006) 1391-1401. [24] H. Vogler, D. Felekis, B. Nelson, U. Grossniklaus, Measuring the Mechanical Properties of Plant Cell Walls, Plants, 4 (2015) 167. [25] S.A. Braybrook, H. Jönsson, Shifting foundations: the mechanical cell wall and development, Current Opinion in Plant Biology, 29 (2016) 115-120. [26] A. Ivakov, S. Persson, Plant cell shape: modulators and measurements, Functional Imaging in living Plants-Cell Biology meets Physiology, (2015) 102. [27] S.A. Wainwright, Mechanical design in organisms, Princeton University Press, 1982. [28] L.J. Gibson, Biomechanics of cellular solids, J Biomech, 38 (2005) 377-399. [29] P. Ryden, K. Sugimoto-Shirasu, A.C. Smith, K. Findlay, W.-D. Reiter, M.C. McCann, Tensile properties of Arabidopsis cell walls depend on both a xyloglucan cross- linked microfibrillar network and rhamnogalacturonan II-borate complexes, Plant physiology, 132 (2003) 1033-1040. [30] F.B. Daher, S.A. Braybrook, How to let go: pectin and plant cell adhesion, Frontiers in plant science, 6 (2015). [31] Z. Hao, D. Mohnen, A review of xylan and lignin biosynthesis: foundation for studying Arabidopsis irregular xylem mutants with pleiotropic phenotypes, Critical reviews in biochemistry and molecular biology, 49 (2014) 212-241. [32] J. Barros, H. Serk, I. Granlund, E. Pesquet, The cell biology of lignification in higher plants, Annals of botany, (2015) mcv046. [33] S. Bouton, E. Leboeuf, G. Mouille, M.-T. Leydecker, J. Talbotec, F. Granier, M. Lahaye, H. Höfte, H.-N. Truong, QUASIMODO1 encodes a putative membrane-bound glycosyltransferase required for normal pectin synthesis and cell adhesion in Arabidopsis, The Plant Cell, 14 (2002) 2577-2590. [34] G. Mouille, M.C. Ralet, C. Cavelier, C. Eland, D. Effroy, K. Hématy, L. McCartney, H.N. Truong, V. Gaudon, J.F. Thibault, Homogalacturonan synthesis in Arabidopsis thaliana requires a Golgi‐localized protein with a putative methyltransferase domain, The Plant Journal, 50 (2007) 605-614. [35] W. Abasolo, M. Eder, K. Yamauchi, N. Obel, A. Reinecke, L. Neumetzler, J.W. Dunlop, G. Mouille, M. Pauly, H. Höfte, Pectin may hinder the unfolding of xyloglucan

288 chains during cell deformation: implications of the mechanical performance of Arabidopsis hypocotyls with pectin alterations, Molecular Plant, 2 (2009) 990-999. [36] A. Weber, S. Braybrook, M. Huflejt, G. Mosca, A.-L. Routier-Kierzkowska, R.S. Smith, Measuring the mechanical properties of plant cells by combining micro- indentation with osmotic treatments, Journal of Experimental Botany, 66 (2015) 3229- 3241. [37] D.J. Cosgrove, Plant cell wall extensibility: connecting plant cell growth with cell wall structure, mechanics, and the action of wall-modifying enzymes, J Exp Bot, 67 (2016) 463-476. [38] L. Beauzamy, J. Derr, A. Boudaoud, Quantifying hydrostatic pressure in plant cells by using indentation with an atomic force microscope, Biophysical journal, 108 (2015) 2448-2456. [39] I. Burgert, J.W. Dunlop, Micromechanics of cell walls, in: Mechanical Integration of Plant Cells and Plants, Springer, 2011, pp. 27-52. [40] I. Burgert, T. Keplinger, Plant micro-and nanomechanics: experimental techniques for plant cell-wall analysis, Journal of experimental botany, (2013) ert255. [41] Y.B. Park, D.J. Cosgrove, Xyloglucan and its interactions with other components of the growing cell wall, Plant Cell Physiol, 56 (2015) 180-194. [42] M.A. Atmodjo, Z. Hao, D. Mohnen, Evolving views of pectin biosynthesis, Annual Review of Plant Biology, 64 (2013) 747-779. [43] G. Levesque-Tremblay, J. Pelloux, S.A. Braybrook, K. Müller, Tuning of pectin methylesterification: consequences for cell wall biomechanics and development, Planta, 242 (2015) 791-811. [44] S. Wolf, S. Greiner, Growth control by cell wall pectins, Protoplasma, 249 (2012) 169-175. [45] M. Dick-Pérez, Y. Zhang, J. Hayes, A. Salazar, O.A. Zabotina, M. Hong, Structure and interactions of plant cell-wall polysaccharides by two-and three-dimensional magic- angle-spinning solid-state NMR, Biochemistry, 50 (2011) 989-1000. [46] T. Wang, Y.B. Park, J.C. Daniel, M. Hong, Cellulose-Pectin Spatial Contacts Are Inherent to Never-Dried Arabidopsis thaliana Primary Cell Walls: Evidence from Solid- State NMR, Plant physiology, (2015) pp. 00665.02015. [47] C.T. Anderson, We be jammin’: an update on pectin biosynthesis, trafficking and dynamics, Journal of experimental botany, 67 (2016) 495-502. [48] E. Gendreau, J. Traas, T. Desnos, O. Grandjean, M. Caboche, H. Hofte, Cellular basis of hypocotyl growth in Arabidopsis thaliana, Plant physiology, 114 (1997) 295-305. [49] T. Hogetsu, H. Shibaoka, M. Shimokoriyama, Involvement of cellulose synthesis in actions of gibberellin and kinetin on cell expansion. 2, 6-Dicholorobenzonitrile as a new cellulose-synthesis inhibitor, Plant and Cell Physiology, 15 (1974) 389-393. [50] C. Xiao, T. Zhang, Y. Zheng, D.J. Cosgrove, C.T. Anderson, Xyloglucan deficiency disrupts microtubule stability and cellulose biosynthesis in Arabidopsis, altering cell growth and morphogenesis, Plant physiology, 170 (2016) 234-249. [51] D.M. Cavalier, O. Lerouxel, L. Neumetzler, K. Yamauchi, A. Reinecke, G. Freshour, O.A. Zabotina, M.G. Hahn, I. Burgert, M. Pauly, Disrupting two Arabidopsis thaliana xylosyltransferase genes results in plants deficient in xyloglucan, a major primary cell wall component, The Plant Cell, 20 (2008) 1519-1537.

289 [52] Y. Verhertbruggen, S.E. Marcus, J. Chen, J.P. Knox, Cell wall pectic arabinans influence the mechanical properties of Arabidopsis thaliana inflorescence stems and their response to mechanical stress, Plant and cell physiology, (2013) pct074. [53] A. Peaucelle, S.A. Braybrook, L. Le Guillou, E. Bron, C. Kuhlemeier, H. Höfte, Pectin-induced changes in cell wall mechanics underlie organ initiation in Arabidopsis, Current biology, 21 (2011) 1720-1726. [54] A. Peaucelle, R. Wightman, H. Höfte, The control of growth symmetry breaking in the Arabidopsis hypocotyl, Current Biology, 25 (2015) 1746-1752. [55] S. Li, L. Bashline, L. Lei, Y. Gu, Cellulose synthesis and its regulation, The Arabidopsis Book, (2014) e0169. [56] R. Zhong, Z.H. Ye, Secondary cell walls: biosynthesis, patterned deposition and transcriptional regulation, Plant Cell Physiol, 56 (2015) 195-214. [57] I. Burgert, P. Fratzl, Actuation systems in plants as prototypes for bioinspired devices, Philos Trans A Math Phys Eng Sci, 367 (2009) 1541-1557. [58] S.R. Turner, C.R. Somerville, Collapsed xylem phenotype of Arabidopsis identifies mutants deficient in cellulose deposition in the secondary cell wall, The plant cell, 9 (1997) 689-701. [59] D.H. Burk, B. Liu, R. Zhong, W.H. Morrison, Z.-H. Ye, A Katanin-like protein regulates normal cell wall gBiosynthesis and cell elongation, The Plant Cell, 13 (2001) 807-827. [60] D.M. Brown, L.A. Zeef, J. Ellis, R. Goodacre, S.R. Turner, Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics, The Plant Cell, 17 (2005) 2281-2295. [61] D.M. Brown, Z. Zhang, E. Stephens, P. Dupree, S.R. Turner, Characterization of IRX10 and IRX10‐like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis, The Plant Journal, 57 (2009) 732-746. [62] Y. Yuan, Q. Teng, R. Zhong, M. Haghighat, E.A. Richardson, Z.-H. Ye, Mutations of Arabidopsis TBL32 and TBL33 Affect Xylan Acetylation and Secondary Wall Deposition, PloS one, 11 (2016). [63] L. Jones, A.R. Ennos, S.R. Turner, Cloning and characterization of irregular xylem4 (irx4): a severely lignin‐deficient mutant of Arabidopsis, The Plant Journal, 26 (2001) 205-216. [64] R. Zhong, W.H. Morrison, G.D. Freshour, M.G. Hahn, Z.-H. Ye, Expression of a mutant form of cellulose synthase AtCesA7 causes dominant negative effect on cellulose biosynthesis, Plant Physiology, 132 (2003) 786-795. [65] K. Sugimoto, R.E. Williamson, G.O. Wasteneys, Wall architecture in the cellulose- deficientrsw1 mutant ofArabidopsis thaliana: Microfibrils but not microtubules lose their transverse alignment before microfibrils become unrecognizable in the mitotic and elongation zones of roots, Protoplasma, 215 (2001) 172-183. [66] D.H. Burk, Z.-H. Ye, Alteration of oriented deposition of cellulose microfibrils by mutation of a katanin-like microtubule-severing protein, The Plant Cell, 14 (2002) 2145- 2160. [67] R. Zhong, M.J. Peña, G.-K. Zhou, C.J. Nairn, A. Wood-Jones, E.A. Richardson, W.H. Morrison, A.G. Darvill, W.S. York, Z.-H. Ye, Arabidopsis fragile fiber8, which encodes a putative glucuronyltransferase, is essential for normal secondary wall synthesis, The Plant Cell, 17 (2005) 3390-3408.

290 [68] M.J. Peña, R. Zhong, G.-K. Zhou, E.A. Richardson, M.A. O'Neill, A.G. Darvill, W.S. York, Z.-H. Ye, Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis, The Plant Cell, 19 (2007) 549-563. [69] C. Lee, R. Zhong, E.A. Richardson, D.S. Himmelsbach, B.T. McPhail, Z.-H. Ye, The PARVUS gene is expressed in cells undergoing secondary wall thickening and is essential for glucuronoxylan biosynthesis, Plant and cell physiology, 48 (2007) 1659- 1672. [70] C. Lee, Q. Teng, R. Zhong, Z.-H. Ye, The four Arabidopsis REDUCED WALL ACETYLATION genes are expressed in secondary wall-containing cells and required for the acetylation of xylan, Plant and Cell Physiology, 52 (2011) 1289-1301. [71] Y. Yuan, Q. Teng, R. Zhong, Z.-H. Ye, The Arabidopsis DUF231 domain- containing protein ESK1 mediates 2-O-and 3-O-acetylation of xylosyl residues in xylan, Plant and Cell Physiology, (2013) pct070. [72] Y. Yuan, Q. Teng, R. Zhong, Z.-H. Ye, Roles of Arabidopsis TBL34 and TBL35 in xylan acetylation and plant growth, Plant Science, 243 (2016) 120-130. [73] Y. Yuan, Q. Teng, R. Zhong, Z.-H. Ye, TBL3 and TBL31, two Arabidopsis DUF231 domain proteins, are required for 3-O-monoacetylation of xylan, Plant and Cell Physiology, (2015) pcv172. [74] M. Busse‐Wicher, T.C. Gomes, T. Tryfona, N. Nikolovski, K. Stott, N.J. Grantham, D.N. Bolam, M.S. Skaf, P. Dupree, The pattern of xylan acetylation suggests xylan may interact with cellulose microfibrils as a twofold helical screw in the secondary plant cell wall of Arabidopsis thaliana, The Plant Journal, 79 (2014) 492-506. [75] J.C. Mortimer, G.P. Miles, D.M. Brown, Z. Zhang, M.P. Segura, T. Weimar, X. Yu, K.A. Seffen, E. Stephens, S.R. Turner, Absence of branches from xylan in Arabidopsis gux mutants reveals potential for simplification of lignocellulosic biomass, Proceedings of the National Academy of Sciences, 107 (2010) 17409-17414. [76] F. Goubet, C.J. Barton, J.C. Mortimer, X. Yu, Z. Zhang, G.P. Miles, J. Richens, A.H. Liepman, K. Seffen, P. Dupree, Cell wall glucomannan in Arabidopsis is synthesised by CSLA glycosyltransferases, and influences the progression of embryogenesis, The Plant Journal, 60 (2009) 527-538. [77] P. McCourt, C. Benning, Arabidopsis: A rich harvest 10 years after completion of the genome sequence, The Plant Journal, 61 (2010) 905-908. [78] T.J. Strabala, C.P. MacMillan, The Arabidopsis wood model—the case for the inflorescence stem, Plant Science, 210 (2013) 193-205. [79] J. Zhang, A. Elo, Y. Helariutta, Arabidopsis as a model for wood formation, Current Opinion in Biotechnology, 22 (2011) 293-299. [80] R.A. Burton, M.J. Gidley, G.B. Fincher, Heterogeneity in the chemistry, structure and function of plant cell walls, Nat Chem Biol, 6 (2010) 724-732. [81] J.C. Thimm, D.J. Burritt, W.A. Ducker, L.D. Melton, Pectins influence microfibril aggregation in celery cell walls: An atomic force microscopy study, Journal of structural biology, 168 (2009) 337-344. [82] G.C. Briggs, K.S. Osmont, C. Shindo, R. Sibout, C.S. Hardtke, Unequal genetic redundancies in Arabidopsis–a neglected phenomenon?, Trends in plant science, 11 (2006) 492-498.

291 [83] M.R. Benatti, B.W. Penning, N.C. Carpita, M.C. McCann, We are good to grow: dynamic integration of cell wall architecture with the machinery of growth, CURRENT CHALLENGES IN PLANT CELL WALLS, (2007). [84] G.J. Seifert, C. Blaukopf, Irritable walls: the plant extracellular matrix and signaling, Plant Physiology, 153 (2010) 467-478. [85] S. Wolf, K. Hématy, H. Höfte, Growth control and cell wall signaling in plants, Annual review of plant biology, 63 (2012) 381-407. [86] T. Engelsdorf, T. Hamann, An update on receptor-like kinase involvement in the maintenance of plant cell wall integrity, Annals of botany, 114 (2014) 1339-1347. [87] P. Sarkar, E. Bosneaga, M. Auer, Plant cell walls throughout evolution: towards a molecular understanding of their design principles, Journal of experimental botany, 60 (2009) 3615-3635. [88] M. Rüggeberg, I. Burgert, T. Speck, Structural and mechanical design of tissue interfaces in the giant reed Arundo donax, Journal of The Royal Society Interface, 7 (2010) 499-506. [89] M. Rüggeberg, T. Speck, O. Paris, C. Lapierre, B. Pollet, G. Koch, I. Burgert, Stiffness gradients in vascular bundles of the palm Washingtonia robusta, Proceedings of the Royal Society of London B: Biological Sciences, 275 (2008) 2221-2229. [90] X. Wang, H. Ren, B. Zhang, B. Fei, I. Burgert, Cell wall structure and formation of maturing fibres of moso bamboo (Phyllostachys pubescens) increase buckling resistance, J R Soc Interface, 9 (2012) 988-996. [91] P.G. Dixon, L.J. Gibson, The structure and mechanics of Moso bamboo material, Journal of The Royal Society Interface, 11 (2014) 20140321. [92] M.K. Habibi, A.T. Samaei, B. Gheshlaghi, J. Lu, Y. Lu, Asymmetric flexural behavior from bamboo’s functionally graded hierarchical structure: Underlying mechanisms, Acta biomaterialia, 16 (2015) 178-186. [93] J.G. Lebkuecher, W.G. Eickmeier, Physiological benefits of stem curling for resurrection plants in the field, Ecology, (1993) 1073-1080. [94] R. Elbaum, Y. Abraham, Insights into the microstructures of hygroscopic movement in plant seed dispersal, Plant Science, 223 (2014) 124-133. [95] J. Dumais, Y. Forterre, “Vegetable Dynamicks”: the role of water in plant movements, Annual Review of Fluid Mechanics, 44 (2012) 453-478. [96] A. Rafsanjani, V. Brulé, T.L. Western, D. Pasini, Hydro-Responsive Curling of the Resurrection Plant Selaginella lepidophylla, Scientific reports, 5 (2015). [97] A. Holstov, B. Bridgens, G. Farmer, Hygromorphic materials for sustainable responsive architecture, Construction and Building Materials, 98 (2015) 570-582. [98] W. Vermerris, A. Abril, Enhancing cellulose utilization for fuels and chemicals by genetic modification of plant cell wall architecture, Current opinion in biotechnology, 32 (2015) 104-112. [99] M.C. McCann, M.S. Buckeridge, N.C. Carpita, Plants and bioenergy, Springer, 2014. [100] C. Baker, M. Sterling, P. Berry, A generalised model of crop lodging, Journal of theoretical biology, 363 (2014) 1-12. [101] S. Huang, R.E. Cerny, D.S. Bhat, S.M. Brown, Cloning of an Arabidopsis Patatin- Like Gene, STURDY, by Activation T-DNA Tagging, Plant Physiology, 125 (2001) 573- 584.

292 [102] D.A. Brummell, Cell wall disassembly in ripening fruit, Functional Plant Biology, 33 (2006) 103-119. [103] A. Cruz-Hernández, O. Paredes-Lopez, Fruit quality: new insights for biotechnology, Critical reviews in food science and nutrition, 52 (2012) 272-289. [104] W.G. Willats, J.P. Knox, J.D. Mikkelsen, Pectin: new insights into an old polymer are starting to gel, Trends in Food Science & Technology, 17 (2006) 97-104. [105] C. Xiao, C.T. Anderson, Roles of pectin in biomass yield and processing for biofuels, Front Plant Sci, 4 (2013) 27. [106] T. Gorshkova, N. Brutch, B. Chabbert, M. Deyholos, T. Hayashi, S. Lev-Yadun, E.J. Mellerowicz, C. Morvan, G. Neutelings, G. Pilate, Plant fiber formation: state of the art, recent and expected progress, and open questions, Critical Reviews in Plant Sciences, 31 (2012) 201-228. [107] T. Gorshkova, O. Gurjanov, P. Mikshina, N. Ibragimova, N. Mokshina, V. Salnikov, M. Ageeva, S. Amenitskii, T. Chernova, S. Chemikosova, Specific type of secondary cell wall formed by plant fibers, Russian Journal of Plant Physiology, 57 (2010) 328-341. [108] C.H. Haigler, L. Betancur, M.R. Stiff, J.R. Tuttle, Cotton fiber: a powerful single- cell model for cell wall and cellulose research, Front Plant Sci, 3 (2012) 104. [109] V.S. Chakravarthy, T.P. Reddy, V.D. Reddy, K.V. Rao, Current status of genetic engineering in cotton (Gossypium hirsutum L): an assessment, Critical reviews in biotechnology, 34 (2014) 144-160. [110] S. Schleicher, J. Lienhard, S. Poppinga, T. Speck, J. Knippers, A methodology for transferring principles of plant movements to elastic systems in architecture, Computer- Aided Design, 60 (2015) 105-117. [111] T.R. Faisal, N. Hristozov, T.L. Western, A.D. Rey, D. Pasini, Computational study of the elastic properties of Rheum rhabarbarum tissues via surrogate models of tissue geometry, Journal of structural biology, 185 (2014) 285-294. [112] P. Fratzl, F.G. Barth, Biomaterial systems for mechanosensing and actuation, Nature, 462 (2009) 442-448. [113] D.J. Robertson, S.L. Smith, D.D. Cook, On measuring the bending strength of septate grass stems, American journal of botany, 102 (2015) 5-11. [114] D.U. Shah, Damage in biocomposites: Stiffness evolution of aligned plant fibre composites during monotonic and cyclic fatigue loading, Composites Part A: Applied Science and Manufacturing, (2015). [115] A.-L. Routier-Kierzkowska, R.S. Smith, Measuring the mechanics of morphogenesis, Current Opinion in Plant Biology, 16 (2013) 25-32. [116] A.-L. Routier-Kierzkowska, A. Weber, P. Kochova, D. Felekis, B.J. Nelson, C. Kuhlemeier, R.S. Smith, Cellular force microscopy for in vivo measurements of plant tissue mechanics, Plant physiology, 158 (2012) 1514-1522. [117] M.A. Graewert, D.I. Svergun, Impact and progress in small and wide angle X-ray scattering (SAXS and WAXS), Current opinion in structural biology, 23 (2013) 748-754. [118] C.E. Blanchet, D.I. Svergun, Small-angle X-ray scattering on biological macromolecules and nanocomposites in solution, Physical Chemistry, 64 (2013). [119] S. Dhondt, H. Vanhaeren, D. Van Loo, V. Cnudde, D. Inzé, Plant structure visualization by high-resolution X-ray computed tomography, Trends in plant science, 15 (2010) 419-422.

293 [120] I. Karahara, D. Yamauchi, K. Uesugi, Y. Mineyuki, Three-dimensional imaging of plant tissues using X-ray micro-computed tomography, PLANT MORPHOLOGY, 27 (2015) 21-26. [121] U.P. Agarwal, Raman imaging to investigate ultrastructure and composition of plant cell walls: distribution of lignin and cellulose in black spruce wood (Picea mariana), Planta, 224 (2006) 1141-1153. [122] M.C. McCann, N.C. Carpita, Designing the deconstruction of plant cell walls, Current opinion in plant biology, 11 (2008) 314-320. [123] M.C. Jarvis, M.C. McCann, Macromolecular biophysics of the plant cell wall: concepts and methodology, Plant Physiology and Biochemistry, 38 (2000) 1-13. [124] M.-A. Ha, D.C. Apperley, M.C. Jarvis, Molecular rigidity in dry and hydrated onion cell walls, Plant Physiology, 115 (1997) 593-598. [125] M.J. Peña, P. Ryden, M. Madson, A.C. Smith, N.C. Carpita, The galactose residues of xyloglucan are essential to maintain mechanical strength of the primary cell walls in Arabidopsis during growth, Plant Physiology, 134 (2004) 443-451. [126] Y.B. Park, D.J. Cosgrove, Changes in cell wall biomechanical properties in the xyloglucan-deficient xxt1/xxt2 mutant of Arabidopsis, Plant physiology, 158 (2012) 465- 475. [127] Z. Kong, M. Ioki, S. Braybrook, S. Li, Z.-H. Ye, Y.-R.J. Lee, T. Hotta, A. Chang, J. Tian, G. Wang, Kinesin-4 functions in vesicular transport on cortical microtubules and regulates cell wall mechanics during cell elongation in plants, Molecular plant, 8 (2015) 1011-1023. [128] L. Köhler, H.-C. Spatz, Micromechanics of plant tissues beyond the linear-elastic range, Planta, 215 (2002) 33-40. [129] A. Bichet, T. Desnos, S. Turner, O. Grandjean, H. Höfte, BOTERO1 is required for normal orientation of cortical microtubules and anisotropic cell expansion in Arabidopsis, The Plant Journal, 25 (2001) 137-148. [130] R. Zhong, D.H. Burk, W.H. Morrison, Z.-H. Ye, A kinesin-like protein is essential for oriented deposition of cellulose microfibrils and cell wall strength, The Plant Cell, 14 (2002) 3101-3117. [131] C. Zhu, A. Ganguly, T.I. Baskin, D.D. McClosky, C.T. Anderson, C. Foster, K.A. Meunier, R. Okamoto, H. Berg, R. Dixit, The fragile Fiber1 kinesin contributes to cortical microtubule-mediated trafficking of cell wall components, Plant physiology, 167 (2015) 780-792. [132] C. Lee, M.A. O’Neill, Y. Tsumuraya, A.G. Darvill, Z.-H. Ye, The irregular xylem9 mutant is deficient in xylan xylosyltransferase activity, Plant and cell physiology, 48 (2007) 1624-1634. [133] C.P. MacMillan, S.D. Mansfield, Z.H. Stachurski, R. Evans, S.G. Southerton, Fasciclin-like arabinogalactan proteins: specialization for stem biomechanics and cell wall architecture in Arabidopsis and Eucalyptus, Plant J, 62 (2010) 689-703. [134] S. Hongo, K. Sato, R. Yokoyama, K. Nishitani, Demethylesterification of the primary wall by PECTIN METHYLESTERASE35 provides mechanical support to the Arabidopsis stem, The Plant Cell, 24 (2012) 2624-2634. [135] M. Kumar, L. Campbell, S. Turner, Secondary cell walls: biosynthesis and manipulation, Journal of experimental botany, (2015) erv533.

294 [136] S. Clark, R. Williams, E. Meyerowitz, Control of shoot and floral meristem size in Arabidopsis by a putative receptor-kinase encoded by the CLAVATA1 gene, Cell, 89 (1997) 575-585. [137] H. Kaya, K.-i. Shibahara, K.-i. Taoka, M. Iwabuchi, B. Stillman, T. Araki, FASCIATA genes for chromatin assembly factor-1 in Arabidopsis maintain the cellular organization of apical meristems, Cell, 104 (2001) 131-142.

295 APPENDIX 2. LIGNIN IDENTITY IN S. LEPIDOPHYLLA CORTICAL STEM

TISSUE

After confirming the presence of lignin in S. lepidophylla stem cortical tissue, the

identity of lignin was further analyzed using Phloroglucinol and Mäule staining following

the protocols in (Mitra & Loqué, 2014). Phloroglucinol detects cinnamaldehyde end-

groups of lignin, including guaiacyl lignin. In contrast, Mäule specifically detects

syringyl lignin, and can be used to differentiate between syringyl lignin (red colour) and

guaiacyl lignin (brown colour) in stained samples (Mitra & Loqué, 2014; Weng et al.,

2010).

The intent of the experiment carried out in this thesis was to stain and image sections from apical, middle and basal regions of inner S. lepidophylla stems to see if

Phloroglucinol and Mäule staining patterns differed among stem regions. Hydrated, unfixed stem segments were embedded in polyethylene glycol (PEG) for sectioning, and

PEG was dissolved from sections prior to staining. Unfortunately, during sectioning, middle and basal regions kept crumbling apart, and images were inconclusive as a result.

Therefore, only transverse sections from the apical region of the stem were imaged and stained with Phloroglucinol (Figure 5.5) and Mäule (Figure 5.6). Syringyl lignin was detected throughout S. lepidophylla apical cortical tissue, with some guaiacyl lignin present in a narrow strip on the abaxial stem side (Figure 5.6). Phloroglucinol detected cinnamaldehyde-containing polymers in xylem, as well as a narrow crescent on the abaxial stem side (Figure 5.5).

296 These results were compared with those from a study by (Weng et al., 2010). This group performed “derivatization followed by reductive cleavage lignin analysis” to identify the types of lignin monomers found in vascular and cortical tissue of mature (i.e., outer) S. moellendorffii stems. S. moellendorffii is a desiccation-sensitive species of the

Selaginella genus that has been studied from an evolutionary perspective for lignin biosynthesis (Weng et al., 2010; Weng, Li, Stout, & Chapple, 2008). S. moellendorffii cortical tissue was found to be predominantly enriched in syringyl lignin, whereas vascular tissue was enriched in guaiacyl lignin. Syringyl lignin provides mechanical support to plants, and is found in angiosperm fiber cells (Weng et al., 2010; Weng et al.,

2008). Thus, enrichment of syringyl lignin in S. lepidophylla cortical tissue most likely contributes to mechanical support of the stem. In addition, though not clearly differentiated by Mäule staining (via colour intensity), different degrees of syringyl lignification most likely also contribute to the stiffness gradients observed at both tissue and cell wall levels in S. lepidophylla stems (discussed in Chapters 3 and 4).

297 APPENDIX FIGURE 5.5.

298 Figure 5.5. Phloroglucinol Staining of S. lepidophylla Apical Cortex. Transverse sections of unfixed samples stained with phloroglucinol to detect cinnamaldehyde end- groups show a thin crescent of staining in abaxial cortical tissue. Closer inspection at higher magnifications reveals staining in the xylem of S. lepidophylla vascular tissue

(adaxial image). Scale bars: 200µm for low magnification (10x), 50µm for higher magnification (40x).

299 FIGURE 5.6.

300 Figure 5.6. Mäule Staining of S. lepidophylla Apical Cortex. Transverse sections of unfixed samples stained with Mäule to detect syringyl lignin show staining throughout adaxial and abaxial cortical tissue in both low and high magnification images. Some guaiacyl lignin, indicated by a brownish colour, is visible in a narrow strip on the abaxial stem side. Scale bars: 200µm for low magnification (10x), 50µm for higher magnification

(40x).

301 APPENDIX 3. RAMAN CONFOCAL SPECTROSCOPY ANALYSIS OF S.

LEPIDOPHYLLA CORTICAL STEM TISSUE

Raman confocal spectroscopy is a powerful technique that combines raman spectroscopy with confocal microscopy to visualize spatial distribution of chemical compounds (Schmidt et al., 2010). This is especially useful for plant cell walls to visualize the location and relative quantity of different cell wall polymers in a noninvasive manner without extensive sample manipulation (e.g., without fixation, embedding, or staining) (Butler et al., 2016; Gierlinger et al., 2012; Richter, Müssig, &

Gierlinger, 2011).

In this thesis, raman confocal spectroscopy was carried out to complement the immunohistochemistry experiments conducted to explore S. lepidophylla cell wall composition. Preliminary raman confocal spectroscopy experiments were performed at

McGill University (Biointerface Lab) and at Université de Montréal (LCM) by myself, and also a single sample was analyzed by technicians at the WiTec Raman Imaging facility in Germany. The intent was to image samples from apical, middle and basal stem regions. However, similar to the problem in Appendix 2, middle and basal sections embedded in PEG kept crumbling apart during sectioning. Therefore, the results presented in Figures 5.7 and 5.8 are only for apical cross-sections. Peak identities for plant cell wall polymers are included in Table 5.3, and peaks with possible compositional identities for adaxial and abaxial cortical cells are recorded in Table 5.4. While plant cell wall polymers were detected, spatial distribution of these polymers and their relative abundance were not analyzed (due to time constraints, equipment limitations and need for better expertise in raman image processing and analysis).

302 FIGURE 5.7.

303 Figure 5.7. Example of Raw Data from Raman Confocal Spectroscopy of S.

lepidophylla Apical Stem Cortex. The raman spectrum presented here is from an adaxial

cortical cell wall of an apical, fresh transverse section. Peak signature (cm-1) is on the x- axis, and relative peak intensity (CCD counts) is on the y-axis. Peaks were identified by the ‘peak finder’ tool in the Bruker image analysis software.

304 FIGURE 5.8.

305 Figure 5.8. Raman Confocal Spectroscopy Analysis of S. lepidophylla Apical Cortex.

The raman spectrum presented here is from an adaxial cortical cell wall of an apical, fresh transverse section. Various cell wall components were detected, including pectin, cellulose, and lignin, and peaks were identified using Table 5.3. The raman spectrum presented here was generated by the WiTec Raman Imaging facility in Germany.

306 Table 5.3. Cell Wall Peaks from Raman Confocal Spectroscopy Literature

Polymer Peaks (cm-1) Reference

330, 350, 380, 381, 437, 459, 520, 900, 969, 1000, 1096, Cellulose 1098, 1120, 1121, 1150, 1151,1295, 1339, 1340, 1380, 1-4 1450, 2890, 2897, 2940, 3264

361, 384, 457, 491, 534, 557, 588, 637, 731, 787, 811, Lignin 895, 928, 975, 1033, 1089, 1177, 1192, 1272, 1298, 1334, (general) 1430, 1508, 1590, 1600, 1621, 1650, 1660, 3075

1162-1168, 1262-1275, 1270, 1370-1405, 1452-1465 G-Lignin 1599, 1592-1616, 1665 1-5

370-399, 781-820, 1138-1160, 1333, 1331-1338, 1454- S-Lignin 1460, 1508-1518, 1586-1609, 1594, 1599

819-864, 1163-1179, 1213-1218, 1286-1299, 1452-1459, H-Lignin 1586-1614

315, 377, 421, 494, 496, 534, 584, 618,829, 900, 981, Hemicellulose 1089, 1124, 1250, 1320, 1380, 1419, 1468, 2898, 2918, 2, 5-6 2935, 2991

340, 372, 441, 486, 537, 621, 686, 710, 750, 775, 795, Pectin 834, 853, 887, 953, 990, 1030, 1050, 1079, 1105, 1145, 6-7 1254, 1330, 1393, 1740, 2941

319, 490, 618, 706, 796, 981, 1003, 1051, 1090, 1131, 1158, 1134, 1158, 1190, 1201, 1248, 1270, 1295, 1356, Pigments 6, 8 1373, 1388, 1452, 1527, 1604, 1630, 2852, 2880, 2907, 2932

(1) (Dieing & Hollricher, 2008) (2) (L. Sun, Simmons, & Singh, 2011) (3) (Agarwal & Ralph, 1997) (4) (Perera, Schmidt, Chiang, Schuck, & Adams, 2012) (5) (Agarwal, 2011) (6) (Agarwal, 2014) (7) (Synytsya, Čopı́ková, Matějka, & Machovič, 2003) (8) (Cai, Zeng, Chen, & Larkum, 2002)

307

Table 5.4. Raman Confocal Spectroscopy of S. lepidophylla Apical Cortex Tissue Region Peaks (cm-1)* Possible Peak Identity**

1036 Lignin 1326 Hemicellulose, pectin 1344 Cellulose 1349 Pigments 1366 Lignin 1370 Lignin, pigments Adaxial 1371 Lignin, pigments 1398 Pectin 1411 Cellulose, hemicellulose 1425 Lignin, hemicellulose 1458 Lignin, pigments, pectin 1523 Pigments 1599 Lignin 1247 Pigments 1253 Hemicellulose, pectin 1340 Cellulose 1345 Cellulose, pigments 1347 Cellulose, pigments 1410 Cellulose, hemicellulose Abaxial 1432 Lignin 1499 Lignin 1592 Lignin 1596 Lignin 1599 Lignin 1604 Lignin, pigments -1 * Control (coverglass and D2O with no tissue sample) peak at 1373cm . ** Possible peak identity based on peak signatures in Table 5.3.

308 APPENDIX 4: WATER MOVEMENT DURING ORGAN DEFORMATION IN

THE RESURRECTION PLANTS SELAGINELLA LEPIDOPHYLLA AND

MYROTHAMNUS FLABELLIFOLIUS

Plants use water for chemical and physiological purposes such as growth and mechanical stability. In addition, some plant species exploit water for precise and controlled movement of mature vegetative tissue (Dumais & Forterre, 2012). Vegetative tissue deformation is implicated in various plant functions, including seed dispersal (e.g.,

Equisetum), predation (e.g., Dionaea muscipula), and predator evasion (e.g., Mimosa pudica) (Forterre et al., 2005; Marmottant et al., 2013; Volkov, Foster, Ashby, et al., 2010).

There are two primary modes of water-driven plant movement. The first is osmotic gradients, in which plants do not gain or lose water but rather shunt it from one tissue to another. This is observed in fast-moving plants such as Dionaea muscipula and Mimosa pudica (Forterre, 2013). The second is differential hygroscopic swelling and shrinking between tissue types as a result of water gain or loss through changes in humidity, or through root water uptake and transpiration. Differential swelling and shrinking results in slower movement, as observed in examples such as wheat awns, pinecones and ice plant seed capsules (Rivka Elbaum & Abraham, 2014).

Plant organ movement via differential (de)hydration can be driven by tissue or cell wall level features. In wheat awns and pinecones, for example, movement is driven at the tissue-level by a bilayer system in which tissues differ in terms of their gross cellulose microfibril angle (Erb et al., 2013). In ice plant seed capsules, movement is driven by keel cell structure (geometry and a cell wall bilayer composition leading to differential

309 swelling/shrinking) (Harrington et al., 2011). Bilayer-driven swelling and shrinking at

different length scales (tissue and cell wall) as a result of geometry and composition has

inspired the creation of a variety of biomimetic and actuating devices for use in sectors such

as aerospace (Bar-Cohen, 2006), medicine (Green et al., 2016), and architecture (Reichert et

al., 2015). A better understanding of inherent plant properties leading to water-driven

deformation is critical for producing biomimetic devices with more complex movements or

with multifunctionality.

The resurrection plant Selaginella lepidophylla is a relatively new model of water-

driven deformation for biomimetic devices (Brulé et al., 2016; Velders et al., 2017; Zhang et al., 2016). S. lepidophylla deforms hierarchically (organ, tissue and cell wall levels) in response to water gain or loss, and deformation is reversible and repeatable. Thus, S.

lepidophylla is an excellent model in which to study the relationship of water-responsive

movement among different length scales. Functional gradients of morphology and

composition at tissue and cell wall levels result in differential swelling and shrinking of S.

lepidophylla stem sides, leading to (un)curling movement (Rafsanjani et al., 2015),

Chapters 3 and 4). While morphological and compositional features contributing to

deformation are understood, little is known about water flow through S. lepidophylla stems

and the relationship between water movement and organ deformation.

In Appendix 4.1. below, I present preliminary work conducted toward examining

water movement during deformation of S. lepidophylla stems. Changes in S. lepidophylla

inner stem water content were tracked over time in vivo using near-infrared light imaging on

a LemnaTec High-Throughput Screen Scanalyzer to correlate water content with organ

deformation. This information will be coupled with three-dimensional analysis of organ and

310 tissue structure in hydrated and dehydrated states (x-ray tomography from CLS) to

understand the relationship between stem deformation pattern/rates and water movement

through stem tissue.

In Appendix 4.2., I present preliminary work done toward examining water-

responsive kinematics of Myrothamnus flabellifolius, another well-studied resurrection

plant that exhibits a more complex sequence of water-responsive organ deformation (Jill M

Farrant, 2000; John P Moore et al., 2007; J. P. Moore et al., 2007; Moore et al., 2006).

Timelapse imaging is used to track changes in M. flabellifolius leaf shape over time during water loss. This information will be coupled with near-infrared image analysis from the

LemnaTec Scanalyzer HTS, as well as three-dimensional analysis of organ and tissue structure in hydrated and dehydrated states (x-ray tomography from CLS) to understand the relationship between leaf deformation pattern/rates and water movement through leaf tissue.

Appendix 4.1. Water Movement through S. lepidophylla Stems

S. lepidophylla plants take approximately 20-24 hours to rehydrate or dehydrate

(Bergtrom et al., 1982; Brighigna et al., 2002; Lebkuecher & Eickmeier, 1993). When

isolated from the plant, stems take approximately six to eight hours (inner type) or one to

two hours (outer type) to rehydrate and dehydrate (Rafsanjani et al., 2015). To observe

changes in water content of S. lepidophylla stems over time, we used a plant phenotyping

system, LemnaTec High Throughput Screen (HTS) Scanalyzer, which is capable of

visualizing changes in plant water content using near-infrared light (Acosta-Gamboa et

al., 2017). Isolated, inner dry stems were rehydrated over six hours, and imaged every 30

311 minutes. Time 0hrs represents the point at which the base of stem samples was placed in water to begin the rehydration process. Near infrared light imaging visualizes vegetative tissue water content on a grey-scale spectrum, and changes in pixel intensity correspond to changes in hydration state (dark is hydrated tissue, and light is less hydrated tissue).

Grey-scale images were converted to a colour scale using ImageJ to better observe differences in water content, where 0 (dark blue) is considered fully hydrated, and 255

(red) is considered dehydrated. Given that the range of colours visible on false-coloured stems (Figure 5.9) does not extend to red, I have designated a relative hydration scale where green to white represents dehydrated tissue, and light blue to dark blue represents tissue in various states of hydration.

Average S. lepidophylla inner stem water content at time 0hrs and 6hrs after the start of rehydration are shown in Figure 5.10. At time 0hrs, most pixels fall between 60-125 on the false colour spectrum, with an average integrated density of approximately 700. This pixel intensity range appears as white, green and light blue on false coloured stem images

(Figure 5.9). At time 6hrs, pixel count shows a leftward shift toward 50 (ranges between

50-115) on the false colour spectrum, with an average integrated density of approximately 900. A larger peak is observed between 60-80 on the false colour spectrum, and has an approximate integrated density of 1300. Stems appear as light blue to dark blue (Figure 5.9).

Based on a preliminary analysis of hydrating S. lepidophylla stems, it appears that the bottom portion of the stem (base to middle region) fills rapidly with water (within the first couple of hours after rehydration begins), as observed by the colour change from light to dark blue that denotes an increase in hydration. This change in water content

312 corresponds with a rapid uprighting of the stem. Following this period of rapid

rehydration is a period of slower rehydration of the upper section of the stem (middle to

tip), which corresponds to slow uncurling of this stem segment. This pattern and rate of

water movement makes sense given the gradient of tissue lignification observed in inner

S. lepidophylla stems (as described in Chapters 2-4), and a corresponding developmental

gradient with respect to living and dead tissue. The base to middle region of the stem is

composed of highly lignified dead or dying tissue. This higher degree of lignification

makes this stem segment more hydrophobic, meaning there is less interaction between

water and tissue in this region, and less water is needed to cause tissue swelling and

subsequent stem movement. The rate of water flow through this region is faster than it is

in the middle to tip stem segment because the cells are dead and the lumen are empty. In

contrast, the stem middle to tip is less lignified and more hydrophilic, and tissue swelling

and subsequent stem movement take longer. The rate of water movement is also slowed

by the fact that cells are alive in this region, and rehydration of the cell must be carefully

controlled so as to avoid rapid water uptake and membrane rupture.

Other factors such as cell and lumen size, as well as cell wall thickness most likely also contribute to the differences in water movement observed in base-to-middle and middle-to-tip stem segments. The contribution of these factors will be made clear when x-ray tomography data gathered from CLS are analyzed, and stem morphology is compared between hydrated and dehydrated states.

Appendix 4.2. Kinematics of Hydrating Myrothamnus flabellifolius Leaves

313 M. flabellifolius leaves were isolated from whole plants for kinematic analysis. In

contrast to S. lepidophylla stems that do not visibly change in terms of area during

dehydration and rehydration, M. flabellifolius leaves drastically change in area. In a

hydrated state, leaves appear fan-like in shape, with sclerenchymatous ribs joined

together by parenchymatous tissue (Figure 5.11). As leaves dry, parenchymatous tissue

shrinks and pulls the sclerenchymatous ribs together. This results in an average leaf

surface area reduction of approximately 32%, a 28% decrease in leaf width, and a 9%

decrease in leaf height when leaves are air-dried for six hours (Table 5.5). The reduction

in surface area is consistent across multiple cycles of dehydration and rehydration, with

only a slight difference evident between the first and subsequent cycles of dehydration

(Table 5.6). A possible cause for this difference is that leaves in the first dehydration

cycle were not left in water before timelapse imaging began; they were picked directly

off hydrated plants and assumed to be at 100% relative water content (RWC), but their

RWC may have been slightly lower, resulting in a slightly reduced starting surface area

as compared to that of subsequent dehydration cycles.

The majority of leaf shape change occurs during the first two hours after dehydration begins. Air-dried leaves reach half of their reduced surface area size

(15.82%) within two hours of drying and slightly over two-thirds of their reduced size

(23.46%) within four hours of drying. Thus, shape change is initiated rapidly in M.

flabellifolius leaves, and the majority of leaf closing occurs during the first two hours

after drying begins. Previous studies examining rehydration and dehydration of M.

flabellifolius have investigated kinematic patterns of the whole plant rather than isolated

leaves (Korte & Porembski, 2012; Sherwin & Farrant, 1996). Thus, the dehydration rates

314 of isolated M. flabellifolius leaves reported here cannot be compared to those reported in the literature since the rates of rehydration/dehydration of the entire plant take more time than single leaves. However, the deformation patterns of leaves can still be compared. A similar percent change in leaf height is observed between the kinematic analyses reported here and those in the literature (13% versus 14% respectively), while percent change in leaf width is slightly higher in the literature (57% versus 24% as reported in Table 5.5), most likely owing to the longer timescale of the experiments conducted in the literature

(5 days of timelapse imaging versus six hours of imaging as conducted in this thesis)

(Korte & Porembski, 2012). Likewise, the pattern of fan-like (un)folding of parenchymatous tissue is consistent with the deformation patterns reported in the literature (Korte & Porembski, 2012; John P Moore et al., 2007; J. P. Moore et al., 2007;

Moore et al., 2006). Visualizing changes in leaf RWC using the HTS Scanalyzer will be useful in comparing water movement during the first two hours of dehydration when leaf surface area rapidly changes and subsequent dehydration over time when surface area reduces more slowly.

315 Figure 5.9.

316 Figure 5.9. Visualization of Inner Stem Rehydration using Near-Infrared Light.

Three representative inner stems are shown hydrating over the course of six hours. Stems are false-coloured to highlight the change in stem hydration over time that was derived from mean grey value of samples imaged with near-infrared light. Dark blue to light blue represents different levels of hydrated tissue, while white to green represents different levels of dehydrated tissue. Stems show a rapid unbending of their basal to middle region during the first couple of hours after rehydration begins. This corresponds to a change in water content (white/light blue to dark blue) in this stem region. The middle to tip stem region rehydrates more slowly, which corresponds to a slower rate of uncurling. Scale bar: 10mm.

317 Figure 5.10.

318 Figure 5.10. Change in Stem Hydration Over Time. Area plots for times 0hrs and 6hrs post rehydration were generated based on pixel count and integrated pixel density derived from false coloured images of samples imaged with near-infrared light. 0-255 reflects the false colour scale, where 0 is considered hydrated (dark blue [DB]), and 255 is considered dehydrated (red [R]). A relative scale (~50 [dark blue]-150 [green]) was selected within this colour spectrum to better represent the hydration state of S. lepidophylla stems. At leftward shift in integrated density is observed between 0hr and

6hr post-hydration time points.

319 Figure 5.11.

320 Figure 5.11. Hydrated and Dehydrated Leaf Conformations of M. flabellifolius. In a hydrated state, leaves appear as an open fan-like shape, with parenchymatous tissue joining sclerenchymatous ribs. As the leaf dehydrates, the parenchymatous tissue shrinks and pulls the sclerenchymatous ribs together. This reduces leaf width and area, along with a moderate reduction in leaf height (visualized in the merged image). Creases in parenchymatous tissue become visible as the leaf dehydrates. Scale bar: 2.5mm.

321 Table 5.5. Kinematic Analysis of Dehydrating Myrothamnus flabellifolius Leaves Surface Area (mm2) Width (mm) Height (mm) 0 hrs PD* 110.43 ± 5.93** 13.31 ± 0.38 13.81 ± 0.24 6 hrs PD 75.76 ± 5.40 10.19 ± 0.42 12.64 ± 0.29 % Decrease (6 hrs PD) 32.28 ± 1.68 23.79 ± 1.50 8.64 ± 0.65 % Decrease (1 hr PD) 9.93 ± 0.78 5.82 0.49 4.03 ± 0.41 % Decrease (2 hrs PD) 15.81 ± 1.06 9.92 ± 0.73 5.72 ± 0.53 % Decrease (4 hrs PD) 23.46 ± 1.34 15.97 ± 1.04 7.13 ± 0.59 * PD: post dehydration (once leaves were removed from water and excess water was blotted off) ** Average and standard error based on N of 30 (10 leaves total, 3 rehydration/dehydration cycles per leaf)

322 Table 5.6. Statistical Analyses* of Shape Change in M. flabellifolius Leaves Across Cycles of Wetting and Drying Change in Dimension (0-6hrs PD)

Surface Area Height Width

0hr + P value 0.032 0.493 0.052 Tukey HSD

Trial 2-1 0.038+ 0.586 0.065 Trial 3-1 0.092 0.531 0.116 Trial 3-2 0.907 0.996 0.956

1hr

P value 0.816 0.861 0.812

Tukey HSD

Trial 2-1 0.839 0.987 0.804 Trial 3-1 0.853 0.856 0.901 Trial 3-2 1.000 0.925 0.979

2hrs

P value 0.317 0.763 0.389

Tukey HSD

Trial 2-1 0.309 0.813 0.383 Trial 3-1 0.523 0.788 0.582 Trial 3-2 0.917 0.999 0.937

4hrs

P value 0.154 0.698 0.282 Tukey HSD

Trial 2-1 0.159 0.779 0.306

Trial 3-1 0.305 0.713 0.408

Trial 3-2 0.919 0.993 0.977 * ANOVA, p < 0.05, N = 30 (10 leaves, 3 cycles each); differences among trials were examined using a Tukey HSD post-hoc test + Denotes significant values (p< 0.05)

323