An In-Depth Biochemical Analysis of Spider and Silkworm Silk

Unravelling the secrets of silk: an in-depth biochemical analysis of spider and silkworm silk

Hamish Cameron Craig

A thesis in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Biological, Earth and Environmental Sciences Evolution and Ecology Research Centre

UNSW

February 2019

THE UNIVERSITY OF NEW SOUTH WALES

Thesis/Dissertation Sheet Surname or Family name: Craig First name: Hamish Other name/s: Cameron Abbreviation for degree as given in the University calendar: PhD School: School of Biological, Earth and Environmental Sciences Faculty: Faculty of Science Title: Unravelling the secrets of silk: a detailed examination of silk biology and structure Abstract:

Silk is a protein-based biopolymer produced by many different invertebrate species from amphipods to spiders. Its incredible material properties, biocompatibility and antimicrobial properties make it one of the most desirable natural fibres in the race for new materials, with major potential impacts in everything from biomedical research to its aerospace applications. Although silk has been studied in detail since the latter part of the 20th century the field is still unable to produce truly comparable synthetics due to the complexity of biological factors involved in influencing silks properties. The major focus of this thesis is examining biological and structural factors that impact silk properties within spiders and silkworms. To examine this, I analysed silk across many scales from phylogenetic trends in amino acid composition and material properties, down to the Nano-scale examining the impacts of molecular structure, pioneering new methods of silk analysis through utilisation of dynamic nuclear polarization (DNP) solid-state nuclear magnetic resonance (ssNMR) spectroscopy. Using a comparative meta-analysis, I found that within spiders MaSp composition is a major influencing factor broadly across phylogeny and that glycine and serine concentration have a more influential impact than previously thought, whilst confirming the importance of proline in influencing silks elastic properties. I explore the inconsistency and spread of isotopically labelled alanine within spider’s silk and further our understanding of metabolic pathways into silk proteins and silk structure. I then provide and examine the efficacy of an alternative to isotopic labelling, DNP ssNMR, which reveals unprecedented detail into silks molecular structure, boasting a > 50 fold signal enhancement allowing determination of the structural role of lower abundance amino acids like arginine and revealing the presence of the first known example of a post-translational modified amino acid hydroxyproline within silk. lastly, I explore the major implications of hitherto undocumented voids found within native silkworm silks, the molecular structural influences that help create them, their impact on mechanical property measurements and potential development and utilisation ecologically improving the insulative properties of their cocoons. Ultimately, whilst revealing invaluable new information and methods for the silk field to help it towards its goal of producing comparable bio-mimetics, this thesis highlights the need for a biologically minded and holistic study of biomaterials such as silk to understand all the factors that help organisms achieve such incredible and complex materials.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

Signature Witness Signature Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research. FOR OFFICE USE ONLY Date of completion of requirements for Award:

Originality Statement

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed

Date

Authenticity Statement

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed

Date

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted, I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed

Date

INCLUSION OF PUBLICATIONS STATEMENT

UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure.

Publications can be used in their thesis in lieu of a Chapter if: • The student contributed greater than 50% of the content in the publication and is the “primary author”, ie. the student was responsible primarily for the planning, execution and preparation of the work for publication • The student has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator. • The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis

Please indicate whether this thesis contains published material or not.

This thesis contains no publications, either published or submitted for publication (if this ☐ box is checked, you may delete all the material on page 2)

Some of the work described in this thesis has been published and it has been documented in the relevant Chapters with acknowledgement (if this box is checked, you

☒ may delete all the material on page 2)

This thesis has publications (either published or submitted for publication) incorporated ☐ into it in lieu of a chapter and the details are presented below

CANDIDATE’S DECLARATION I declare that: • I have complied with the Thesis Examination Procedure • where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis. Name Signature Date (dd/mm/yy) Hamish Craig

Postgraduate Coordinator’s Declaration (to be filled in where publications are used in lieu of Chapters) I declare that: • the information below is accurate • where listed publication(s) have been used in lieu of Chapter(s), their use complies with the Thesis Examination Procedure • the minimum requirements for the format of the thesis have been met. PGC’s Name PGC’s Signature Date (dd/mm/yy)

Summary of Collaborators’ Contributions

In all chapters and manuscripts submitted, and in preparation in which I was listed as first author, I was the primary investigator. This included being the main researcher in data collection, analysis and interpretation, and preparation of the manuscript. Michael Kasumovic and Sean Blamires were my primary supervisors, Aditya Rawal was my main advisor in the training, collection and analysis of NMR involved in this manuscript. Shinichi Nakagawa and Dakota Piorkowski were my main collaborators in the first chapter, Shinichi providing advice and expertise in conducting a meta-analysis and Dakota assisting with the back search and editing the manuscript. James Hook and Marc-Antoine Sani contributed to the access, design, and assisted in data collection of DNP NMR experiments as well as editing the manuscript in Chapter 3. Rangam Rajkhowa provided access to silkworm silk samples and assisted in mechanical property measurements in chapter 4. Yin Yao and Nicholas Ariotti provided training and assistance in the collection and processing of AFM and electron tomography data within chapter 4.

Acknowledgements

In the words of the late Stephen Hawking “Science is not only a discipline of reason but, also, one of romance and passion”. This is very true of my experience throughout my PhD and I am extremely grateful to be in a position to be able to purse my passions and thirst for gaining and sharing knowledge. Much like a romantic relationship my PhD has had its ups and downs, its hardships and its moments of splendour, but ultimately I’ve emerge a different and more experienced and developed person. The great irony of pursuits such as these however is the increasing awareness of how little you actually know. I find this as humbling and inspiring as it is frustrating, when one’s goals are a perpetually moving bench mark advancing further into the unknown. I hope that in reading my thesis I can impart at least some the knowledge I spent a better part of 3.5 years passionately gathering to share with the field.

I consider myself blessed to have found two different but ultimately complimentary supervisors for my PhD, Michael Kasumovic and Sean Blamires. Without their combined effort I wouldn’t be in the position I am today or to be able to present such interesting and varied work. Firstly, Mike, I would like to express my tremendous gratitude for all the help and support you have given me throughout my PhD, not only academically but as a friend. You were always happy to let me interrupt whatever you were doing to be a sound board for my frustrations, troubles, thoughts and feelings. Always in my corner, willing to help, and a perpetually positive influence across this past 3.5 years of my life and for that I thank you and am truly grateful. Sean without you I wouldn’t be the scientist I am today, without your knowledge, diligence and connections my thesis and PhD would definitely not be the same. I’ve found myself to only have increasing respect and admiration for both you and your input over the course of my thesis. Thank you for all the effort and time you put in helping me make this a reality, you were an essential source of assistance and guidance that I would not have been able to complete this project without.

I would also like to give a huge thank you to my co-supervisor Aditya Rawal, without all your help I would not have the knowledge, training and understanding of NMR spectroscopy and analysis that is at the core of my entire thesis. Thank you for taking so much time out of your days to help me when I was only coming by to ask 1 “quick” question, being my constant sound board for any questions or ideas I had about the chemistry in my thesis.

I would also like to give a big thankyou to all my collaborators Shinichi Nakagawa, James Hook, Marc Sani, Yin Yao, Nicholas Ariotti and Rangam Rajkhowa for all your help, support, expertise and training that was necessary to make such an interdisciplinary thesis possible.

Day to day I am extremely lucky to have been surrounded by an amazing group of people within the EERC. I would especially like to thank Teagan Gale, Emma Asbridge for taking me in early on in my PhD and being my constant companions with me through thick and thin. I would also like to thank Justin Chan who along with Teagan and Emma became my lunch time escape into a vast array of amazing conversations and debates that I will forever cherish as part of my time doing this thesis. Thank you all for making my PhD a fun and fulfilling experience, as we all move our separate ways I hope we keep in contact and share some great banter again over lunch sometime.

I’d like to thank my chosen family; my friends who potentially unbeknownst to them have helped me develop and grow as a person more so than anyone over this time. These amazing boys took me in as a very reserved, unhappy and closeted person and helped me flourish into the man I am today. Aaron, Andrew, Brenton, Chris, Jesse, Josh, Robert, Simon and Damien thank you for all the love and support you have given me through the ups and downs over my time doing this research.

Lastly, I’d like to thank my family, Mum and Dad, Nick and Kate thank you for your love and support over this time. I understand it was hard to living so far apart but I always felt your love and pride in me for following my passions. Mum and Dad I’d like to thank you especially for all the effort you put in getting me here in the first place, school, tutors, college, rent , basically everything; without you I definitely wouldn’t be the person I am today or be standing where I’m standing and for that I will be eternally grateful.

Publications and Presentations

Publications

Craig, H.C., Piorkowski, D., Blamires, S. J., Nakagawa, S., Kasumovic, M.M (in prep). A phylogenetic comparative analysis of amino acid composition and spider silk mechanical properties

Craig, H.C., Kasumovic, M.M., Blamires, S., Rawal, A. (in prep) Evidence of a protein allocation trade-off and macronutrient partitioning in the production of protein-based extended phenotypes

Craig, H.C., Blamires, S., Sani, M., Kasumovic, M.M., Rawal, A., Hook, J.M. (Submitted). DNP NMR spectroscopy reveals new structures, residues and interactions in wild spider silk

Craig, H.C., Kasumovic, M.M., Rawal, A., Rajkhowa, R., Blamires, S.J The origin, function and implications of biologically induced cavitation in wild silkworm silk

Presentations

Craig, H.C. (2016) A phylogenetic comparative analysis of amino acid composition and spider silk mechanical properties. Paper presented at the International Congress of Arachnology, Golden, Colorado, USA

Abstract

Silk is a protein-based biopolymer produced by many different invertebrate species from amphipods to spiders. Its incredible material properties, biocompatibility and antimicrobial properties make it one of the most desirable natural fibres in the race for new materials. Although silk has been studied since the late 20th century, the field is still unable to produce truly comparable synthetics due to the complexity of biological factors involved in influencing silk properties. The major focus of my thesis is thus to examine the biological and structural factors that impact silk properties within spiders and silkworms. To examine this, I analysed silk, specifically Major ampullate spidroin (MaSp) protein and the silk of silk moths, across many scales from phylogenetic trends in amino acid composition and material properties, down to the Nanoscale, examining the impacts of molecular structure. In Chapter 2, I used a comparative meta- analysis to look for correlations between the major amino acids within silk and silks material properties across phylogeny. I found that MaSp composition is a major influencing factor broadly across the spider phylogeny. Additionally, I found that glycine and serine percentages have a larger impact than previously thought, whilst I confirmed the importance of proline in improving silks properties. In Chapter 3, I explored how isotopically labelled alanine moves through a spider’s metabolic pathways into their silk, furthering our understanding of spider’s metabolic response to dietary fluctuations in protein, the pathways involved in spidroin production and silk structure. In Chapter 4, I provided and examined the efficacy of an alternative to isotopic labelling, DNP ssNMR, which reveals unprecedented detail into silk’s molecular structure. This allowed the determination of the structural role of lower abundance amino acids like arginine and revealed the presence of the first known example of a post-translational modified amino acid, hydroxyproline, within spider silk. Lastly, in Chapter 5, I explored the major implications of voids found within native silkworm silks, the molecular structural influences that help create them, and their impact on mechanical property measurements. I also discuss the role these voids have in ecologically improving the insulative properties of silkworm cocoons. Ultimately, whilst revealing invaluable new information and methods for the silk field to help it towards its goal of producing comparable biomimetics, my thesis highlights the need for a biologically minded and holistic study of biomaterials such as silk, to understand all the factors that help organisms achieve such incredible and complex materials.

Table of Contents

Thesis/Dissertation Sheet ...... i Originality Statement ...... ii Authenticity Statement ...... iii Copyright Statement ...... iii Summary of Collaborators’ Contributions ...... iv Acknowledgements ...... v Publications and Presentations ...... vii Abstract ...... viii Chapter 1 ...... 1 Silk: the ultimate biological material ...... 1 Chapter 2 ...... 6 A phylogenetic comparative analysis of amino acid composition and spider silk mechanical properties ... 6 Abstract ...... 6 Introduction ...... 7 Methods ...... 11 Data collection ...... 11 Phylogeny and Analysis ...... 11 Results ...... 12 Discussion...... 12 Conclusion ...... 18 Chapter 3 ...... 19 Elucidation of metabolic pathways, amino acid synthesis and macro-nutrient partitioning in the production of spider’s silk as an extended phenotype ...... 19 Abstract ...... 19 Introduction ...... 20 Methods ...... 22 Results ...... 24 Discussion ...... 27 Utilisation of a new metabolic pathway suggests a protein allocation trade-off ...... 29

Starvation and macronutrient partioning ...... 31 Conclusion ...... 32 Chapter 4 ...... 33 DNP NMR spectroscopy reveals new structures, residues and interactions in wild spider silk ...... 33 Abstract ...... 33 Introduction ...... 33 Methods ...... 35 Materials ...... 35 Sample Preparation ...... 35 Dynamic Nuclear Polarization Enhanced 13C and 15N Cross Polarization MAS (DNP CP MAS) Solid- State NMR Spectroscopy ...... 36 Results and Discussion ...... 37 DNP Enhancement ...... 37 Variation in crystallinity ...... 43 Hydroxyproline ...... 44 Structural role of arginine ...... 45 Glue glycoprotein ...... 46 Conclusion ...... 46 Chapter 5 ...... 47 The origin, function and implications of biologically induced cavitation in wild silkworm silk ...... 47 Abstract ...... 47 Introduction ...... 48 Methods ...... 50 Silk Degumming ...... 50 FTIR Spectroscopy ...... 50 Solid-State NMR (ssNMR) Spectroscopy ...... 50 Atomic force microscopy ...... 51 Pore fraction analysis ...... 51 Quasistatic Nano-indentation ...... 52 Electron Tomography ...... 53 Heat flow simulations ...... 53 Results ...... 53

1D 13C CPMAS NMR and FTIR Spectroscopy ...... 53 2D 13C–1H HETCOR NMR ...... 54 Atomic Force Microscopy and Pore fraction analysis ...... 57 Nano-indentation ...... 58 Electron Tomography ...... 59 2D Heat flow simulations ...... 60 Discussion...... 61 Silk fibre chemistry (CPMAS and FTIR) ...... 61 Silk fibre chemistry (1H-13C heteronuclear correlations)...... 61 Characterisation of cavitation ...... 63 Origin of cavitation in silkworm silk ...... 64 Effect of Cavities on silk material properties ...... 64 Possible biological role for cavitation ...... 65 Conclusion ...... 66 Chapter 6 ...... 67 Conclusions ...... 67 References ...... 67 Appendices ...... 78 Phylogenetic comparative analysis search terms ...... 78 List of references used in phylogenetic comparative analysis...... 79

Chapter 1

Silk: the ultimate biological material

The study of biological materials sits at the interface of a vast array of disciplines, from the biological basis of their production, to the nuances in their chemical structure, and the resultant mechanics of the solid material1-2. The interdisciplinary nature of this field makes the study of biological materials as interesting as it is complex. Limpet teeth, abalone nacre, mussel byssus and invertebrate silks are just a handful of examples of such materials; the unique and impressive properties of which ultimately showcase the incredible propensity of natural selection to create and refine materials in nature3-5.

Silks are one of the best know and versatile examples of these materials and have independently evolved in many invertebrate groups from marine amphipods to spiders5-9. Although there appears to be homology between these silks as a protein it important to note that the glands that produce them have evolved independently on different segments between each major silk producing group (i.e. anteriorly in lepidoptera, opisthosomal limbs in spiders, etc)10. Silks are utilised by invertebrates for a variety of reasons, as protection (egg cases and during metamorphosis), supportive elements in structures (webs, retreats and hives), and in prey capture (glow worms and spiders)11-12. Spiders and silkworms are the best- known examples for silk production, spiders for the variety and unrivalled material properties of their silk, and silkworms for the sheer volume of quality material they produce and its associated commercial value.

Silk itself is a protein-based nanocomposite biopolymer, meaning it’s formed from chains of amino acids that fold into disparate nanostructures (crystalline and amorphous) which come together in a unique way to form the overall fibre2, 13. Synthesised in specialised silk glands, the silk proteins (fibroins in silkworms and spidroins in spiders) are stored in the ampulla of the gland as a highly concentrated solution prior to its transition to a solid fibre during spinning14-16. Silkworms, unlike spiders, can only produce one type of silk from a pair of silk glands that take up a large portion of their body ending at a pair of spinnerets in their mouth14. Spiders on the other hand can produce up to 8 different types of silk, each with a unique function, and so possess up to 8 pairs of silk glands in their abdomen15. Each type of silk (silkworm and

the eight-spider silks) is distinct in both their chemical composition and the resultant structural and mechanical properties17-19.

Of the eight types of silk produced by spiders, their wrapping silks, cylindriform and aciniform silk20, are most analogous to silkworms’ silk in terms of biological function, however their material properties can vary significantly. Spiders also possess flagelliform silk that generally form capture threads, aggregate (wet) and cribellate (dry) silk which form the sticky threads used in prey capture and pyriform silk that is used to attach their silk to surfaces19, 21-22. The seventh and eighth kind of silk are found in spiders dragline silk, which is made up of major and minor ampullate silk (MA and Mi silk, respectively). These silks, particularly MA silk, are generally of most interest as a result of their incredible combination of strength and extensibility resulting in unrivalled toughness23. MA silk makes up the major supportive elements of a spider’s web (i.e. frame silks and radial silks in orb webs) and is generally significantly stronger than Mi silk but lower in extensibility24. As a result of these properties, MA silk is the major focus of much of this thesis.

There are numerous factors that are known to affect the distinct structural and mechanical properties of MA silk. These range from genetic and amino acid sequences making up the silk proteins, which affect how these proteins fold and aggregate during spinning, to the multitude of glandular mechanisms involved in the formation of the fibre1, 25-26. These interactive factors can thus play a significant role in the various properties we can observe in silk. Moreover, as a result of these factors, spider silk is not only distinct from silkworm silk, but varies between closely related species, between the same species, and can even vary significantly on an individual level27-28.

Despite variation in the factors mentioned above, there still remain common bio-chemical threads that make spider’s major ampullate spidroins and silkworm silk fibroins similar. Both silks are made of multiple proteins - Spiders MA silk is made up of two proteins, major ampullate 1 and 2 (MaSp1 and 2, respectively)29-30, silkworm silk is made up of heavy and light chain fibroins as well as p25 fibroin (which assists in linking the heavy and light fibroins)31. Both MaSp’s and the heavy chain fibroin consist of highly conserved amino and carboxyl (denoted as N- and C-) terminal domains within which are found at either end an extensive repetitive regions18, 32. This repetitive region consists of repeated amino acid motifs that are responsible for the formation of the nanostructures within silk. Spider silk’s repetitive region is made up of four major motifs that form two distinct structural regions: the so called crystalline and amorphous

regions. The glycine rich amorphous region is formed from a GGX motif found in both MaSp1 and 2 and a GPGXX motif found only in MaSp2 (where A=alanine, G=glycine, P=proline and X can be Q=glutamine or

Y=tyrosine or L=leucine. The crystalline region is formed from a poly-alanine (A)n motif found in both

MaSp1 and 2 and poly-alanine-glycine (GA)n motif generally only found within MaSp1. Silkworm silk on the other hand is formed from either an extensive region of (GAGAGS)n that forms crystalline structures and a combination of glycine rich (GA)n and (GX)n that form the amorphous regions like in the domesticated species Bombyx mori 17, 33, or has a similar structure to spider silk with distinct glycine rich amorphous region and poly-alanine (A)n crystalline region seen in many wild silkworm species (Saturniidae)17, 23.

Molecular computational models suggest that crystalline and amorphous regions together play a significant role in the material properties and deformation characteristics of silk2, 34-35. Entropic unfolding of the amorphous region, and strain hardening within the crystalline region are thought to work together to give silk its amazing capacity to resists tension whilst plastically deforming2, 34, 36. Significant variation within the molecular structures that form these regions and how they interact along the fibre axis play a crucial role in determining the properties in a given silk37. Understanding this structure-function relationship as a result has been at the forefront of interest within the field for over a decade2, 23, 38. Uncovering the structures formed by each repetitive motif and how they influence their respective structural domain has shed significant light on observed variation in silk between species.

The (A)n and/or (GA)n motifs are common across all silks, the dense hydrogen bonding networks formed by stacking these poly-alanine and/or poly-alanine-glycine runs into beta-sheet crystals is imperative for strain hardening. Strain hardening involves the repeated dislocation of hydrogen bonds within the nanocrystallites, as more dislocations occur, a resistance to the nucleation of further dislocations develops17, 39-40. This increasing resistance to strain is what ultimately results in silk’s high toughness. Moreover, it therefore follows that the major structural factors observed to affect this region are the density and alignment of these nanocrystallites along the fibre axis36, 41-43, i.e. the number of hydrogen bonds that need to be repeatedly dislocated and their orientation to the direction of the applied stress.

The amorphous region is arguably where we find the most structural diversity across the different silks. It can be formed by many motifs, from the (GA)n motif in the heavy chain of B. mori to the more intricate GPGXX within spiders MaSp244. The amorphous region is distinct from the crystalline region due to its

relatively high level of structural disorder, however it is just as dependant on the hydrogen bonding networks forming its structures35. The entropic unfolding occurring in the amorphous region involves a gradual break down of the hydrogen bonds that make up the random, turn, and helical structures formed by the aforementioned motifs35. Silks with higher elasticity generally contain more sophisticated structures that help maximise the hydrogen bonds per unit area within its amorphous region thereby increasing the overall entropy of the silk. Simple motifs like GGX seen in the MaSp1 can form relatively simple random coil or α-helical structures; whereas more derived motifs like GPGXX within MaSp2 adds type-1 and -2 β-turns that allow the formation of more complex β-turn spirals that act like nanosprings and contain a significantly higher number of hydrogen bonds44-46.

Despite the developing understanding of the structure-function relationship within silk, our ability to recreate silk artificially has remained somewhat of a pipe dream due to the largely singular focus on fibre chemistry. The complexity of studying materials like silk in a biological system, unlike the relatively controlled environments used to understand simple inorganic substances, means that silks properties - both structural and mechanical - are confounded and ultimately driven by the complex biological mechanisms responsible for its formation26, 47. There are many examples of how ecological factors can influence silk properties through post-secretion processes, for example spiders tailoring the properties of the silk forming their webs to the given needs or limitations of their environment such as prey availability or excessive wind41, 48. However, the biological factors that control silk properties are still relatively unexplored. Some aspects are beginning to become resolved, like differences that can be generated by varying the sheer forces along the bends and tapering portion of the silk gland25, 28, 49. By varying the speed at which silk is pulled from the silk gland researchers can influence the order and alignment of the nanocrystallites along the fibre axis which results in variation in the mechanical properties of the silk25. Another well-established biological mechanism influencing silk properties lies in the capacity of spiders to vary the expression of the proteins forming their silk, for example changing the MaSp2 expressed in spider silk results in a significant increase in silk’s elasticity50-51.

Beyond these examples however we find spinning induced variation within silk properties that cannot be readily explained. For example, the significant individual variation observed in the mechanical properties of a single silkworm silk fibre; exposure to wind inducing the mechanical performance of MA silk to vary independently of MaSp expression41 or diet inducing changes in MaSp expression without necessarily

affecting silk’s mechanical properties50. Understanding silk holistically and being able to account for these sources of variation is paramount for the field to achieve its ultimate goal of creating a viable synthetic analogue to silk52. The ultimate purpose of this thesis is thus to explore biological impacts on silk structural and material properties and to emphasise the need for a whole-organismal understanding of silk from how it is influenced by the environment as an extended phenotype down to fine scale nuances in the protein nanostructures.

In Chapter 2, I first look for correlations between the major amino acids within silk and silk’s material properties broadly across the spider phylogeny using a phylogenetic comparative analysis. This will help understand how different species vary in their protein expression and the relationship this has on silk properties. In Chapter 3, I will use different techniques to introduced labelled alanine into a spider’s diet to explore how isotopically labelled alanine moves through a spider’s metabolic pathways and into their silk. This will further our understanding of the metabolic response spiders have to dietary fluctuations in protein, the pathways involved in spidroin production, and silk structure. In Chapter 4, I will explore the efficacy of an alternative to isotopic labelling – dynamic nuclear polarization (DNP) solid state nuclear magnetic resonance spectroscopy (ssNMR). This technique is successfully used in many other studies of biological materials from viruses to collagen53-54. I will examine if this technique has the same potential to explore the same molecular details and detect important structural variation between three wild silks of Argiope keyserlingi, Latrodectus hasselti and Nephila plumipes. Finally, in Chapter 5, I will explore the major implications of voids found within wild silks of Antheraea assemensis and Samia cythia ricini and compare them to the domesticated Bombyx mori. This comparison will allow me to explore the molecular and structural influences that help create the voids and their impact on the measurements used to characterise the material properties of their silk. This will help to understand the significant intraspecific and individual variation that can be observed in the tensile properties of silkworm silk and provide a better understanding of the biological influences that are generally overlooked when exploring silkworm silk as a material.

Chapter 2

A phylogenetic comparative analysis of amino acid composition and spider silk mechanical properties

Craig, H.C., Piorkowski, D., Blamires, S. J., Nakagawa, S., Kasumovic, M.M

Abstract

The Major Ampullate (MA) silk of spiders possesses a unique combination of extensibility and strength by weight. This protein-based biopolymer outperforms all synthetic equivalents. It is thus no surprise that there is interest in understanding the underlying reason for MA silk’s unique performance across science and engineering disciplines. Considerable efforts have been made toward piecing together the hierarchical structure of MA silk to determine if the material properties could be synthetically replicated. A key part of this is the role that specific amino acid sequences play in inducing specific mechanical outputs. Here I performed a phylogenetic comparative analysis to examine the relationship between MA silk’s most predominant amino acids and its material properties across a spider phylogeny. I found that the presence of high glycine and proline, via differential spidroin expressions, had the biggest impact on material properties across the phylogeny. The binomial expression of proline adds weight to our hypothesis that spidroin expression explains silk material properties. Serine significantly influences silks toughness potentially as a result of its role and relative abundance in the C and N terminus domains. This study emphasizes the importance of the relative ratio of key amino acids in predicting material properties in silks.

Introduction

Biological materials science sits at the interface of biology, chemistry, and material science making it both an extremely exciting forefront for research but also extremely challenging. Navigating this field requires a holistic understanding of any given system from subtle changes in genetic expression to the resultant changes in the physical mechanics of the material.

The best example of this is the major ampullate (MA) silk of spiders, which makes up the major supportive elements of their webs and holds the title of world’s toughest natural material55. The complex hierarchical protein structure of spider MA silk gives it a unique combination of extensibility and strength that results in incredible toughness which can even outperform Kevlar56. These impressive mechanical properties make spider MA silk a highly sought-after material for a multitude of purposes, from bulletproof clothing to medical scaffolds for cellular engineering19, 57. Unfortunately, the complexity of biological factors influencing the chemistry and resultant mechanical properties of silk fibres make it a veritable nightmare for chemical analysis compared to the relatively controlled systems found in other fields of chemistry.

As for the mechanics, there are many ways to quantify the mechanical properties of silk; however, the most common method is quasistatic tensile testing which involves stretching the fibre under high tension along its axis until it breaks. Such tests enable the production of a stress-strain curve (Figure 1a) from which one may extract information such as the fibre’s Young’s modulus (also known as, elastic modulus or stiffness), ultimate stress (a.k.a ultimate strength or breaking stress) and ultimate strain (a.k.a breaking strain) each of which are used to calculate the silk’s toughness. Young’s modulus refers to the elastic deformation of the fibre and is defined as the capacity of the silk to stretch before permanent deformation occurs and is therefore calculated from the initial slope of the stress-strain curve (Figure 1a). Ultimate Stress or the rigidity/strength of the silk corresponds to the capacity of the silk to resist the force applied as it is being pulled. Ultimate Strain or extensibility is the percentage capacity of silk to be stretched relative to its original size before breaking. Toughness is the combination of stress and strain or how strong the silk is as a factor of how much it can stretch and calculated by working out the area underneath the stress-strain curve (figure 1a).

Like other insect and spider silks, MA silk is a protein-based nanocomposite spun from a liquid dope precursor that is secreted and stored within the ampulla of the major ampullate glands 58-59. The silk itself is comprised of a fibrillar core coated in a thin skin that may be easily removed by chemical

treatment. High-resolution imaging techniques such as Atomic force microscopy (AFM) reveal that the 1-5 μm MA silk fibres are made up of smaller nano-scale fibrils oriented along the fibre’s axis49, 60. The fibrillar substructure is comprised of hard nanocrystalline regions, thought to impart the silk’s strength, suspended within a rubbery amorphous matrix thought to be responsible for silk’s elasticity and extensibility. These regions have come to conventionally be known as the crystalline and amorphous regions respectively 61-65.

Two protein monomer classes are thought to be found in MA silk, Major Ampullate Spidroin 1 & 2 (MaSp1 and MaSp2), with evidence of repeated recombination’s of these proteins among spiders29, 66. The proteins are comprised of highly conserved amino (N) and carboxyl (C) terminal regions, between which highly repetitive regions, make up the bulk (~95%) of the protein67-69. The majority of the repetitive region is made up of the short repetitive amino acid motifs GGX, (A)n and (AG)n within

MaSp1 and the addition of a GPGXX and loss of (AG)n within MaSp2 (Figure 1b, where A= alanine, G=glycine, P=proline and X can be Q=glutamine or Y=tyrosine or L=leucine)70-71. GGX and GPGXX form helical and nano-spring structures (Figure 1c) within the amorphous region responsible for the silk’s

45-46 elasticity . (A)n and (AG)n form tightly packed β-pleated sheets that form nanocrystallites. The extensive networks of dense hydrogen bonding within the nanocrystallites impart MA silk with its strength (Figure 1c) 72-73.

It is thought that MaSp1 was the first protein to appear in MA Silk, with MaSp2 evolving relatively recently in the derived orb weavers, although there is some suggestion for “MaSp2-like” proteins found Mygalomorphs, the best genomic analysis indicates that these Mygalomorph proteins actually more closely resemble egg case proteins 74-75. The development of MaSp2 in the orb weavers has had a significant effect on the elastic properties of MA silk and has enabled the development of two- dimensional orb webs. By including proline within MaSp2 beta-turns, spirals are formed within the amorphous region (Figure 1c) which facilitate silk extensibility46, 76. It may indeed be the primary reason that the two-dimensional orb-web has become the primary method of prey capture used among araneids (Figure 2)59, 77.

Although we now have a rudimentary understanding of the structure-function relationship of MA silk2, 78, many external factors can influence its properties. Some of these factors are well documented, such as the influence of reeling speed (or how fast the silk is pulled from the spinneret) on the strength of silk through friction induced alignment of the nanocrystallites along the fibre axis1, 25; or the impact of water concentration within the silk gland79. Nevertheless, the influences acting on silk during

spinning are not well understood. For example exposure to wind induces the mechanical performance of MA silk to vary independently of MaSp expression41. Moreover, diet induces changes in MaSp expression without necessarily affecting mechanical properties80. These findings suggest that other factors impact the organization of the amorphous region, which could have a greater impact on mechanical performance than the MaSp1 or MaSp2 composition, or amino acid motifs80.

Here I performed a comparative analysis to ascertain the relationship between the amino acids alanine, glycine, proline glutamine and serine (the five most abundant amino acids in MA silk,70) and the silk’s material properties (examining specifically Young’s modulus, ultimate stress, strain and toughness), broadly across a spider phylogeny. Were I to find significant correlations between any given amino acid(s) and particular material properties across the phylogeny, we may conclude that the amino acid(s) in question play a significant role in forming specific nanostructures and that these structures impact silk performance. The lack of any such correlations, on the other hand, implicate spinning, or other, unidentified, effects as having the greatest influence over MA silk material properties across different spiders.

Figure 1. a) An example of a typical stress-strain curve generated during silk tensile testing, demonstrating where each major material property value is extracted. b) consensus amino acid sequence of Nephila clavipes MaSp1 and MaSp2 repetitive region showing the major motifs, GGX in blue, GPGXX in purple, (GA)n in yellow and (A)n in red. where A= alanine, G=glycine, P=proline and X can be Q=glutamine or Y=tyrosine or L=leucine). c) colour coded table of the structures formed by each of the above motifs and where they fall within either the amorphous or crystalline region.

Methods

Data collection

Although this is a comparative analysis81, I referenced to PRISMA82 (preferred reporting items for systematic reviews and meta-analyses) reporting guidelines as a searching and screening guideline. I searched Web of Knowledge and Scopus for primary articles using the following search terms: (i) Silk, (ii) Amino acid, (iii) Protein, (iv) Composition, (v) Mechanical and (vi) Properties. To further refine the search, ‘NOT’ terms were used to exclude many irrelevant publications. The full list can be found in the appendices.

I obtained the full text of studies that included the tensile properties and/or the amino acid composition of spider silks. Inclusion was limited to primary publications that included tabulated data on at least 3 of the 4 tensile properties, and/or 4 of the amino acids, or any combination including both amino acid and tensile properties. The tensile properties that I recorded were Young’s modulus (modulus), ultimate strength (stress), ultimate strain (strain), and toughness. I also recorded the amino acid profile of each spider’s silk. However the final analysis was later limited to alanine, glutamine, proline, glycine and serine due to the limited data available on amino acid found at exceptionally low percentages.

Phylogeny and Analysis

To control for effects of phylogenetic relatedness (non-independence) in our comparative analysis83, I used the most up-to-date phylogenetic information of spider relationships available in the literature 84-86.The tree was constructed using the ape package in R with the branch lengths being computed using the compute.brlen function using the Grafen method with a power of 19.

Once constructed, the missing data points (~33.5% of the data) of were imputed using an ancestral state reconstruction method in the rPhylopars package in R, which takes into account phylogenetic relatedness to append missing values87.

To take into account the error associated with imputed values, I used the variance of the interpolated data points to simulate the data set 1000 times to form an array88. The statistical analysis was then run and pooled from this array to ensure that the regressions were robust to the interpolated data points. The amino acid percentage of alanine, glycine, proline, glutamine and serine were each

compared to the four material properties (modulus, ultimate strength, ultimate strain and toughness) using a weighted phylogenetic generalized least squares (PGLS) regression. Weights corresponded to the number of individuals in each species; this analysis is mathematically very similar to a meta- analysis where each data point is weighted 81.

Results

Our search found 1672 publications, 514 of which were duplicates between the two search engines. Of the 1158 studies remaining, 1114 were excluded due to lack of relevance or available data or methods that involved chemical treatment or alteration of the spider’s silk (Figure 22, appendices). After an additional 22 papers were included from a back search, data was retrieved on 85 species from 66 studies for inclusion in the analysis, representing 44 genera across 21 spider families. Nephila and Argiope were the most represented genera with 129 individuals making up ~41 % of the data. The full list of references used in this study can be found in the appendices.

MA silk’s Glycine content (Figure 2 & Table 1) had the greatest effect on the mechanical properties, and was positively correlated with Young’s modulus and stress and negatively correlated with strain (Figure 3 and Table 1). Proline composition increases in the more derived spiders (Figure 2). Proline composition was negatively correlated with Young’s modulus and positively correlating with strain. Conversely, Serine correlated negatively with silk toughness overall (Figure 3 and Table 1). Alanine and glutamine were not significantly correlated with any mechanical properties included in the analyses.

Discussion

While decades of research have now established a basic understanding of the structure-function relationship of MA silk, questions such as the relative role of certain amino acids in influencing particular mechanical properties, remain unresolved. Here I performed a phylogenetic comparative analysis to ascertain the relationship between the amino acids alanine, glycine, proline glutamine and serine and the material properties Young’s modulus, ultimate stress, strain and toughness across a spider and some significant correlations.

Figure 2. phylogenetic tree and standardized trait values generated from the literature search showing each species within this study resolved to genus. The tree depicts the major clades involved within this study and development/ potential utilization of the two MaSp proteins. MaSp’s displayed at the top left and major clades on the bottom left of the highlighted areas. The colour of the box corresponds to the relative value of each trait lower values in white and higher in red.

Figure 3. pooled PGLS regressions of each major amino acid (alanine, glycine, proline, glutamine and serine) against each material property; where modulus is Young’s modulus, stress is ultimate stress and strain is ultimate strain. Significant correlations are displayed with red regression lines and data points are scaled and weighted in size by no. of individuals.

. . summary PGLS regressioncorresponding with the regressionsin Figure 2 Table 1 Table

Young’s modulus, or the silk’s ability to resist being permanently deformed by an applied force, is affected by the concentration of both proline and glycine. Young’s modulus decreases with an increasing proline composition. This is likely because proline is a branching amino acid that forms a zwitterion (contains both a positively and negatively charged group)76. Accordingly, it readily bonds across and within the protein chain and alters protein structure, transforming beta-sheets into beta- coils, spirals and turns. Moreover, it reduces protein alignment in MA silk when present in high compositions76. The changes in silk structure and alignment as a result of the presence of proline induces slip-stick stretching within protein chains, which is manifested as a decrease in Young’s modulus37, 47, 89-90. This proline-modulus correlation is therefore likely a direct result of this added stability and elasticity influencing the elastic phase of silk as it is stretched13, 37, 91.

Young’s modulus increased as glycine percentage increased. I believe this is a reflection of a greater glycine composition being found among species with silks exclusively, or almost exclusively, utilizing MaSp1. The major component of MaSp1’s amorphous region is the GGX motif, and this motif forms

46 the various turns and a 310-helical structures (Figure 1b,c) . Silks with such structures have fewer hydrogen bonding sights and low elasticity compared to those with proteins that form -turn structures. Moreover, the utilization of this GGX in tandem with the GA motif in MaSp1 results in approximately 10% more glycine content found within MaSp171. The mere presence of higher glycine along with a lack of elasticity and stability as a result of predominant GPGXX motifs within MaSp1 means primitive spider silks will tend to have both a low modulus and high glycine composition. Furthermore, additional glycine is found within MaSp1 in the edge of the crystalline region, which is involved in the formation -sheets as a result of the (GA)n motifs that flank a (A)n motif (these are not present in MaSp2, Figure 1c). The increased involvement of glycine in MaSp1’s less elastic and less stable amorphous region and its role in the formation of non-elastic structures may also explain the inverse relationship between glycine and Young’s modulus.

Strain, a measure of the amount of deformation the silk can undergo before breaking proportionate to its original size, was also influenced again by both Proline and Glycine. The GPGXX motif of MaSp2 induces silk proteins to form type 1 and type 2 -turns within the amorphous region of MA silk, which are thought to enhance extensibility whilst increasing the silk’s stability (Figure 1b,c)13, 37, 47. Similarly to Young’s modulus, an increase in glycine appeared to impede the deformation capacity of the silk. This is likely a result of the glycine content associated with MaSp1 expression and the general low elasticity in the silks of ancestral spiders, which do not utilize MaSp2.

Ultimate stress, or the maximum capacity of the silk to resist tension whist being stretched, was positively correlated with MA silk’s Glycine compositions. This is interesting as Alanine is the major component of the nanocrystalites, which are thought to be the major contributor to strength in MA silk fibres72. Our finding, nonetheless, may be explained as a consequence of high expressions of the

72, 92 (GA)n motif forming beta-sheets within the crystalline region resulting in stiffer silks . MaSp 1 dominant silks are generally stiffer and this motif has recently been shown to form extremely tight hydrogen bonding at the exterior of the nanocrystallites in Nephila plumipes, which are known to secrete silks that predominantly express MaSp193(see chapter 4).

Toughness, or the silk’s ability to absorb energy whilst deforming without fracturing, was negatively correlated with the serine content of the silk. Serine appears to have a slight trend with Young’s modulus and negatively trending with ultimate stress and extensibility, and thus the overall toughness of the silk. I find this result intriguing as serine does not form any of the major structural motifs in the repetitive region of MA silk. Serine is generally found more commonly within MaSp2, and this may explain the positive correlation I found between serine and Young’s modulus71. Curiously, this trend does not continue with silk’s extensibility, which would be expected if serine was to influence the amorphous region structurally like proline. Serine is found in abundance within the N-terminus of both MaSp1 and MaSp2. This region plays a regulatory function over dimerization and folding of both spidroins during spinning 67-68, 94. Increases in serine may correspond to shorter repetitive regions as a result of the increase in N- terminal domains per unit area this necessitates. The size of the repetitive region is known to affect silk material properties, and therefore we may be observing this indirectly through the serine content within silk69 Moreover, considering that serine is the third to fourth most abundant amino acid in silk, it’s interesting that little is known about its structural role at least within the silk’s repetitive region.

I noticed that the silks of the derived orb weaving spiders, represented as dark orange in our phylogeny (see Figure 1), generally had either high (10-14%) or low (1-4%) proline compositions and these correlated with the Young’s moduli and extensibilities of the silks. Moreover the distribution of the high or low proline silks were not predictable throughout the phylogeny. These findings are consistent with predictions of Savage & Gosline, Liu et al. and Blamires et al. 50, 76, 89.

I accordingly hypothesize that there are species-specific utilizations of MaSp2. The proline found in MaSp2 is considered to be relatively metabolically expensive to synthesize from constituent molecules48, 95, so spiders may be potentially forced into trading off between somatic maintenance and the MaSp2 expression, which likely impacts the material properties of silks50. This bimodal distribution is likely a reflection of MaSp2 expression and supports the main argument as to why spiders don’t just use predominantly MaSp2; that is, given the production costs of MaSp2, selection optimizes the use of MaSp2 only according to the ecological and biological conditions around each individual species. As a result, we observe species with the capacity to use MaSp2 down-regulating its expression as there may be no selective pressure to improve the elastic properties of their silk. A good example of this occurs within species of the genus Nephila which can be seen to use less proline (and therefore MaSp2) than any of their closest relatives within the phylogeny (Figure 2)76. Nephila are known for their size and use of barrier webs to assist in prey capture, both of which are factors that likely influence their need to increase the elasticity of their silk96.

Conclusion

Here I performed a comparative analysis comparing the amino acid compositions and mechanical properties of 66 studies covering 85 spider species, 44 genera and 21 families to shed light on the relative role of specific amino acids on the properties of spider MA silk. I found that MaSp composition mainly influences the elastic properties (modulus and strain) of silk and has some effect on the strength properties through glycine. Serine appears to be playing a larger role in silk material properties than originally thought and should be further explored to understand its exact structural influence. Our comparative analysis highlights the importance of how understanding the relative abundance of key amino acids can be utilized to predict material properties within secreted biological materials.

Chapter 3

Elucidation of metabolic pathways, amino acid synthesis and macro-nutrient partitioning in the production of spider’s silk as an extended phenotype

Craig, H.C., Kasumovic, M.M., Blamires, S., Rawal, A

Abstract

Energy/nutrient intake has often been associated with fitness when considering optimal foraging theory. This association, however, falls apart where fitness maximisation and energy intake maximisation are subject to differing constraints. Silk as an extended phenotype plays a significant role in the fitness of many invertebrates most notably spiders. A key area of interest is how spiders can maintain the quality of their silk despite significant variation in prey quality and availability. Measuring the flow of 13C isotopically enriched alanine into the silk of Nephila plumipes by solid state Nuclear magnetic resonance spectroscopy (NMR) allowed the exploration of variations in metabolic pathways in conditions where constraints to fitness maximization and energy intake maximization were implemented. As the labelled 13C alanine was converted to other aminoacids necessary for silk production by various metabolic pathways, I examined the differences in the abundances of different amino acids which are synthesised from alanine under nutrient deprivation (where the spiders were only directly fed a solution of labelled alanine) and control conditions (spiders were fed with both labelled alanine and crickets together). Idemonstrate the metabolic pathways that result in the shift of label from alanine and track the flow of labelled carbon into those amino acids. Overall, a significant amount of variation in label uptake between individuals is observed, reflecting either limtiations in standardard method of feeding label or biological variation in label utilisation. In general, however, greater levels of enrichment and spread of 13C label from alanine to other amino acids was achieved in individuals solely fed the 13C labelled alanine. The relatively low enrichment of serine compared to glycine may suggest a more direct pathway to glycine from alanine than currently thought and hints at a protein allocation trade-off between alanine being shifted into glycine rather than being utilised for gluconeogenesis. Lastly, the uniformity in labelling of proline and glutamine provides further

support for a protein allocation trade-off and somatic macronutrient partitioning in spiders as a result of the inability of -oxidation of fatty acids to contribute to amino acid synthesis within starved individuals.

Introduction

Procuring a sufficient amount and balance of nutrients is crucial for animals to maximise their Darwinian fitness. Energy/ nutrient intake has often been conflated with fitness when considering optimal foraging theory, this association, however, dissolves in situations where fitness maximisation and energy intake maximisation are subject to differing constraints97-99. Traditional optimal foraging models predict that animals hunt to maximize energy intake97. Recent, modelling100, nevertheless, challenges this presumotion and show the importance of balancing the uptake of competing nutrients, such as proteins and carbohydrates, in the foraging decisions animals make. The power of these so called ‘nutritional framework models’ lies in their ability to make predictions about how animals select among differing foods (i.e. how animals chose to move through nutrient space)101. Moreover, by overlapping so called ‘fitness landscapes’ over an animal’s chosen nutrient space it can be determined how its selection of nutrient throughout nutrient space affects chosen fitness outcomes. For example, fecundity is enhanced when wolf spiders forage to maximize protein to carbohydrate uptake ratio102-103.

Animals may need to trade off among foraging for the nutrients essential for somatic maintenance and other important activities, such as finding mates and reproductive output. When animals create structures, these structures may be considered an extended phenotype (EP)104-106. Examples of such structures include the male pufferfish’s “nest”, the mating bowers of bowerbirds, and the webs and/or cocoons created by spiders, caddiflies, glow worms, and certain hymenopterans and lepidopterans11, 105, 107-108. When extended phenotypes are constructed primarily of proteins, such as silk, they may affect how a given animal forages for and utilizes nutrients. For instance, it has been shown that web building spiders move through protein-carbohydrate nutrient space in a unique way so as to maximize its web and silk performance109. The assumptions made in building this model was that spiders directly utilize the amino acids acquired in their diet to invest in the proteins partitioned between somatic maintenance and silk production110, and the balance of these needs determines the nutrient acquisition decisions the spiders make109. Nevertheless the direct physiological modulation of protein through metabolic pathways is plausible and had not been considered in devising these models111-112.

It is reasonable to expect that since silks are imperative for maximising fitness through their role in protection (e.g., cocoons), prey capture, or sexual signalling (e.g., spiders web)105, 113-115, foraging to aquire nutrients, primarily protein, that maximizes silk and web performance affects a spider’s fitness. However, since the successful capture and consumption of prey by spiders is dependant on many factors, none more important than the energy absorbing capacity of the silks within their webs116-117, nutrient acquisition itself can feedback on foraging success and thus future nutrient acquisition. It is thus imperative that spiders manipulate and match the quality of their silks and web architectures to respond to any fluctuations in prey types and availability118-120. Experiments have shown that spiders can use a combination of prey nutrient content and prey vibratory stimuli in webs as cues to manuipulate web architecture and silk properties118, 121. Nevertheless, the mechanisms by which spiders can modulate the properties of their silks remains unresolved.

Spider dragline (or major ampullate/MA) silk itself is made from two protein monomer classes, Major ampullate spidroin 1 and 2 (MaSp1 and MaSp2) which boast large molecular weights in the range of 250-400 kDa122. Synthesized in specialized silk glands these spidroins are stored as a highly concentrated solution within the ampulla of the silk gland59. Alanine and glycine make up the majority of the amino acids (often up to 40%) within these spidroins, followed by serine, glutamine, proline, and tyrosine70. There is a disparity in the relative energy cost associated with synthesizing each of these amino acids by metabolic means, accordingly it is thought that the simpler amino acids like alanine and glycine are formed at a lower metabolic cost when compared to relatively complex amino acids such as proline and glutamine112. Moreover, certain amino acids, like tyrosine, are almost exclusively only available through direct dietary intake. Changes in the relative expression of MaSp1 and MaSp2 has thus been hypothsized as a mechanism by which spiders can regulate their silk protein investment under certain nutritional constraints. Given that the balance of these two proteins will affect the secondary structures found in the silk with the greater crystalinity associated with MaSp1 promoting greater strength, or MaSp2 structures promoting higher extensibily and toughness, it may also represent a means by which the performance of silk is manipulated by spiders across different diets.

Looking closely at a spider’s diet, the macronutrient profile of insects is generally high in protein and fat and relatively low in carbohydrates. The catabolism of proteins into amino acids and the - oxydation of fatty acids therefore plays a key role in normal spider metabolism for the production of glucose through gluconeogenesis and the production of electron carriers used for the production of

energy through the tricarboxylic citric acid (TCA) cycle. The TCA cycle is the highly conserved set of enzymatic reactions resposible for the production of NADH and FADH2 which are later used for the production of ATP123. Importantly, the TCA cycle is the primary metabolic process connecting somatic maintainance and the synthesis of many of the amino acids necessary for silk production112. This connection provides an opportunity to examine macronutrient partitioning between somatic processes such as energy production and amino acid synthesis used for silk production.

In this experiment, I fed the spider Nephila plumipes 13C uniformly labelled alanine to track how this key dietary amino acid is utilized in dragline silk production, and explore the metabolic processing and subsequent redistribution of labelled carbon into other amino acids under a starved (nutrient resticted) and fed state. As the majorly conserved metabolic pathways are highly predictible and well studied124, this allows me to understand the catabolism of alanine and resultant incorporation and spread of labelled carbon into silk through the observed differences intra and inter-residue labelling ratios within the silks NMR specta. I hypothesise: (1) that in a starved state I will find a higher relative utilisation of labelled alanine within their silk, as well as a higher shift of labelled carbon from alanine into their silk’s other major amino acids, (2) that the starved state will result in changes in the uniformity of lablelling within proline and glutamine due to known effects of -oxydation of fatty acids on the ratio of labeling within glutamine/ proline Cg :Cb, (3) that the differences in diet will result in changes in the silk’s MaSp composition and therefore relative abundance of label moving into proline and glutamine.

Methods

Spider collection and housing

Six adult Nephila plumipes were collected within a 5 metre radius of one another in a sheltered area of dry sclerophyll forest near Clovelly beach, Sydney NSW. The individuals were found aggregating together around a compost heap providing ample supply of flying insects and likely expirencing similar environment during development and maturation. These adult N. plumipes were then stored in a large 2.5×2.5×2.5 meter mesh enclosure at the University of New South Wales. Using such an enclosure provided individuals with a space that mimicked a natural environment and allowed individuals to build natural looking webs. Each spider was marked using non-toxic paint to allow individual tracking within the enclosure.

13C labelling and silk collection

40uL of a saturated 13C uniformly labelled alanine solution was fed directly to three of the spiders each day for eight days using a micro pipette. The other three spiders were also provided with 40uL of the same solution, except this solution was injected using a hyper-fine uL syringe into CO2 anaesthetized crickets which were then placed on the web and the web vibrated using a tuning fork to stimulate a predation response from the spider. An eight-day limit was placed on the experiment to limit the

125 stress on the spiders. Major ampullate silk was forcibly reeled, see ref , from the CO2 anesthetised spiders at a constant rate of 1m min-1 every other day starting on day 2 of feeding for a week and was spooled for approximately 45mins to 1hr until the spider stopped providing any silk, this allowed us to ensure that the 13C labelling was adequately incorporated into the silk124.

Solid-state NMR (ssNMR) spectroscopy

The 13C solid-state NMR experiments were measured on a Bruker Avance III spectrometer, with a 7 Tesla superconducting magnet, operating at frequencies of 300 MHz and 75 MHz for the 1H and 13C nuclei respectively. The MA silk samples from each silking of individual spiders, with sample weights ranging from 2.7 mg to 11 mg and were packed into in a 4mm zirconia MAS rotor with Kel-F cap with a Teflon insert to centre pack the sample within the rotor. The rotors were spun at speeds of 6.5 KHz to 13 KHz at the magic angle (magic angle spinning-MAS). The 13C spectra were recorded using 13C cross-polarization magic-angle spin (CPMAS) NMR at 6.5 KHz MAS with the total suppression of spinning sidebands (TOSS) scheme incorporated (CPTOSS) 126. 1H decoupling was provided at 80 KHz using the SPINAL 64 decoupling scheme. A recycle delay of 3 s and 2048 transients were acquired to provide sufficient signal to noise.

The 2D 13C-13C correlation spectrum was acquired using the CP–DARR technique127-128, at 13 KHz MAS, with a DARR mixing time of 25 ms. The spectrum was acquired with 1024 t1 increments of 16.6 μs each, 64 transients, and using a recycle delay of 3 s that resulted in an experimental time of 48 h. The 13C chemical shifts were referenced to Tetramethylsilane (TMS) using adamantane as a secondary reference.

Results

13C alanine label incorporation into the MA silk

The 13C solid-state NMR provided a rapid and non-destructive way to examine the uptake and spread of 13C alanine label into the silk over the seven days of MA silk collected from N. plumipes. The 13C CPMAS experiments reveal the fate of labelled alanine into the silk and reflect what has been observed previously within similar experiments exploring label uptake into silk. A comparison of the labelled and natural 13C CPMAS and 13C- 13C DARR silk spectra which residues the labelled carbons of alanine end up in after it is metabolised (Figure 4 & 5). Expectedly, AlaCa(C=49.1ppm), AlaC - sheet(C=21.1ppm), AlaC -helix (C=17.5ppm) and AlaCO (C=173ppm) are the primary nuclei being labelled, followed by GlyC (C=43.0ppm) and GlyCO (C=170ppm), then GlnC (C=53.2ppm), GlnC

(C=28.8ppm), Gln C(C=32.7ppm), GlnC(C=176.9ppm), the GlnCO is obscured by the AlaCO, then

ProC (C=61.5ppm), ProC(C=30.7ppm), ProC (C=27.1ppm), ProC and ProCO are obscured by the

AlanC and AlaCO respectively, and SerC (C=64.1ppm) , SerC (C=55.4ppm) the SerCO is also obscured by the AlaCO (Figure 4 & 5). Moreover, this study is consistent with previous work that shows that tyrosine is not labelled when feeding spiders labelled alanine as can be clearly seen in the comparison of labelled and natural silk (Figure 4)124. Moreover, the 13C- 13C DARR cross-correlation peaks helped disentangle peaks that are relatively hard to resolve in the CP MAS. The ProCa/ProCb peak (C=61.5/30.7ppm) resolved the intensity at ~61ppm in the CPMAS to proline which was previously thought to be SerCb. The ProCa/Cb cross-correlation peaks along with the GlnCa/Cb (C =53.2/28.8ppm) also helped tease apart the Glutamine and Proline residues that fall closely together between 27ppm and 33ppm (Figure 5). Importantly the presence of distinct crosscorrelation peaks also suggests relative uniformity in the labelling of each amino acid. Significant reduction or lack of labelling at any one carbon site within each amino acid would prevent the production of noticeable cross peaks using this method.

In figure 6 Isee how the 13C label increases in incorporation into the silk over time, there is individual variation in the uptake in both treatments. Individual NP1 and NP5 though in different treatments show a consistent increase in label uptake over the three days suggesting gradual incorporation of the label over time. The other four individuals show a greater jump in initial labelling and show differing patterns of incorporation. The cricket fed individuals have a less consistent increase as can be seen in individual NP4 in which the day 5 has less labelling then day 3, whereas individuals NP6 and NP2 in

the label only fed treatment are more consistent with increasing label incorporation over time. The Label only fed (starved) treatment tend to have greater label incorporation overall than the cricket fed treatment (Figure7a), moreover the label only fed treatment (starved) shows greater shifts into amino acids other than alanine, most notably glycine (Figure 7b).

13 Figure 4. C CPMAS comparing natural (unlabelled) silk and labelled silk scaled to same alanine C intensity , displaying the major amino acids within silk. natural silk in red and labelled silk in blue.

Figure 5. 13C-13C CP DARR extract from 0-75ppm showing homo- and hetero-nuclear correlations within the directly labelled silk, cross correlation peaks allowing the differentiation of the residues with similar chemical shifts.

Figure 6. time series of 13C CPMAS spectra showing the increase in label incorporation into the silk over 7 days between the cricket fed and label only fed treatments. Spectra range in colour from light blue to dark blue as time points increase.

Figure 7. a) 13C CPMAS spectra comparing the amount of label incorporation on the 7th day across each individual within each treatment. Spiders fed crickets are in blue and label only fed individuals are in red. b) same CPMAS Spectra scaled to the carbonyl to examine the spread label from alanine into the other major amino acids. As a result of this scaling, the baseline appears noisy in the cricket fed treatments as a result of the disparity in labelling. the peak assignments are assigned from the DARR (figure 5) and follow from 7a. NB the minor peaks of ProCa, SerCb are indistinguishable from the noise in the cricket fed treatments.

Discussion

Label Incorporation

The 13C CPMAS experiments revealed that the uptake of labelled alanine into the silk proved to be extremely variable between individuals, with significant variation both within and between treatments (Figures 6 & 7). Generally, the direct labelling treatment appeared to have higher overall incorporation into the silk, however a single individual (NP1) in the cricket fed treatment showed equivalent labelling to the directly labelled treatment; again this demonstrates the high individual- level variability in label uptake. The general higher levels of labelling in the label only individuals is likely due to their sole reliance on the labelled alanine for dietary intake, that being said it is surprising that there wasn’t a greater difference between treatments as the crickets effectively dilute the labelled alanine transferred into the silk.

The 13C CPMAS and 13C-13C DARR (Figure 4 & 5) show that the 13C label was shifted into glycine, serine, glutamine, and proline which is a known effect of the catabolism of alanine via alanine aminotransferase into pyruvate, which is a key intermediate for several metabolic pathways (Figure 8)124. However, this is the first notable case of substantial proline labelling through alanine, this is likely due to the lack of cricket supplementation within the label only treatment and therefore total reliance on alanine for sustenance124. Importantly, the amino acids synthesized from alanine appeared to be uniformly labelled (Figure 5). This is not only vital for looking at hetero and homo-nuclear correlations during NMR experiments but also in understanding the flow of alanine through metabolic pathways into the other amino acids129.

Implications for Metabolism and MaSp composition

I found an important difference between the cricket fed and directly labelled treatments in the distribution of the label into other amino acids. This can be best seen in comparing the relative intensities of the AlaCa (49.1ppm) and GlyCa (43.0ppm) peaks between treatments. Although the label generally increase over time (Figure 6), the relative ratio of labelled alanine to glycine decrease over time as more label is shifted into glycine and the other amino acids (Figure 7b). The spiders fed labelled crickets shift less label into glycine, proline, glutamine and serine. This can be attributed to the fact that crickets themselves contained various amino acids, the highest of which is glutamate130, which is the precursor to both proline and glutamine (Figure 8). Moreover, normal protein catabolism within

carnivores contributes various intermediates into the tricarboxylic citric acid (TCA) cycle (Figure 8)123, 131. This means that the non-labelled carbons from the cricket preferentially, or out of a chance, are incorporated into the silk proteins through metabolic actions, leading to the dilution of the overall distribution of the labelled alanine. In contrast, in the label only treatment, as free amino acids within the body of the spider decrease over time the TCA cycle becomes increasingly reliant on labelled alanine which enters the cycle as OAA after carboxylation of pyruvate (Figure 8). Moreover this also means that I am ultimately unable to disentangle this dilution effect from differing utilisation of MaSp1 and 2 between each treatment.

Utilisation of a new metabolic pathway suggests a protein allocation trade-off

Previously in silkworms, the carbon of alanine was shown to be redistributed into serine and glycine through pyruvate to OAA (figure 8)132. The label is moved into these amino acids as an intermediate step within the gluconeogenesis pathway. In herbivorous and omnivorous animals, glycine formation would usually occur from glycolysis which is the inverse process of breaking down glucose to contribute to amino acids synthesis123, 133. However, as N. plumipes is an obligate carnivore, gluconeogenesis should be the major metabolic pathway being utilised as insect bodies are generally high in protein and fat, and low in carbohydrates130. Interestingly, Iprovide evidence that this is not what is happening in our spiders, as 13C within serine did not appear to increase at a similar relative rate as it did within glycine, despite serine being an intermediate step in the synthesis of glycine (Figure 6, 7 & 8). This indicates a more direct metabolic pathway from alanine into glycine. The alanine- glyoxylate aminotransferase (AGT) pathway is utilised across many species134-135, although this is yet to formally demonstrated in spiders, our results strongly suggest this is likely. Furthermore, a search of genbank (NCBI) shows that the mitochondrial AGT gene is present within Parasteatoda tepidariorum a basal member of Araneoidea which contains the derived orb weavers including N. plumipes85. The glyoxylate-alanine aminotransferase pathway operates in a similar manner to alanine aminotransferase, however, it is responsible for the production of glycine and pyruvate from alanine and glyoxylate134. If glycine is being produced directly from alanine this may also help explain the relatively poor labelling of SerC (C =64.1ppm) when compare to SerC (C =55.4ppm), as any serine formed from glycine can only receive labelled C and CO which would result in a reduction in the labelling of SerC (Figure 4 & 5). Importantly, given that glycine is the most abundant amino acid in silk, it lends to the idea that having multiple possible pathways for glycine synthesis into silk would be advantageous. If this is occurring this would be an example of a protein allocation trade-off, as the

alanine within starved individuals appears to be moved into glycine via AGT pathway for silk production, as opposed to gluconeogenesis and the TCA cycle for the production of glucose and electron carriers needed for somatic maintenance. While our results cannot confirm this conclusively, they do suggest that investigating the presences of enzymes used for the metabolic synethsis of amino acids utilised within silk would be a fruitful avenue for further research.

Figure 8. A map of labelled carbons movement from alanine through the various metabolic pathways and their intermediates into the major amino acids being observed labelled in the N. plumipes silk. arrows indicating the direction of flow between each intermediate. Arrows under furmurate and succinate demonstrate their molecular symmetry allowing carbon randomisation. * correspond to labelled carbons from alanine, purple asterisk show carbons that can come from either Acetyl CoA or OAA after furmurate/ succinate randomisation within the TCA cycle. Blue asterisk corresponds to label carbons that can only come from Acetyl CoA. Grey asterisk corresponds to carbons transferred through the proposed AGT pathway.

Starvation and macronutrient partioning

The relative uniformity of carbon labelling among glutamine and proline in the label only fed treatment is vital in not only understanding the flow of carbon into silk but more importantly, the spider's metabolic response to only being fed a single amino acid129. To understand this, its important to first understand the impact of starvation on the TCA cycle and the relative impact that has on the flow of alanine’s labelled carbon into other amino acids. Generally, during starvation -oxidation of fatty acids results in a supression of carbons entering the TCA cycle via Acetyl CoA from protein/amino acids. When using labelled alanine, this would result in a change in the relative abundance of labelled carbon found in the C and C of glutamate and therefore glutamine and proline (Figure 8). A disparity in the labelling ratio of C:C in glutamate caused by starvation is a well-known effect observed in invertebrates129, 133. Starvation in invertebrates can result in a 4 fold increase in the contribution of 13C

129 label through OAA into the TCA cycle, resulting in an equivalent increase in C, C labelling . This is not observed within glutamine or proline within the nutrient-deprived spider’s silk as the relative intensities of the C peaks are well defined (GlnC C= 32.7ppm and ProC C= 27.1ppm, Figure 7) and are approximately the same relative intensity as the C/Cb. Assuming a trade-off between the contribution of alanine to OOA and Acetyl CoA, the uniformity in labelling of glutamine and proline can only be achieved if the ratio of OAA to Acetyl CoA entering the TCA cycle is approximately 2:3 respectively. Ultimately this demonstrates that there is no suppression of labelling entering from

Acetyl CoA and therefore no -oxidation of fatty acids occurring. This suggests that spiders prioritise amino acid synthesis from the abundance of alanine in their system, as generally fatty acids (with the exception of odd chain fatty acids) cannot be used to form amino acids123. More interestingly this appears to be an example of somatic macronutrient partioning or an isolation of silk production and fat in spiders. Spiders would have had to modulate other somatic processes to conserve fat whilst investing the small amount of dietary alanine to maintaining silk production whilst in a starved state. Macronutrient partitioning is a good example of a way to maximise fitness by selectively choosing where to invest different forms of energy within the body136-137; in this case the results suggest that the spiders are likely investing the limited dietary protein they are obtaining into silk production and any energy produced from the TCA cycle whilst conserving fat potentially for other processes like the production and maintenance of their eggs.

Conclusion

Despite high variation in label uptake into silk, it is clear that dietary supplementation with crickets during labelling schemes affects both the uptake and spread of label into the other amino acids in the silk. Overall, it can be seen that without the additional source of amino acids from the crickets, researchers can achieve greater levels of label uptake and higher levels of enrichment of the other amino acids. Looking closely at the spread of amino acids Ican track the flow of labelled carbon into each amino acid and Ifind evidence that a more direct route may exist into glycine from alanine than was previously thought. Upregulation in the utilisation the this alanine-glyoxylate aminotransferase would result in a protein allocation trade-off between meeting the glycine needs for silk production and shifting alanine into the gluconeogenesis pathway and TCA cycle responsible for energy production. Moreover, the relative uniformity in labelling of both proline and glutamine suggests that in the nutrient-deprived treatment, amino acid synthesis was being prioritised from the abundance of alanine. The fact that fat isn't being oxidised would necessitate changes in other factors contributing to the balance of energy utilisation and expenditure, furthermore suggests that spiders are partitioning and adjusting somatic processes and behaviours to save energy to ensure that what little energy they are receiving in a “starved” state can be utilised for synthesis of the amino acids within the spidroins. More broadly speaking, the presence of these kinds of trade-offs furthers our understanding of how animals, and particularly carnivores, may choose to modulate/ partition macronutrients and metabolites in systems that requires significant effort in the production of extended phenotypes that directly impact fitness.

Chapter 4

DNP NMR spectroscopy reveals new structures, residues and interactions in wild spider silk

Craig, H.C., Blamires, S., Sani, M., Kasumovic, M.M., Rawal, A., Hook, J.M. This chapter is currently in review.

Abstract

DNP solid state NMR spectroscopy allows non-targeted analysis of wild spider silk in unprecedented detail at natural abundance, revealing hitherto unreported features across several species. A >50-fold signal enhancement for each silk, enables the detection of novel H-bonding networks and Arginine conformations, and the post-translational modified amino acid, Hydroxyproline.

Introduction

Biological materials refined by natural selection outperform all comparable synthetics. Limpet teeth, abalone nacre, mussel byssus and spider silk are examples of materials displaying impressive material properties 3-5, 138. Spider major ampullate (MA) silk is an archetypical biological material as it’s a bio- inert, has natural anti-microbial properties139, and most notably its low density, high tensile strength and exceptional extensibility make it incredibly tough13, 56, 90-91. Although synthetic materials of comparable toughness can be made in the laboratory, no synthetic material combines these properties in a similar way 2, 140. Consequently, there is immense interest in MA silk's structure 52, 141- 142. Advances have been made in our understanding of MA silk performance by probing aspects of the silk’s structure and function. Genomic analysis shows that MA silk consists of two proteins: Major Ampullate Spidroin 1 (MaSp1, MWt: 250 kDa) and Major Ampullate Spidroin 2 (MaSp2, MWt: 312 kDa), featuring two structural domains: the crystalline domain made up of a poly-alanine (A)n motif, and in MaSp1 this is often flanked by a polyalanine-Glycine (AG)n motif (Figure 13)13, 23, 91, 143; and the amorphous domain comprises of less structurally ordered Glycine-rich regions. Important work has

established that these motifs form tightly packed antiparallel β-sheets forming nano-crystallites parallel to the fibre axis2, 12, 62, 65, 72. The H-bonds of the crystallite network are posited to be responsible for its strength. The crystallites are known to form the amorphous domain23, 144-145, which is dominated by a GGX motif in MaSp1 that form various turns and 310-helices. MaSp2 features the addition of a GPGXX motif, forming type-II β-turns that sequentially come together along the protein backbone in a highly extensible β-spiral (Figure 13)44, 46, 146. These observations have led to the conclusion that the less ordered H-bonding patterns, the β-turns and the α-helices within the amorphous domain, are responsible for silk’s extensibility.

Of the analytical techniques available, ssNMR spectroscopy has been paramount for understanding structure-function relationships in spider silks2. Unlike other techniques, ssNMR facilitates detailed examination of silk structure and dynamics, and the various 1D and 2D experiments are fundamental to detail atomistic features of the protein backbone and higher order structures by direct and indirect examination of the intermolecular bonding arrangements37, 146-149. Two main limitations to using ssNMR are (1) large masses of silk are required, limiting the scope of experiments of the inter- and intra-species structural variation. Force-feeding spiders 13C and/or 15N enriched amino acids72, 124, 147 does not increase the amount of silk available, but can lead to enhancement of NMR signal to noise ratio of those residues. This technique, however, has additional limitations, as diet impacts silk composition and properties109, 150-151, and it is unclear how the dietary manipulation and exposure to relatively high concentrations of specific amino acids affects the spiders and their silk.150 (2), the timeframe of the experiments to achieve an appropriate uptake of labelled amino acid(s) into silk makes it impossible to gain structural information on wild native state silks.

Here I show how Dynamic Nuclear Polarization (DNP) enhances the solid state 13C and 15N NMR signal by greater than 50 fold152-155, allowing the elucidating of previously undetected protein secondary structural features in wild silks without 13C or 15N isotopic labelling, across several species of spiders.

Methods

Materials

Silk samples were collected from wild caught individuals of three species of spider: Latrodectus hasselti, Argiope keyserlingi, and Nephila plumipes. I varied the number of individual spiders silked to reach the following sample weights. For L. hasselti I collected 12 mg of silk from one individual; for A. keyserlingi I collected 14 mg from two individuals; for N. plumipes I collected 18 mg from three individuals. Individual A. keyserlingi and N. plumipes were collected in the field in Brisbane, Queensland and silked in a makeshift laboratory upon collection; A. keyserlingi and L. hasselti were collected around Sydney, NSW, and silked at the University of New South Wales, Sydney, upon collection. Little intraspecific variation is known or expected within the silk of each species so we will only focus on one sample for the purpose of this study. All spiders were silked using procedures

148 outlined by Blamires et al. (2016) this involved anaesthetising each spider using CO2 gas and carefully pulled a single dragline fiber from the spinneret onto a spool that was spun at a constant speed (1 m·min−1) for ~1 h whereupon 14-18 mg of silk was collected. All silks were extracted under controlled temperature (~25 °C) and humidity (~50% R.H.) in still air, so reeling and the post-spin environment did not influence their subsequent chemical or mechanical properties. Each species was chosen for their distinct differences in MaSp composition and proline content as this has been shown to have significant impact on material properties. A. keyserlingi having the highest MaSp2/ prolination followed by L. hasselti then N. plumipes respectively151.

2 H2O (D2O) and 1,1,2,2-tetrachloroethane (TCE) were purchased from Sigma-Aldrich (Castle Hill, Australia). AMUPol and TEKPol were purchased from Cortecnet (Voisins-Le-Bretonneux, France).

Sample Preparation

Each silk sample was wetted in an Eppendorf tube with equivalent volume of 10 mM AMUPol

(D2O/H2O 80:20 solution) at a 1:1 mass ratio (i.e. 14 mg of silk to 14 L of AMUPol solution) for 10 mins then packed into a 3.2 mm sapphire rotor with a silicon plug and a zirconia spinning cap.

The process was repeated with 10 mM TEKPol in TCE, however with 36 L being added to 18 mg of Nephila silk for comparison. Part of the TCE solution was absorbed into the silk as indicated by visible sample swelling. The treated spider silk was packed as above into a 3.2 mm sapphire rotor with a

silicon plug and a zirconia spinning cap. The results for this can be found as a part of the appendices (Figure 23, appendices).

Dynamic Nuclear Polarization Enhanced 13C and 15N Cross Polarization MAS (DNP CP MAS) Solid-State NMR Spectroscopy

DNP NMR measurements were performed on a 400 MHz (9.4 T) Bruker AVANCE-III-DNP system (Germany) equipped with a 263 GHz gyrotron and a triple resonance 3.2 mm low temperature Magic Angle Spinning (MAS) probe. DNP was achieved by irradiating the sample with 130 mA microwaves matching the AMUPol electron frequency156. The samples were inserted into the probe and spun to MAS rates of 8 kHz (± 3Hz) with sample temperature set at 110 K with liquid nitrogen for all experiments, except if indicated in a figure caption.

DNP-enhanced 13C and 15N CP MAS were acquired with 102 kHz 1H excitation, followed by a 1H linear amplitude ramp (50% to 100%) of cross-polarization with 1.5 ms contact time and 102 kHz SPINAL-64 1H decoupling during acquisition157-158. The RF amplitudes for 13C and 15N were 59.5 kHz and 44.6 kHz,

1 respectively. A recycle delay of 10 s was chosen as ca. 3 x T1( H) which was obtained indirectly from saturation-recovery 13C CP MAS experiments. 64 scans were accumulated for the 1D experiments using 25 Hz and 80 Hz line broadening for 13C and 15N spectra, respectively. The microwave-off acquisitions for 13C CP MAS were performed under identical conditions.

DNP-enhanced 1H-13C and 1H-15N 2D correlation (HETCOR) experiments were acquired using similar CP conditions. The frequency-shifted Lee-Goldberg (FSLG) scheme was used for 1H homonuclear decoupling during the indirect evolution time with an RF field amplitude of 102 kHz. States-TPPI detection was used and a scaling factor of 0.56 was applied to correct for 1H chemical shift scaling. 13C

15 HETCOR were obtained with 256 t1 increments of 62.5 μs with 2 scans accumulated and N HETCOR were obtained with 126 t1 increments of 62.5 μs with 8 scans accumulated. The 2D spectra were processed with a 4k by 1k complex point matrix using 100 Hz of line broadening. 13C DNP enhancement factors (εDNP) were determined by scaling the intensities of the spectrum without and with MW irradiation.

Results and Discussion

DNP Enhancement

The DNP enhanced 13C and 15N CP-MAS spectra for the silks of three spider species, Latrodectus hasselti, Argiope keyserlingi and Nephila plumipes were acquired rapidly for 13C, ~10 mins compared to ~8 hours using standard ssNMR151), and resulted in a signal improvement for 13C DNP gain (εDNP) of 64, 59 and 50, respectively (Figure 9). This exceptional signal improvement allowed examination of each silk’s protein structure in considerably more detail than previously possible and exposed hitherto unexplored intermolecular interactions and undetected groups (Figure 10 & 12). Peaks in the NMR spectra were assigned with reference to previous work124, 151, however, DNP enhancement allowed precise differentiation of lower abundance residues such as Arginine (Figure 9), while rapid and detailed 2D 1H-13C and 1H-15N heteronuclear correlation (HETCOR) analysis at natural abundance revealed additional novel findings. Although previous ssNMR experiments with 13C-labelled silks2 has provided details on secondary structure and folding patterns along the protein back-bone, the DNP gain has allowed, for the first time, categorization of major structural motifs and identification of novel hydrogen bonding (H-bonding) patterns (Figure 10c) in wild silks close to their native state.

Figure 9. 400 MHz 13C CP-MAS DNP ssNMR spectra at natural abundance of L. hasselti, A. keyserlingi and N. plumipes silks, displaying DNP gain (ε), experimental time (t), sample weight (W) and the comparison of the same sample irradiated (left panel, blue) and un-irradiated silk (left panel, red). Some amino acid peak assignments are also shown. Right panel shows the 15N CP-MAS DNP ssNMR spectral assignments of the

Arginine (Arg) residue. Inset A showing enhanced view allowing the differentiation of the Tyrosine (Tyr) Cz and Arg Cd peaks. Inset B Sum of the spectra and spectra -8kHz to remove carbonyl spinning side band (*ssb) exposing presence of glycoprotein sugar residue in the L. hasselti silk

Table 2. High Sensitivity Advanced Amino Acid Mass Spectrometry (AAA-MS) results showing the average relative mole percentage of the major amino acids within each species silk. (NB: different silk samples collected at tandem were used for this purpose to preserve the original samples used for DNP NMR analysis and to prevent any influence of the added AMUPol)

demonstrating motif

sheet of the crystalline

within the β

ampullate silk at natural abundance.

major

bonding

O) O) neighbouring withits C

Enhanced of view carbonyl region displaying ability

N. N. plumipes

and

A. A. keyserlingi

, ,

L. L. hasselti

s), s), of

=150

sheets sheets and helices, as well as the relative strength of H

Enhanceddisplaying view the interaction of the amino group (NH

and intermolecular correlations and new findings are highlighted in red.

gnments.

C C HETCOR DNP ssNMR spectra (contact time, T

2D 2D

. . a)

Figure 10 Details of amino acid assignments, intra todifferentiate highly detailed peak assi differentiation and differentiation of Glycine NH residues into β regionbetween species.

Figure 11. Expansions and assignments of the 1H-13C HETCOR experiments showing 1D 13C and 1H slices taken through the cross correlation peaks of interest (“Hydroxyproline“ from A. keyserlingi and “NH downfield shifts“ from N. plumipes). δH and δC in top left corners represent the point (in ppm) at which the 1D slices where extracted, where a projected sum was provided within the NH Downfield shifts

Continued

Figure 12. Expansion and assignment of 1H-13C HETCOR experiments: carbonyl shifts from N. plumipes showing 1D 13C and 1H spectra taken through each major cross correlation peak. Deconvolution was provided for potentially ambiguous overlapping peaks where necessary, deconvoluted peaks provided in black, with the simulated projection presented in red.

Figure 13. molecular structure of spider silk showing the amorphous region in blue and crystalline region in red. Location of the (A)n and (GA)n mofits within the crystalline region displayed. a) hypothesised utilization of the R145 conformation of arginine assisting in inter-chain cross linkage. b) Difference in hydrogen bonding length detected in N. plumipes between the (A)n and (GA)n motifs. c) -turn structure produced with hydroxyproline containing added H-bonding site.

Variation in crystallinity

There are detailed structural differences within the crystalline regions of each silk. Variation in the strength of hydrogen bonding within the crystallites between each species was detected in the Ala

Cα/NH δH shift correlating with the known effects of MaSp composition and the influence of Proline on β-sheet formation2, 109 (Figure 10c & 11). Of the species examined, A. keyserlingi had the highest MaSp2 induced prolination and subsequently showed the longest NH···O H-bonding (~1.97Å) associated with its Alanine (A)n β-sheets with a δH shift of 8.2 ppm, compared to L. hasselti and N. plumipes which had δH 8.7 and 9.1 ppm, giving them an NH···O length of ~1.88Å and ~1.83Å respectively159. N. plumipes thus had the shortest NH···O β-sheet H-bond lengths and the most compact crystallites, followed by those of L. hasselti then of A. keyserlingi. This is consistent with the predicted utilizations of MaSp2 for each of these species and also correlates well with the documented interspecific differences in silk stiffness109, since this property is attributed to the crystalline region2, 109.

1 Another major difference of N. plumipes silk was an additional peak (Cα/NH ) at δC 49.5/δH 14 ppm with integration of the area accounting for ~28% of the Cα/NH residue and was accompanied by a peak of equal intensity at δC 42.3/δH 13.5 ppm, accounting for ~26% of the overall Glycine (G) Cα/NH residues. The extent of δH down field shift in both these residues implies that the (GA)n motif flanking

159-160 the (A)n forms extremely short NH···O H-bonding, approaching 1.5 Å, so is associated with exceptionally tight packing of the antiparallel β-sheets at the exterior of the crystallites. This difference is a consequence of the density of H-bonding between the surface (created by the (GA)n motif) and interior (created by the (A)n motif) of the crystallites within N. plumipes’ MA silk (Figure 13b).

Hydroxyproline

Hydroxyproline (Hyp), hitherto unknown in spider silk, was detected in A. keyserlingi silk, as evident from the peak at δc 72/ δH 3.0 ppm, assigned to Hyp Cγ, with the associated Hyp Cβ peak at δc 41/ δH 1.5 ppm. Both peaks are consistent with the effects that the additional OH group has on the bond angles of Proline and known chemical shift data for Hyp161-162. This non-proteinogenic (non-coding) amino acid is usually synthesized by the metabolic hydroxylation of Proline and is produced by many animals, being the major residue of human collagen. As a structural protein, collagen is similar to spider silk, so Hyp in silk may be utilized in a structurally similar way. Hyp provides the functionality of Proline for the formation of β-spirals whilst also contributing to the stability of the silk’s amorphous region, with the added –OH participating in H-bonding within the β-spirals. Because Hyp cannot be coded for within the MaSp or silk genes it is likely tied to the expression of high proline spidroins such as MaSp2 and Flag, which explains why there is no evidence of Hyp in the MA silk of N. plumipes, with a lower proportion of proline and MaSp2 (Table 2, Figure 10a)150. A. keyserlingi, on the other hand, expresses significantly more MaSp2 and, therefore its MA silk contains comparatively more Proline151. However, as seen in the amino acid profile of each species’ silk (Table 1), the abundance of Hyp is not necessarily linked to the abundance of Proline itself and might be dependent on physiological factors. Although Hyp is produced in small amounts, it likely contributes to the stabilisation of the amorphous region, enhancing elasticity in this species’ silk through additional H-bonding through the additional OH (Figure 13c) 48.

Structural role of arginine

Little work has been done on the structural characterisation of nitrogen interactions in spider silk, at natural abundance. However, DNP enhanced 15N CP-MAS ssNMR spectra of the three species silks found previously undocumented variations within the Arginine residues and evidence of Arginine H- bonding within the amorphous region (Figure 9, right panel & 13a). Distinct peaks at δN 70 ppm and

δN 85 ppm are consistent with the Arginine NH1,2 and N, components of Arginine’s guanidinium group. Arginine usually comprises ~2% of the amino acid composition of MA silk and is primarily found in the amorphous region of MaSp2 in Latrodectus sp. and MaSp1 in Nephila sp. but is less utilised among Argiope sp. (Table 2)62, 71. Until now, little has been reported on the structural role of Arginine due to its low concentration in silk and its relatively irregular occurrence in any primary motif.

Distinct structural differences were found between L. hasselti and the other two species. Arginine Nε and NH1,2 peaks in A. keyserlingi and N. plumipes are consistent with that expected of free Arginine.

However, Arginine of L. hasselti shows the absence of the peak at δN 85 ppm and the presence of an intense peak at δN 70 ppm (Figure 9 & 13a). This is ascribed to an up-field shift of Arginine Nε coinciding

15 with that of Arginine NH1,2 resulting in an amplified N intensity in the spectrum of L. hasselti silk. The difference can be attributed to the slower rotational motion of the guarnidino group, specifically rotation around the Nε–Cδ bond, because of its involvement in H-bonding networks. The high level of stabilisation of the Arginine Nε in L. hasselti’s silk, is consistent with R145, a conformation of Arginine in which the guarnidino NH1,2 forms both a weak H-bond as well as ionic bidentate H-bonds (Figure 13a)163-164. Arginine in L. hersperus’s MA silk is usually found following the GPG triad in the GPGXX motif (GPGRX, where R is Arginine). H-bonding to the Arginine side chain likely plays an important role in the stabilisation of the amorphous region by linking the β-spirals formed by the GPGXX motif to the surrounding amorphous region. Arginine is used in a structurally similar way in collagen, with its extended side chain adopting conformations that facilitate charged interactions between neighbouring collagen molecules.

Glue glycoprotein

Another undocumented feature is assigned to a carbohydrate moiety in L. hasselti’s MA silk. High intensity peaks at δc ~100 ppm (Figure 9 & 10a) consistent with the anomeric (C-1) carbon, together with that at δc 72 ppm, assigned to -CH2-OH, is reasoned to be of N-acetylgalactosamine, a significant component of Glycoprotein A, in black widow defensive secretion165-166.

Conclusion

In conclusion, the key findings of H-bonding motifs, and of novel residues and their conformations indicate that the requirement to collect large sample masses and/or isotopically labelling is relinquished when examining spider silk by DNP NMR. Future studies with greater sample sizes at lower sample weights can examine in greater detail the extent of structural variation in spider silk. This enables a wider range of non-MA silks including synthetic silks to be studied and a broader examination of the extent of silk’s structural and functional diversity. Moreover, DNP enhancement can be used along with isotopic labelling to isolate more subtle structural features that play a significant role in silk function53.

Chapter 5

The origin, function and implications of biologically induced cavitation in wild silkworm silk

H.C. Craig, Y. Yao, N. Ariotti, M.M. Kasumovic, R. Rajkhowa, A. Rawal, S.J. Blamires,

Abstract

A significant effort has been put in to understanding the structure-function relationship of silk using various techniques, such as NMR, probing silks molecular structure in an effort to correlate that with its material properties. Recent research into wild silkworm silk reveals that it can have equivalent properties to spider silk, however macrostructural flaws generated during spinning generally impede its maximal functional potential. Here I investigate possible sources of variation within silkworm silk’s chemistry and macrostructure that would result in any significant variation in the material properties along the silk fibre’s axis. To do this I compared the silk from the inner and outer portions of Bombyx mori, Antheraea assamensis and Samia cythia ricini cocoons to uncover if they are altering their silk in some manner in the different layers. Nuclear magnetic resonance spectroscopy and Fourier transform infra-red spectroscopy results showed that the silk fibres are chemically homogenous along the fibre axis meaning the variation in mechanical properties likely stems from macrostructure. This was backed up with quasistatic nanoindentantion results indicating that, at the local scale, there is no difference in material properties between the inner and our portions of the silk. AFM analysis revealed extensive cavitation within the silk of S. c. ricini and A. assamensis which was further investigated by electron tomography. Pore fraction analysis revealed that the cavitation varied significantly between each species and within the silk of A. assamensis. Cavitation has known effects on polymer performance and would be a significant source of material property variation. Moreover, if cavitation isn’t taken into account when running tensile tests, the difference in cross-sectional area will have a significant impact on the calculations used to determine the strength and toughness of the silk fibre. Lastly, I add evidence that shows that these pores are biologically induced within the silk gland and serve an adaptive function. Using simulations I show that cavitation decreases heat transfer through the silk thereby increasing the insulative properties of the wild silkworm silks. Ultimately this may indicate

that natural selection may be acting to maintain cavitation within the wild silk despite its negative impact on the material properties of the silk.

Introduction

Silk has independently evolved and is utilised by a diverse range of invertebrate species from the well- known spider and silk moth silks to the lesser known amphipod, lacewing fly and bee silks6-7, 167. Each of these silks has a variety of uses, and as such, exhibit an array of impressive material properties depending on the needs of the organism producing them. Understanding the processes of silk formation and the resultant molecular and long-range structure that enables silk to achieve its impressive material properties is a necessary step to generate high performance biomimetic fibres52. Of the better-known examples, silkworm silk is generally stiffer (having a higher Young’s modulus) and has an overall lower strength and toughness than silks produced by spiders17. This can be attributed to the silks from these different species having experienced different selective pressures168. For example, the mechanical performance of the major ampullate silk of orb web spiders is shaped largely by the need to absorb the impacts of flying prey, whereas the material properties of silkworm silk are driven by the need to protect the moth forming in the cocoon during metamorphosis168.

Mulberry silk produced by Bombyx mori is the best known and extensively studied silk produced by silkworms169-176. Mulberry silk itself is made up of heavy chain, light chain, and p25 linker fibroins31. Heavy chain fibroins are thought to make up the bulk of the silk fibre and is majorly comprised of highly repetitive GAGAGS and GAGX motifs where A=alanine, G=glycine, S=serine, and X can be tyrosine, valine or serine147. Many spectroscopic techniques such as solid-state NMR, Ramen, and FTIR spectroscopy have established that these repetitive protein sequences conform into two major structural domains: the GAGAGS motif conforms to a nanocrystalline domain, made up of relatively large crystallites formed from tightly packed -sheets and the GAGX motif results in an amorphous domain made up of helical and less ordered structures17, 147, 169, 172-173. The identity of these domains has been verified by X-ray diffraction analyses17, 38, 177, which has also established that the density and alignment of the nanocrystallites play a role in influencing material properties of the silk13, 150, 178-179 Moreover, most research on the structure-function relationship of silks using the abovementioned techniques have concluded that higher crystallinity results in an overall increased stiffness (Young’s modulus), and variation in crystallite size and orientation influences extensibility and overall toughness17, 180.

More recent work has started to examine non-mulberry or ‘wild’ silks (Saturniidae). The silks from these species are distinct in the fact that they come from undomesticated species whereas Bombyx mori has been domesticated for over 5000 years. Interestingly the silks from wild species such as Muga or Eri spun by Antheraea assamensis and Samia cynthia ricini respectively, exhibit significantly different material properties than that observed in mulberry silk17. This silk from wild species more closely resemble the spidroins found in spider silk with poly-alanine (A)n runs forming the crystalline domain surrounded by an extensive glycine rich region forming the amorphous domain181. Interestingly, the similarity in amino acid sequence functionally corresponds to both these silks exhibiting strain hardening, whereas the mulberry silk does not. Strain hardening is a process that occurs in semicrystalline polymers and involves an increasing resistance to strain as hydrogen bonds are dislocated within the crystallites as the fibre is stretched17, 39-40. As a result of this disparity in the crystalline regions structure and resultant function, mulberry silk generally has a higher tensile strength, whereas the silks from wild species have higher elasticity (breaking strain) and toughness182- 183.

A closer examination of the material properties of wild silkworm silks reveals that through repeated mechanical tests on unbroken sections of stretched silk, tensile properties comparable to that of spider silk can be produced184. The breakage of stretched silk fibres is thought to be caused by two major forms of macrostructural deformation: crazing which is the formation of many small cracks on the surface of the fibre and cavitation which is the formation of voids within the fibre185-186. In wild silks this suggests that one or a combination of these macroscopic flaws at the point of fracture prevents the fibre from achieving an extensive plastic phase, impeding the functional potential of the silk. It may be the case that with no direct and continual selective pressure to remove or reduce the formation of such flaws, silk from wild species is functionally as tough as it needs to be to perform its biological function.

The various studies into silkworm silk have often failed to consider biologically induced variation such as the effects of macroscopic structural variation in silkworm silk13, 17, 182, 187. Silkworm silk is spun as a continuous silken thread forming the cocoon from the outside in, with the acceleration and deceleration of the drawing motion playing an important role in influencing material properties by affecting crystal alignment25, 188-189. Importantly, the notable amounts of material property variation (generally attributed to error) observed both on an intraspecific and an individual level suggest there is more to material properties of silkworm silks than what can currently be explained by the simple

structure-function relationship such as differences in crystalinity17, 184. Moreover, it remains to be determined exactly what these sources of the material property variation in silkworm silk are. As such, I hypothesise two possible alternative explanations for the material property variance observed in silkworm silks: (1) that a difference in molecular structure along the fibre is responsible for a proportion of the variation detected in the tensile properties, or (2) differences in spinning as the cocoon proceeds to be completed, impacts the silk via changes in shear forces as the internal diameter of the cocoon decreases. I thus investigated herein the chemical microstructure, macrostructure, and material properties along the fibre length of silks from three different silkworm species.

Methods

Silk Degumming

The silk cocoons washed thoroughly with water, cut into four parts and then gently peeled by hand to separate outer and inner layers. Separated layers from few cocoons were degummed separately.

Degumming was performed in a solution of 2 g/L of Na2CO3 and 1 g/L pure olive oil soap as a wetting agent (The Natural Olive Oil Soap Factory). Degumming was done in a laboratory dyeing machine (Ahiba Nuance) with a material (g) to Liquor (mL) ratio of 1:50. Treatment was for 30 min at 980C. Degumming for B. mori was performed once while it was performed twice for non B. mori silks. Degummed samples were thoroughly rinsed to remove all residual alkali with DI water and dried in a fume hood.

FTIR Spectroscopy

The FTIR Spectra of each silk were recorded using a PerkinElmer FTIR spectrum 100 with a diamond ATR accessory. The spectra were collected with a spectral window of 650−4000 cm−1, a resolution of 4 cm−1, and 32 scans. Baseline correction was performed in spectrum10 and deconvolution of the amide I peak was done using decomposition tool within DMfit190.

Solid-State NMR (ssNMR) Spectroscopy

The 13C solid state NMR experiments were measured on a Bruker Advance III spectrometer, with a 7 Tesla superconducting magnet, operating at frequencies of 300 MHz and 75 MHz for the 1H and 13C nuclei respectively. The inner third and outer third of a single silk filament, from degummed cocoons, of Bombyx mori, Anthereae assamensis and Samia cynthia ricini weighing ~ 80 mg of sample each, were packed into in a 4mm zirconia MAS rotor with Kel-F cap with a Teflon insert to centre pack the sample within the rotor. The rotors were spun at speeds of 6.5 KHz to 12 KHz at the magic angle (magic angle spinning-MAS). The 13C spectra were recorded using 13C cross-polarization magic-angle spin (CPMAS) ssNMR at 6.5 KHz MAS with the total suppression of spinning sidebands (TOSS) scheme incorporated (CPTOSS). 1H decoupling was provided at 80 KHz using the SPINAL 64 decoupling scheme. A recycle delay of 3 s and 2048 transients were acquired to provide sufficient signal to noise. The 2D 13C{1H} Heteronuclear–Correlation spectra were acquired with Frequency-Switch Lee -Goldberg (FSLG) scheme with a field strength of 86 kHz, during the 1H evolution time for homonuclear decoupling. A short cross polarization time of 0.15 ms at 12 kHZ MAS was used to probe the short range (<0.5 nm) 13C-1H interactions. The 2D spectra were acquired with 64 t1 increments of 68.25 μs each, 512 transients, and using a recycle delay of 3 s that resulted in an experimental time of 28 h. The 13C chemical shifts were referenced to Tetramethylsilane (TMS) using adamantane as a secondary reference.

Atomic force microscopy

The AFM sample was prepared by depositing the thin section (after microtome) onto a silicon substrate. The substrate was previously fixed onto a steel AFM sample holder using silver paste, then it was dried inside a fumehood for at least one hours before it is further cured overnight. The AFM measurements were performed using the Bruker Dimension SPM ICON using tapping mode. Firstly, the probe was tuned at its resonance frequency with 30-35nm free air oscillation amplitude. A small offset was made to the left side of the resonance curve. The scan size was set to 40um for an overall view of the sample, with smaller 4 and 10 microns for more detailed measurements. The pixel resolution was to 512 for all scans. However, the scan rate was varied between 0.25 to 0.5Hz depending on the specimen, with more porous sections scanned at lower speed to allow better track of the pores. The amplitude set point is adjusted accordingly as to avoid excessive tapping force on the surface, but also sufficient to track any changes in surface topography.

Pore fraction analysis

The 30µm2 AFM images showing the entire silk cross section were imported into the image analysis program ImageJ and each image was scaled using the scale bar191. The total area of the silk was calculated by tracing the perimeter of the silk which also provides the internal area. The colour threshold selection settings were then used to select the dark pores within the silk from which the area of the pores was generated. The pore fraction was then calculated using:

푉푣 훷 = 푉푇

Where 훷 is the total fractional area taken up by pores, 푉푣 is the total volume of the pores and 푉푇 is the total volume of the silk.

Quasistatic Nano-indentation

For nanoindentation, a cocoon layer before degumming was used where the filament is arranged naturally in a cross laid form. The cocoon piece was put in a mould containing TAAB TLV resin. Sections of cocoons of 100-200 nm were prepared using an ultra-microtome (Leica EM UC6) for nanoindentation tests. Nano-Indentation test were carried out using Hysitron “TI 950 Tribo-Indenter”. The equipment was fitted with a standard three-sided pyramidal (Berkovich) probe and the probe shape (tip area) function was calibrated independently before the test. Indenter was forced into the specimen at 200 μN/s for 5 sec, held at a peak load of 1000 μN (Pmax) for 2 sec and unloaded at 200 μN/s. Force and displacement were recorded during the test. Hysitron’s data analysis software was used to estimate hardness (H). The software uses, Pmax and tip area function for hardness estimation. P H = max A

Where, ‘Pmax’ is 1000 μN and ‘A’ is the contact area. Unloading segments of each indentation were analysed using Hysitron’s data analysis software which follows, Oliver-Pharr model to fit the initial unloading portion (95% - 80%) of the force-displacement curve and extracts the Reduced Modulus (Er).

(S  ) Er = (2 A)

Where ‘S’ is the contact stiffness and ‘A’ is the projected contact area.

Average hardness (H) and reduced modulus (Er) 12 indents on each sample were calculated.

Electron Tomography

Silk samples were embedded as described previously. Thick sections (250 nm) were cut on an UC6 ultramicrotome (Leica Microsystems) placed onto carbon coated 200 mesh copper grids (Ted Pella Inc., USA). Grids were incubated with 10 nm gold fiducial markers in solution for 5 minutes, washed twice in water then carbon coated. Single axis tilt series were acquired on a 200kV Talos Arctica (ThermoFisher, USA) from -60o to +60o at 2o increments operated at room temperature. Images were acquired with a 4K x 4K Falcon 3 camera at a binning of 1 under the control of Tomography (ThermoFisher) software. Tilt series were reconstructed using weighted back-projection in IMOD (Kremer 1996). Three-dimensional modelling was performed with the Isosurface Render program in IMOD.

Heat flow simulations

To simulate heat flow through porous and non-porous silk I used the two- dimensional heat flow simulation program Energy2D192. The simulations were set up such that two regions with the thermal conductivity, specific heat and density of air (0.0026 W/m·°C, 718 J/Kg·°C and 1.17Kg/m3 respectively) were joined by a section with a fixed area that had the thermal conductivity, specific heat and density of silk (0.3 W/m·°C, 1000 J/Kg·°C and 1350 Kg/m3 respectively). The left region of air represented the external environment and was set to a consistent temperature of 35°C simulating a normal a hot day in the regions these wild silk moths are found. The right region of air represents the internal space within the cocoon and was allowed to fluctuate starting from 0°C. The simulation of the silk involved three alternatives, firstly silk with no cavities (voidless) , then a simulation of a transverse section that resembled the cavitation seen in the wild silk AFM cross-sections and a third simulation of a longitudinal section resembling the cavitation seen in the electron tomography along the silk fibers axis. The simulation was run for 2hrs until the “voidless” silk approached 30°C.

Results

1D 13C CPMAS NMR and FTIR Spectroscopy

Both the 13C CPMAS and FTIR spectra show no notable difference between the inner and outer silks portions of the silk (Figure 14 & 15). Both analysis exhibit well documented differences between the three silks with regard nuances in crystallinity, the chemical shift assignments for the CPMAS are listed in Table 3.

2D 13C–1H HETCOR NMR

The short contact time 13C-1H HETCOR spectra reveal significant differences in alanine bonding environments between each species. The wild silks of A. assamensis and S. c. ricini show distinct

AlaC/GlyH cross correlation peak at δC20.79 /δH3.78 ppm and δC20.99 /δH4.00ppm respectively this is not seen within B. mori (Figure 16). This indicates a close inter-chain association (< ~0.3nm) between glycine and alanine within the wild type silks and not the domestic B. mori. Moreover each silk displays unique AlaCa/AlaNH chemical shifts due to the influence of the NH-O hydrogen bonding environments within the crystalline region; the relevant AlaNH chemical shifts are shown in Figure 16b along with the hydrogen bond lengths calculated as per Holland et al.,2013159. B. mori has on average the shortest h-bond lengths (~1.81Å) associated with in its crystalline region followed by S. c. ricini (~1.825Å) then A. assamensis which displays two distinct H-bonding environments the majority having the longest h- bond lengths (1.908Å) and a second less abundant form having the shortest associated h-bond length (1.613Å) of the three species.

Figure 14. FTIR spectra showing deconvolution Amide I region of inner and outer silkworm silks of B. mori, A. assamensis and S.c. ricini. Summary of the deconvolution results provided (right).

Figure 15. 13C CPMAS spectra of the inner and outer silks across the three silkworm species, B. mori, A. assamensis and S. c. ricini. Major amino acid residues listed above their respective intensities. Inner silk shown in red and outer silk shown in blue.

Table 3. Summary of the amino acid residue chemical shifts from the 13C CPMAS spectra. Blank spaces indicate chemical shifts that are unresolvable due to their convolution with other intensities.

Figure 16. 1H-13C HETCORs for each silk. a) spectra from 10-70ppm showing major hetero/ homo-nuclear correlation peaks. b) expansion of a focusing on the Alanine and Glycine NH chemical shifts highly sensitive to NH-O H-bonding arrangements. Calculated average H-bond length displayed for each major intensity.

Atomic Force Microscopy and Pore fraction analysis

The atomic force microscopy of transverse microtomed sections of inner and outer silk reveals a significant number of cavities within both the inner and outer sections of wild silkworm silks (Figure 17). The domestic silk also has cavities however they appear to be considerably smaller and fewer in number. The cavities ranged in size from tens of nm2 up to ~2.2 um2, the pore fraction was calculated for each microtomed section and is displayed is Table 4 along with the total area of the silk (including the pores).

Figure 17. transverse cross-sectional view of microtomed silkworm silk sections displaying differences in the distribution and number of pores between inner and outer silks as well as b/w species. 57

Table 4. summary of the total area and pore fraction generated from the imageJ analysis on the inner and out AFM silkworm silk cross-sections.

Nano-indentation

Nano-indentation performed on the wild and domestic silk reveals no significant difference in modulus or hardness between the inner and outer silks. B. mori was found to have overall the highest hardness however showed no significant difference to the wild silk in reduced modulus (Figure 18).

Figure 18. a) Example loading unloading curves used to calculate results of quasistatic nanoindentation tests run on inner and outer silk sections of B. mori, A. assamensis and S.c. ricini. b) Average reduced modulus (shown in light blue) and hardness (shown in dark blue) for each species silk.

Electron Tomography

The existence of these cavities is confirmed via electron tomography and is therefore not a possible artefact of the microtoming process used to make the transverse section of the silk. The electron tomography also shows in greater detail the morphology of these cavities in longitudinal view and reveals the presence of many nano-fibril bridges that span the cavities throughout the volume of silk (Figure 19).

Figure 19. Electron tomography of A. assamensis silk reveals fine ultrastructure a) an optical slice through the central region of the reconstructed tomogram of the muga silk highlighting the abundance of large pores observed by AFM. Scale bar = 200 nm. b) Numerous nano-fibril bridges were observed spanning across pores throughout the volume highlighted with arrows . Scale bar 100 nm. c) A magnified 3D rendering of the region highlighted in b. Scale bar = 50 nm.

2D Heat flow simulations

A 2D heat flow simulations ran on Energy2D reveals that the presence of pores within the silk impedes the movement of heat through the silk. As can be seen in Figure 20 the presence of pores in the transverse (b) and longitudinal section simulations (c), like what is seen in the wild A. assamensis and S. c. ricini silks, causes a reduction of overall thermal conductivity resulting in a ~2°C reduction in the internal air temperature over a 2 hour time period when exposed to a consistent 35°C ambient air temperature externally. Whereas the voidless simulation similar to B. mori silk allowed the flow of heat more readily.

Figure 20. Heat flow simulations and results run to simulate heat transfer in porous versus nonporous silk where a nonporous silk, b transverse cross-sectional simulation of silk like that seen

in the AFM and c is a longitudinal simulation like that seen in the electron tomography. “External air” is a region simulating air held at a constant temperature of 35°C, the “Internal air” simulates the internal environment of the cocoon, the thermometers in this region measure the temperature fluctuations and displayed it graphically (right).

Discussion

Silk fibre chemistry (CPMAS and FTIR)

The 13C CP-MAS and FTIR spectra of the three silks reveal that although there are significant differences between B. mori and the wild type silks of S. c. ricini and A. assamensis, there are no major differences in chemical structure between the inner and outer silks of each species. The major differences exhibited between species are well documented and largely explained by the differing use of a poly-

17 alanine (A)n motif in the wild silks versus poly-alanine-glycine (GA)n in B. mori . As a result I find that 13C spectra of B. mori silk has a significantly higher relative abundance of glycine seen in the increase in GlyC and GlyCO when compared to A. assamensis and S. c. ricini (Figure 15 & Table 3). Moreover, this indicates that any variation in material properties along the silk fibre isn’t associated with gross chemical differences along the fibre axis.

Silk fibre chemistry (1H-13C heteronuclear correlations)

To looking more closely into the structure of each silk, I ran 1H-13C HETCORs – a technique that has been relatively overlooked within silk research as there is a preference for 13C-13C experiments taking precedent looking at back-bone dynamics. HETCORs reveal more about the intra-chain and H-bonding interactions within silk that are imperative for silk’s amazing material properties, and help characterise structural differences with in silks that may be relevant to observed differences in silk material properties159. This technique is also useful in the examination of sidechain dynamics however due to a lack of sensitivity in solid state NMR experiments little can be concluded about such structure within this study. The HETCORs do however show cross correlation peaks identified between AlaC and GlyH

(A. assamensis δC20.79 /δH3.78 ppm and S. c. ricini δC20.99 /δH4.00ppm) within the silks from wild species but not within B. mori at first this seemed counterintuitive; this is because the short contact time HETCOR looks at relatively short range interactions (~0.3nm) and the two wild silks use the (A)n motif, so no close intra- or inter-chain correlation between glycine and alanine should not be expected within their crystalline region17. The only way this cross-correlation peak can exists is if a portion of the (A)n -sheets of the silk from the two wild species are forming thin (~2 molecular layers thick) lamella with close contact to the glycine rich amorphous region (Figure 21b). Short (A)n runs within the protein back bone are known to differentiate into two conformations within the greater protein structure; short-form and long-form -sheets. In (A)n runs longer than six amino acids, ~20% of the

alanine conform to the short form structure whilst the other 80% create larger crystals (Figure 21a) formed from the long form conformation65, 176. This cross correlation peak is the first time these conformations have been observed using NMR on native silk and may prove to be a useful tool in helping determine the presence and relative abundance of thin lamella -sheets that are otherwise difficult to characterise by other methods such as XRD65.

Figure 21. Simplified molecular model of silk showing the two distinct forms of crystallite detected from the NMR analysis. “a” corresponding to the large long form crystallites and “b” an example of the thin lamellar short form b-sheets.

Further analysis of the HETCORs alanine and glycine NH peaks, which are highly sensitive to hydrogen bonding arrangements159, show a difference in the relative length of the H-bonding interactions, and therefore, the density of H-bonding networks of the silks crystallites between each species. B. mori shows the shortest hydrogen bonding with both the alanine and glycine NH peaks exhibiting a H of 9.13ppm, indicating the majority of the residues have an NH – O hydrogen bond length of 1.81Å. This was closely followed by S. c. ricini with an alanine NH H shift of 8.9ppm, indicating an average hydrogen bond length of 1.825Å. The majority of A. assamensis alanine NH falls lower at a NH H shift of 8.54ppm with a second further down field shift at 11.0ppm; this indicates that the majority of the

NH-O hydrogen bonding lengths are 1.908Å with a smaller portion forming much tighter H-bonding at 1.61Å.

The NMR results therefore indicate that B. mori silk has the largest densest crystallites formed by the

(GA)n motif, followed by S. c. ricini which has both the previously discussed large long form (A)n crystals and thin short form -sheets . A. assamensis has the longest average H-bonding length associated with its long form crystallites, but potentially shorter lengths within its short form -sheets which have previously been shown to be able to become as short as 1.38Å176. These findings are consistent with the literature in regard to proposed structural differences that result in the characteristic material properties of these silks17, 182. I.e. B. mori silk’s higher crystallinity/larger crystallites results in it being stiffer/harder than the other wild silks, whereas the smaller crystallites facilitate the strain-hardening effect observed in the wild silks. Moreover, this variation in crystallites formed not only helps explain differences in the material properties observed between the silk of each species, but will become relevant in the discussion of cavity formation (below) as the size and organisation of crystallites in polymers has known effects on the process of cavitation during fibre formation.

Characterisation of cavitation

Although the above chemical analysis can explain broad differences in material properties between species it can’t explain the intraspecific and individual variation observed in silkworm silk material properties that I aimed to investigate, hence the use of AFM. The AFM of microtomed silk sections revealed a significant level of cavitation within the wild type silks and substantial variation both between each species and between the inner and outer section within A. assamensis. The cavities ranged in size from a few nm2 up to ~2.2 um2. B. mori had the lowest pore fraction (Φ = 0.008) followed by S. c. ricini which had ~3x the pore fraction (Φ = 0.035, 0.038; inner and outer respectively) then A. assamensis which had ~6x the pore fraction in the inner silk and ~13x in the outer (Φ = 0.066, 0.136 inner and outer respectively).

The cavities were confirmed via Electron tomography, which provides a detailed look at the pores longitudinally and reveals the presence of free nano-fibrils bridging across the cavities. These bridges are characteristic of cavitation within polymers as nano-fibrils that aren’t nicely parallel and that would otherwise be held closely together are pulled apart and stretched across the cavity during its nucleation. That being said, this kind of cavitation would be expected more so in polymers that are

formed from spherulites and not necessarily silkworm silk having a more complex protein-based nano- fibrillar origin.

Origin of cavitation in silkworm silk

The minor cavitation within B. mori has previously been attributed to the reduction of free volume and dehydration as the silk fibroin goes from a liquid phase into a solid along the silk duct193. Greater changes in free volume are thought to result in higher amounts of cavitation within polymers194. Looking at the NMR produced within this study and the NMR and XRD in other studies B. mori is likely to undergo the greatest reduction in free volume as it has the largest densest crystals and relatively smallest amorphous region, followed by S. c. ricini then A. assamensis. This then appears to be the exact opposite expected relationship to what is observed within the AFM with regard to levels of cavitation. Moreover, it cannot readily explain the variation observed between A. assamensis inner and outer silk as I have proven that the silk is chemically homogenous along the fibre axis. However, if the crystals are majorly formed within the fibroin prior to fibre hardening it may be the reduction in free volume of the amorphous region that has a more important role in establishing these voids during fibre hardening.

In early and relatively overlooked work looking at voids within wild silk, the major voids were thought to stem from vacuoles produced within the liquid fibroin within the silk gland195. Our results help support this idea, however the presence of so many nano-fbrils branching across the voids suggest that these vacuoles within the liquid fibroin may also serve as a nucleation point for further elongation of the voids during fibre elongation. The fact that such stark differences in pore fraction within and between species are observed also supports this second idea that vacuoles are responsible for producing these voids. In contrast, if it was a result of the chemistry of the silk as a bio-polymer we, would expect more consistent levels of void formation particularly with regard to A. assamensis inner and outer sections which show transposable 13C NMR spectra.

Effect of Cavities on silk material properties

A major consequence of the pores is their impact on cross sectional area and the calculations used for quasistatic tensile testing of the fibres. The cross-sectional area of the silk is used to scale the stress strain curves generated from tensile testing and has an impact on the relative stress and toughness values produced during the tests. In the case of A. assamensis the level of cavitation and variation in

pore fraction is so high it would significantly impact the material property values produced for the silk. Moreover, the relative structural impact of the voids themselves has to be considered. The size and distribution of cavities can have significant impacts on how the silk performs, if distributed correctly the presence of cavities can improve the strain hardening effects observed within the wild silks. However, voids can also serve as nucleation points for further cavitation that ultimately lead to macrostructural failure and silk fibre breakage. Ultimately this means that these cavities are likely a major source of the material property variation that I observe when running quasistatic tensile tests. This is further supported by the fact that the nano-indentation is consistent between the inner and outer portions of the silk (Figure 18), meaning that at the highly local level, there is no difference in silk material properties (which also nicely reflects what is expected from the chemical analysis). On the large scale, however, morphological factors are playing a more significant role in influencing “bulk” properties.

Possible biological role for cavitation

The biological nature of this cavitation being formed form vacuoles may suggest an adaptive response to temperature as opposed to an intrinsic property of the silk as a biopolymer. Excessively high environmental temperatures are known to significantly impact moth fecundity as they approach adulthood196; Moreover, temperature effects on pupae are known to have significant morphological effects on adult phenotype197. So it stands to reason that increasing the insulative properties of the cocoon would serve to impede the impacts of significant changes in environmental temperature198. This is supported by the fact that void formation is minimal in B. mori, a domesticated species which has been sheltered from relatively harsh conditions for over 5000 years, whilst void formation is extensive and wide spread in wild species 199-200. To test this Iran a heat flow simulations of silk in high environmental temperatures shows that showed a ~2°C reduction in temperature increase when cavities are present which would likely be additive across the scale of a whole cocoon (Figure 20). This significantly supports the idea that the cavitation may serve a thermoregulatory purpose. Moreover, this also helps explain why cavitation has not been removed by natural selection within wild silks despite its negative impact on the material properties of the silk. Further in-depth research is needed to determine this conclusively as there is a significant phylogenetic divide between the porous wild silks (Saturniidae) and that of B. mori (Bombycidae) that may also tie in with the observed differences in pore fraction.

Conclusion

Ultimately, my results further our understanding of the observed variation in the material properties of silk between the inner and outer portion of the cocoon. With local material property measurements using AFM and the NMR displaying no discernible difference between inner and outer silks, whilst the bulk tensile property measurements exhibiting significant variation, it is clear the observed differences relate to the silks macro structure as opposed to significant differences in molecular structure. Moreover, though flaws produced during spinning likely play a significant role in the propensity of the silk fibre to fracture, the pore fraction and distribution are an easily observed confounding factor that, in the future, can and should be controlled for via direct imaging via SEM or AFM. More interestingly my results open up a new line of research into the potential degree of control silkworms may have in the degree of porosity with their silk and any resulting material property trade off. Overall, this research highlights the importance of continuing to further our understanding of the biological influences affecting biomaterials such as spider and silkworm silk and not just focusing on the structure function relationship, which has become the major focus within the field of biological material science.

Chapter 6

Conclusions References

1. Vollrath, F.; Madsen, B.; Shao, Z. Z., The effect of spinning conditions on the mechanics of a spider's dragline silk. P Roy Soc B-Biol Sci 2001, 268 (1483), 2339-2346. 2. Yarger, J. L.; Cherry, B. R.; van der Vaart, A., Uncovering the structure-function relationship in spider silk. Nature Reviews Materials 2018, 3 (3). 3. Barber, A. H.; Lu, D.; Pugno, N. M., Extreme strength observed in limpet teeth. J R Soc Interface 2015, 12 (105). 4. Meyers, M. A.; Lin, A. Y.-M.; Chen, P.-Y.; Muyco, J., Mechanical strength of abalone nacre: Role of the soft organic layer. Journal of the Mechanical Behavior of Biomedical Materials 2008, 1 (1), 76-85. 5. Lintz, E. S.; Scheibel, T. R., Dragline, Egg Stalk and Byssus: A Comparison of Outstanding Protein Fibers and Their Potential for Developing New Materials. Advanced Functional Materials 2013, 23 (36), 4467-4482. 6. Sutherland, T. D.; Campbell, P. M.; Weisman, S.; Trueman, H. E.; Sriskantha, A.; Wanjura, W. J.; Haritos, V. S., A highly divergent gene cluster in honey bees encodes a novel silk family. Genome research 2006, 16 (11), 000-000. 7. Kronenberger, K.; Dicko, C.; Vollrath, F., A novel marine silk. Naturwissenschaften 2012, 99 (1), 3-10. 8. Sutherland, T. D.; Young, J. H.; Weisman, S.; Hayashi, C. Y.; Merritt, D. J., Insect silk: one name, many materials. Annual review of entomology 2010, 55, 171-188. 9. Paradis, E.; Claude, J.; Strimmer, K., APE: analyses of phylogenetics and evolution in R language. Bioinformatics 2004, 20 (2), 289-290. 10. Sehnal, F.; Craig, C., Chapter 235 - Silk Production. In Encyclopedia of Insects (Second Edition), Resh, V. H.; Cardé, R. T., Eds. Academic Press: San Diego, 2009; pp 921-924. 11. Richards, A. l. M. In Observations on the New Zealand glow-worm Arachnocampa luminosa (Skuse) 1890, Transactions of the Royal Society of New Zealand, 1960; pp 559-574. 12. Hansma, H. G., Atomic Force Microscopy and Spectroscopy of Silk from Spider Draglines, Capture-Web Spirals, and Silkworms. In Biotechnology of Silk, Asakura, T.; Miller, T., Eds. Springer Netherlands: Dordrecht, 2014; pp 123-136. 13. Keten, S.; Buehler, M. J., Nanostructure and molecular mechanics of spider dragline silk protein assemblies. J R Soc Interface 2010, 7 (53), 1709-21. 14. Tashiro, Y.; Morimoto, T.; Matsuura, S.; Nagata, S., Studies on the Posterior silk gland of the silkworm, Bombyx mori: I. Growth of posterior silk gland cells and biosynthesis of fibroin during the fifth larval instar. The Journal of cell biology 1968, 38 (3), 574-588. 15. Vollrath, F.; Knight, D. P., Liquid crystalline spinning of spider silk. Nature 2001, 410 (6828), 541. 16. Jin, H.-J.; Kaplan, D. L., Mechanism of silk processing in insects and spiders. Nature 2003, 424 (6952), 1057. 17. Guo, C.; Zhang, J.; Jordan, J. S.; Wang, X.; Henning, R. W.; Yarger, J. L., Structural Comparison of Various Silkworm Silks: An Insight into the Structure–Property Relationship. Biomacromolecules 2018, 19 (3), 906-917.

18. Gosline, J.; Guerette, P.; Ortlepp, C.; Savage, K., The mechanical design of spider silks: from fibroin sequence to mechanical function. J Exp Biol 1999, 202 (23), 3295-3303. 19. Kluge, J. A.; Rabotyagova, U.; Leisk, G. G.; Kaplan, D. L., Spider silks and their applications. Trends in Biotechnology 2008, 26 (5), 244-251. 20. Vasanthavada, K.; Hu, X.; Falick, A. M.; La Mattina, C.; Moore, A. M.; Jones, P. R.; Yee, R.; Reza, R.; Tuton, T.; Vierra, C., Aciniform spidroin, a constituent of egg case sacs and wrapping silk fibers from the black widow spider Latrodectus hesperus. Journal of Biological Chemistry 2007, 282 (48), 35088-35097. 21. Kullmann, E. J., The convergent development of orb-webs in cribellate and ecribellate spiders. American Zoologist 1972, 12 (3), 395-405. 22. Kovoor, J.; Zylberberg, L., Fine structural aspects of silk secretion in a spider (Araneus diadematus). I. Elaboration in the pyriform glands. Tissue and Cell 1980, 12 (3), 547-556. 23. Xu, M.; Lewis, R. V., Structure of a protein superfiber: spider dragline silk. Proceedings of the National Academy of Sciences 1990, 87 (18), 7120-7124. 24. Hayashi, C. Y.; Shipley, N. H.; Lewis, R. V., Hypotheses that correlate the sequence, structure, and mechanical properties of spider silk proteins. International Journal of Biological Macromolecules 1999, 24 (2-3), 271-275. 25. Pérez-Rigueiro, J.; Elices, M.; Plaza, G.; Real, J.; Guinea, G., The effect of spinning forces on spider silk properties. J. Exp. Biol. 2005, 208 (14), 2633-2639. 26. Vollrath, F., Biology of spider silk. International Journal of Biological Macromolecules 1999, 24 (2-3), 81-88. 27. Boutry, C.; Blackledge, T. A., Biomechanical variation of silk links spinning plasticity to spider web function. Zoology 2009, 112 (6), 451-460. 28. Madsen, B.; Shao, Z. Z.; Vollrath, F., Variability in the mechanical properties of spider silks on three levels: interspecific, intraspecific and intraindividual. International journal of biological macromolecules 1999, 24 (2-3), 301-306. 29. Thamm, C.; Scheibel, T., Recombinant Production, Characterization, and Fiber Spinning of an Engineered Short Major Ampullate Spidroin (MaSp1s). Biomacromolecules 2017, 18 (4), 1365-1372. 30. Rising, A.; Johansson, J.; Larson, G.; Bongcam‐Rudloff, E.; Engström, W.; Hjälm, G., Major ampullate spidroins from Euprosthenops australis: multiplicity at protein, mRNA and gene levels. Insect molecular biology 2007, 16 (5), 551-561. 31. Tanaka, K.; Kajiyama, N.; Ishikura, K.; Waga, S.; Kikuchi, A.; Ohtomo, K.; Takagi, T.; Mizuno, S., Determination of the site of disulfide linkage between heavy and light chains of silk fibroin produced by Bombyx mori. Biochimica et Biophysica Acta (BBA)-Protein Structure and Molecular Enzymology 1999, 1432 (1), 92-103. 32. He, Y.-X.; Zhang, N.-N.; Li, W.-F.; Jia, N.; Chen, B.-Y.; Zhou, K.; Zhang, J.; Chen, Y.; Zhou, C.-Z., N-terminal domain of Bombyx mori fibroin mediates the assembly of silk in response to pH decrease. Journal of molecular biology 2012, 418 (3-4), 197-207. 33. Lefèvre, T.; Rousseau, M.-E.; Pézolet, M., Protein secondary structure and orientation in silk as revealed by Raman spectromicroscopy. Biophysical journal 2007, 92 (8), 2885-2895. 34. Meyers, M. A.; McKittrick, J.; Chen, P.-Y., Structural Biological Materials: Critical Mechanics- Materials Connections. Science 2013, 339 (6121), 773-779. 35. Termonia, Y., Molecular modeling of spider silk elasticity. Macromolecules 1994, 27 (25), 7378-7381. 36. Xiao, S.; Xiao, S.; Gräter, F., Dissecting the structural determinants for the difference in mechanical stability of silk and amyloid beta-sheet stacks. Physical Chemistry Chemical Physics 2013, 15 (22), 8765-8771. 37. Shi, X.; Yarger, J. L.; Holland, G. P., Elucidating proline dynamics in spider dragline silk fibre using 2H-13C HETCOR MAS NMR. Chem Commun 2014, 50 (37), 4856-4859. 38. Asakura, T.; Kuzuhara, A.; Tabeta, R.; Saito, H., Conformational characterization of Bombyx mori silk fibroin in the solid state by high-frequency carbon-13 cross polarization-magic angle

spinning NMR, x-ray diffraction, and infrared spectroscopy. Macromolecules 1985, 18 (10), 1841- 1845. 39. Du, N.; Yang, Z.; Liu, X. Y.; Li, Y.; Xu, H. Y., Structural origin of the strain‐hardening of spider silk. Advanced Functional Materials 2011, 21 (4), 772-778. 40. Van Dommelen, J. v.; Parks, D.; Boyce, M.; Brekelmans, W.; Baaijens, F., Micromechanical modeling of the elasto-viscoplastic behavior of semi-crystalline polymers. Journal of the Mechanics and Physics of Solids 2003, 51 (3), 519-541. 41. Blamires, S. J.; Wu, C. C.; Wu, C. L.; Sheu, H. S.; Tso, I. M., Uncovering spider silk nanocrystalline variations that facilitate wind-induced mechanical property changes. Biomacromolecules 2013, 14 (10), 3484-90. 42. Xiao, S.; Stacklies, W.; Cetinkaya, M.; Markert, B.; Gräter, F., Mechanical response of silk crystalline units from force-distribution analysis. Biophysical journal 2009, 96 (10), 3997-4005. 43. Cetinkaya, M.; Xiao, S.; Gräter, F., Effects of crystalline subunit size on silk fiber mechanics. Soft Matter 2011, 7 (18), 8142-8148. 44. Becker, N.; Oroudjev, E.; Mutz, S.; Cleveland, J. P.; Hansma, P. K.; Hayashi, C. Y.; Makarov, D. E.; Hansma, H. G., Molecular nanosprings in spider capture-silk threads. Nature Materials 2003, 2, 278. 45. Jenkins, J. E.; Creager, M. S.; Butler, E. B.; Lewis, R. V.; Yarger, J. L.; Holland, G. P., Solid-state NMR evidence for elastin-like β-turn structure in spider dragline silk. Chem Commun 2010, 46 (36), 6714-6716. 46. Gray, G. M.; van der Vaart, A.; Guo, C.; Jones, J.; Onofrei, D.; Cherry, B. R.; Lewis, R. V.; Yarger, J. L.; Holland, G. P., Secondary Structure Adopted by the Gly-Gly-X Repetitive Regions of Dragline Spider Silk. Int J Mol Sci 2016, 17 (12). 47. Blackledge, T. A.; Perez-Rigueiro, J.; Plaza, G. R.; Perea, B.; Navarro, A.; Guinea, G. V.; Elices, M., Sequential origin in the high performance properties of orb spider dragline silk. Sci Rep-Uk 2012, 2. 48. Blamires, S. J.; Wu, C. L.; Blackledge, T. A.; Tso, I. M., Post-secretion processing influences spider silk performance. J R Soc Interface 2012, 9 (75), 2479-87. 49. Greving, I.; Cai, M. Z.; Vollrath, F.; Schniepp, H. C., Shear-Induced Self-Assembly of Native Silk Proteins into Fibrils Studied by Atomic Force Microscopy. Biomacromolecules 2012, 13 (3), 676-682. 50. Blamires, S. J.; Wu, C. L.; Tso, I. M., Variation in protein intake induces variation in spider silk expression. Plos One 2012, 7 (2), e31626. 51. Blamires, S. J.; Wu, C.-L.; Blackledge, T. A.; Tso, I.-M., Environmentally induced post-spin property changes in spider silks: influences of web type, spidroin composition and ecology. Biological journal of the Linnean Society 2012, 106 (3), 580-588. 52. Wolff, J. O.; Wells, D.; Reid, C. R.; Blamires, S. J., Clarity of objectives and working principles enhances the success of biomimetic programs. Bioinspiration & biomimetics 2017, 12 (5), 051001. 53. Chow, W. Y.; Li, R.; Goldberga, I.; Reid, D. G.; Rajan, R.; Clark, J.; Oschkinat, H.; Duer, M. J.; Hayward, R.; Shanahan, C. M., Essential but sparse collagen hydroxylysyl post-translational modifications detected by DNP NMR. Chemical Communications 2018, 54 (89), 12570-12573. 54. Sergeyev, I. V.; Itin, B.; Rogawski, R.; Day, L. A.; McDermott, A. E., Efficient assignment and NMR analysis of an intact virus using sequential side-chain correlations and DNP sensitization. Proceedings of the National Academy of Sciences 2017, 201701484. 55. Blackledge, T. A.; Hayashi, C. Y., Silken toolkits: biomechanics of silk fibers spun by the orb web spider Argiope argentata (Fabricius 1775). J Exp Biol 2006, 209 (13), 2452-2461. 56. Agnarsson, I.; Kuntner, M.; Blackledge, T. A., Bioprospecting Finds the Toughest Biological Material: Extraordinary Silk from a Giant Riverine Orb Spider. PLoS One 2010, 5 (9). 57. Yao, D.; Liu, H.; Fan, Y., Silk scaffolds for musculoskeletal tissue engineering. Experimental biology and medicine (Maywood, N.J.) 2016, 241 (3), 238-245. 58. Davies, G. J. G.; Knight, D. P.; Vollrath, F., Structure and function of the major ampullate spinning duct of the golden orb weaver, Nephila edulis. Tissue & Cell 2013, 45 (5), 306-311.

59. Blackledge, T. A., Spider silk: a brief review and prospectus on research linking biomechanics and ecology in draglines and orb webs. Journal of Arachnology 2012, 40 (1), 1-12. 60. Cranford, S. W., Increasing silk fibre strength through heterogeneity of bundled fibrils. Journal of the Royal Society Interface 2013, 10 (82). 61. Holland, G. P.; Lewis, R. V.; Yarger, J. L., WISE NMR characterization of nanoscale heterogeneity and mobility in supercontracted Nephila clavipes spider dragline silk. J Am Chem Soc 2004, 126 (18), 5867-72. 62. Jenkins, J. E.; Sampath, S.; Butler, E.; Kim, J.; Henning, R. W.; Holland, G. P.; Yarger, J. L., Characterizing the secondary protein structure of black widow dragline silk using solid-state NMR and X-ray diffraction. Biomacromolecules 2013, 14 (10), 3472-83. 63. Yang, M.; Nakazawa, Y.; Yamauchi, K.; Knight, D.; Asakura, T., Structure of model peptides based on Nephila clavipes dragline silk spidroin (MaSp1) studied by 13C cross polarization/magic angle spinning NMR. Biomacromolecules 2005, 6 (6), 3220-6. 64. Yamaguchi, E.; Yamauchi, K.; Gullion, T.; Asakura, T., Structural analysis of the Gly-rich region in spider dragline silk using stable-isotope labeled sequential model peptides and solid-state NMR. Chem Commun (Camb) 2009, (28), 4176-8. 65. Riekel, C.; Branden, C.; Craig, C.; Ferrero, C.; Heidelbach, F.; Muller, M., Aspects of X-ray diffraction on single spider fibers. Int J Biol Macromol 1999, 24 (2-3), 179-86. 66. Lewis, R. V., Spider Silk - the Unraveling of a Mystery. Accounts Chem Res 1992, 25 (9), 392- 398. 67. Motriuk-Smith, D.; Smith, A.; Hayashi, C. Y.; Lewis, R. V., Analysis of the conserved N- terminal domains in major ampullate spider silk proteins. Biomacromolecules 2005, 6 (6), 3152-9. 68. Rising, A.; Hjalm, G.; Engstrom, W.; Johansson, J., N-terminal nonrepetitive domain common to dragline, flagelliform, and cylindriform spider silk proteins. Biomacromolecules 2006, 7 (11), 3120- 4. 69. Ittah, S.; Barak, N.; Gat, U., A proposed model for dragline spider silk self-assembly: insights from the effect of the repetitive domain size on fiber properties. Biopolymers 2010, 93 (5), 458-68. 70. Lewis, R. V., Spider silk: ancient ideas for new biomaterials. Chem Rev 2006, 106 (9), 3762- 74. 71. Gatesy, J.; Hayashi, C.; Motriuk, D.; Woods, J.; Lewis, R., Extreme diversity, conservation, and convergence of spider silk fibroin sequences. Science 2001, 291 (5513), 2603-5. 72. van Beek, J. D.; Hess, S.; Vollrath, F.; Meier, B. H., The molecular structure of spider dragline silk: folding and orientation of the protein backbone. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (16), 10266-71. 73. Yang, Z. T.; Liivak, O.; Seidel, A.; LaVerde, G.; Zax, D. B.; Jelinski, L. W., Supercontraction and backbone dynamics in spider silk: C-13 and H-2 NMR studies. Journal of the American Chemical Society 2000, 122 (37), 9019-9025. 74. Bittencourt, D.; Dittmar, K.; Lewis, R. V.; Rech, E. L., A MaSp2-like gene found in the Amazon mygalomorph spider Avicularia juruensis. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 2010, 155 (4), 419-426. 75. Starrett, J.; Garb, J. E.; Kuelbs, A.; Azubuike, U. O.; Hayashi, C. Y., Early events in the evolution of spider silk genes. PLoS One 2012, 7 (6), e38084. 76. Liu, Y.; Sponner, A.; Porter, D.; Vollrath, F., Proline and processing of spider silks. Biomacromolecules 2008, 9 (1), 116-21. 77. Boutry, C.; Blackledge, T. A., Evolution of supercontraction in spider silk: structure-function relationship from tarantulas to orb-weavers. J. Exp. Biol. 2010, 213 (20), 3505-3514. 78. Tokareva, O.; Jacobsen, M.; Buehler, M.; Wong, J.; Kaplan, D. L., Structure-function-property- design interplay in biopolymers: Spider silk. Acta Biomaterialia 2014, 10 (4), 1612-1626. 79. Yazawa, K.; Ishida, K.; Masunaga, H.; Hikima, T.; Numata, K., Influence of water content on the β-sheet formation, thermal stability, water removal, and mechanical properties of silk materials. Biomacromolecules 2016, 17 (3), 1057-1066.

80. Blamires, S. J.; Liao, C. P.; Chang, C. K.; Chuang, Y. C.; Wu, C. L.; Blackledge, T. A.; Sheu, H. S.; Tso, I. M., Mechanical performance of spider silk is robust to nutrient-mediated changes in protein composition. Biomacromolecules 2015, 16 (4), 1218-25. 81. Nakagawa, S.; Lagisz, M.; Hector, K. L.; Spencer, H. G., Comparative and meta-analytic insights into life extension via dietary restriction. Aging Cell 2012, 11 (3), 401-9. 82. Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D. G.; Group, P., Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009, 151 (4), 264-9, W64. 83. Revell, L. J.; Harmon, L. J.; Collar, D. C., Phylogenetic signal, evolutionary process, and rate. Syst Biol 2008, 57 (4), 591-601. 84. Bond, J. E.; Garrison, N. L.; Hamilton, C. A.; Godwin, R. L.; Hedin, M.; Agnarsson, I., Phylogenomics resolves a spider backbone phylogeny and rejects a prevailing paradigm for orb web evolution. Curr. Biol. 2014, 24 (15), 1765-71. 85. Wheeler, W. C.; Coddington, J. A.; Crowley, L. M.; Dimitrov, D.; Goloboff, P. A.; Griswold, C. E.; Hormiga, G.; Prendini, L.; Ramírez, M. J.; Sierwald, P.; Almeida-Silva, L.; Alvarez-Padilla, F.; Arnedo, M. A.; Benavides Silva, L. R.; Benjamin, S. P.; Bond, J. E.; Grismado, C. J.; Hasan, E.; Hedin, M.; Izquierdo, M. A.; Labarque, F. M.; Ledford, J.; Lopardo, L.; Maddison, W. P.; Miller, J. A.; Piacentini, L. N.; Platnick, N. I.; Polotow, D.; Silva-Dávila, D.; Scharff, N.; Szűts, T.; Ubick, D.; Vink, C. J.; Wood, H. M.; Zhang, J., The spider tree of life: phylogeny of Araneae based on target-gene analyses from an extensive taxon sampling. Cladistics 2017, 33 (6), 574-616. 86. Garrison, N. L.; Rodriguez, J.; Agnarsson, I.; Coddington, J. A.; Griswold, C. E.; Hamilton, C. A.; Hedin, M.; Kocot, K. M.; Ledford, J. M.; Bond, J. E., Spider phylogenomics: untangling the Spider Tree of Life. PeerJ 2016, 4, e1719. 87. Nakagawa, S.; Freckleton, R. P., Missing inaction: the dangers of ignoring missing data. Trends Ecol Evol 2008, 23 (11), 592-596. 88. Nakagawa, S.; de Villemereuil, P., A general method for simultaneously accounting for phylogenetic and species sampling uncertainty via Rubin's rules in comparative analysis. Syst Biol 2018. 89. Savage, K. N.; Gosline, J. M., The role of proline in the elastic mechanism of hydrated spider silks. J. Exp. Biol. 2008, 211 (12), 1948-1957. 90. Cranford, S. W.; Tarakanova, A.; Pugno, N. M.; Buehler, M. J., Nonlinear material behaviour of spider silk yields robust webs. Nature 2012, 482 (7383), 72-6. 91. Keten, S.; Buehler, M. J., Atomistic model of the spider silk nanostructure. Applied Physics Letters 2010, 96 (15), 153701. 92. dos Santos-Pinto, J. R.; Lamprecht, G.; Chen, W. Q.; Heo, S.; Hardy, J. G.; Priewalder, H.; Scheibel, T. R.; Palma, M. S.; Lubec, G., Structure and post-translational modifications of the web silk protein spidroin-1 from Nephila spiders. J Proteomics 2014, 105, 174-85. 93. Craig, H. C., Blamires, S. J. ,Sani, M., Kasumovic, M. M., Rawal, A., Hook, J. M., DNP NMR Spectroscopy Reveals New Structures, Residues and Interactions in Wild Spider Silks Chemical Communications 2019, (In Review). 94. Kurut, A.; Dicko, C.; Lund, M., Dimerization of Terminal Domains in Spiders Silk Proteins Is Controlled by Electrostatic Anisotropy and Modulated by Hydrophobic Patches. Acs Biomater Sci Eng 2015, 1 (6), 363-371. 95. Elser, J. J.; Sterner, R. W.; Gorokhova, E.; Fagan, W. F.; Markow, T. A.; Cotner, J. B.; Harrison, J. F.; Hobbie, S. E.; Odell, G. M.; Weider, L. J., Biological stoichiometry from genes to ecosystems. Ecol Lett 2000, 3 (6), 540-550. 96. Blamires, S. J.; Lee, Y. H.; Chang, C. M.; Lin, I. T.; Chen, J. A.; Lin, T. Y.; Tso, I. M., Multiple structures interactively influence prey capture efficiency in spider orb webs. Anim. Behav. 2010, 80 (6), 947-953. 97. Schoener, T. W., Theory of feeding strategies. Annual review of ecology and systematics 1971, 2 (1), 369-404.

98. Illius, A. W.; Tolkamp, B. J.; Yearsley, J., The evolution of the control of food intake. Proc Nutr Soc 2002, 61 (4), 465-72. 99. Simpson, S. J.; Sibly, R. M.; Lee, K. P.; Behmer, S. T.; Raubenheimer, D., Optimal foraging when regulating intake of multiple nutrients. Anim. Behav. 2004, 68 (6), 1299-1311. 100. Raubenheimer, D.; Simpson, S. J., Integrative models of nutrient balancing: application to insects and vertebrates. Nutrition research reviews 1997, 10 (1), 151-179. 101. Harrison, S. J.; Raubenheimer, D.; Simpson, S. J.; Godin, J.-G. J.; Bertram, S. M., Towards a synthesis of frameworks in nutritional ecology: interacting effects of protein, carbohydrate and phosphorus on field cricket fitness. Proceedings of the Royal Society of London B: Biological Sciences 2014, 281 (1792), 20140539. 102. Jensen, K.; Mayntz, D.; Toft, S.; Raubenheimer, D.; Simpson, S. J., Nutrient regulation in a predator, the wolf spider Pardosa prativaga. Anim Behav 2011, 81 (5), 993-999. 103. Wilder, S. M.; Rypstra, A. L., Diet quality affects mating behaviour and egg production in a wolf spider. Anim Behav 2008, 76 (2), 439-445. 104. Simpson, S. J.; Raubenheimer, D., The Hungry Locust. In Advances in the Study of Behavior, Slater, P. J. B.; Rosenblatt, J. S.; Snowdon, C. T.; Roper, T. J., Eds. Academic Press: 2000; Vol. 29, pp 1- 44. 105. Blamires, S. J., Plasticity in extended phenotypes: orb web architectural responses to variations in prey parameters. The Journal of Experimental Biology 2010, 213 (18), 3207. 106. Kotiaho, J. S.; Alatalo, R. V.; Mappes, J.; Nielsen, M. G.; Parri, S.; Rivero, A., Energetic costs of size and sexual signalling in a wolf spider. Proceedings of the Royal Society B: Biological Sciences 1998, 265 (1411), 2203-2203. 107. Kawase, H.; Okata, Y.; Ito, K., Role of huge geometric circular structures in the reproduction of a marine pufferfish. Sci Rep 2013, 3, 2106. 108. Borgia, G., Why Do Bowerbirds Build Bowers? American Scientist 1995, 83 (6), 542-547. 109. Blamires, S. J.; Hasemore, M.; Martens, P. J.; Kasumovic, M. M., Diet-induced co-variation between architectural and physicochemical plasticity in an extended phenotype. J. Exp. Biol. 2017, 220 (Pt 5), 876-884. 110. Wilder, S. M., Spider nutrition: an integrative perspective. In Advances in insect physiology, Elsevier: 2011; Vol. 40, pp 87-136. 111. Dong, H.-L.; Zhang, S.-X.; Tao, H.; Chen, Z.-H.; Li, X.; Qiu, J.-F.; Cui, W.-Z.; Sima, Y.-H.; Cui, W.- Z.; Xu, S.-Q., Metabolomics differences between silkworms (Bombyx mori) reared on fresh mulberry (Morus) leaves or artificial diets. Scientific reports 2017, 7 (1), 10972-10972. 112. Craig, C. L., Spiderwebs and Silk: Tracing Evolution From Molecules to Genes to Phenotypes. Oxford University Press: Oxford, U.K., 2003. 113. Venner, S.; Casas, J., Spider webs designed for rare but life-saving catches. Proceedings. Biological sciences 2005, 272 (1572), 1587-1592. 114. Maklakov, A. A.; Bilde, T.; Lubin, Y., Vibratory courtship in a web-building spider: signalling quality or stimulating the female? Anim. Behav. 2003, 66 (4), 623-630. 115. Blackledge, T. A., Signal Conflict in Spider Webs Driven by Predators and Prey. Proceedings: Biological Sciences 1998, 265 (1409), 1991-1996. 116. Sensenig, A. T.; Lorentz, K. A.; Kelly, S. P.; Blackledge, T. A., Spider orb webs rely on radial threads to absorb prey kinetic energy. Journal of the Royal Society, Interface 2012, 9 (73), 1880- 1891. 117. Sensenig, A. T.; Kelly, S. P.; Lorentz, K. A.; Lesher, B.; Blackledge, T. A., Mechanical performance of spider orb webs is tuned for high-speed prey. The Journal of Experimental Biology 2013, 216 (18), 3388. 118. Blamires, S. J.; Chao, Y.-C.; Liao, C.-P.; Tso, I. M., Multiple prey cues induce foraging flexibility in a trap-building predator. Anim. Behav. 2011, 81 (5), 955-961. 119. Blamires, S. J.; Tso, I. M., Nutrient-mediated architectural plasticity of a predatory trap. PLoS One 2013, 8 (1), e54558.

120. Tso, I. M.; Chiang, S. Y.; Blackledge, T. A., Does the giant wood spider Nephila pilipes respond to prey variation by altering web or silk properties? Ethology 2007, 113 (4), 324-333. 121. Blamires, S. J.; Piorkowski, D.; Chuang, A.; Tseng, Y.-H.; Toft, S.; Tso, I.-M., Can differential nutrient extraction explain property variations in a predatory trap? Royal Society open science 2015, 2 (3), 140479. 122. Ayoub, N. A.; Garb, J. E.; Tinghitella, R. M.; Collin, M. A.; Hayashi, C. Y., Blueprint for a high- performance biomaterial: full-length spider dragline silk genes. PLoS One 2007, 2 (6), e514. 123. Nelson, D. L.; Lehninger, A. L.; Cox, M. M., Lehninger principles of biochemistry. Macmillan: 2008. 124. Creager, M. S.; Izdebski, T.; Brooks, A. E.; Lewis, R. V., Elucidating Metabolic Pathways for Amino Acid Incorporation Into Dragline Spider Silk using (13)C Enrichment and Solid State NMR. Comparative biochemistry and physiology. Part A, Molecular & integrative physiology 2011, 159 (3), 219-224. 125. Work, R. W.; Emerson, P. D., An apparatus and technique for the forcible silking of spiders. Journal of Arachnology 1982, 1-10. 126. Guignard, L.; Padilla, A.; Mispelter, J.; Yang, Y. S.; Stern, M. H.; Lhoste, J. M.; Roumestand, C., Backbone dynamics and solution structure refinement of the 15N-labeled human oncogenic protein p13MTCP1: comparison with X-ray data. J Biomol Nmr 2000, 17 (3), 215-30. 127. Takegoshi, K.; Nakamura, S.; Terao, T., 13C–1H dipolar-assisted rotational resonance in magic-angle spinning NMR. Chemical physics letters 2001, 344 (5-6), 631-637. 128. Takegoshi, K.; Nakamura, S.; Terao, T., 13 C–1 H dipolar-driven 13 C–13 C recoupling without 13 C rf irradiation in nuclear magnetic resonance of rotating solids. The Journal of chemical physics 2003, 118 (5), 2325-2341. 129. Thompson, S. N.; Lee, R. W. K., Metabolic fate of alanine in an insect Manduca sexta: effects of starvation and parasitism. Biochimica et Biophysica Acta (BBA) - General Subjects 1993, 1157 (2), 259-269. 130. Finke, M. D., Complete nutrient composition of commercially raised invertebrates used as food for insectivores. Zoo Biology: Published in affiliation with the American Zoo and Aquarium Association 2002, 21 (3), 269-285. 131. Schermerhorn, T., Normal glucose metabolism in carnivores overlaps with diabetes pathology in non-carnivores. Frontiers in endocrinology 2013, 4, 188. 132. Zhang, Y.; Zhao, D.; Meng, Z.; Dong, Z.; Lin, Y.; Chen, S.; Xia, Q.; Zhao, P.; Bendena, B., Wild silkworm cocoon contains more metabolites than domestic silkworm cocoon to improve its protection. Journal of Insect Science 2017, 17 (5). 133. Cohen, S. M., Carbon-13 NMR study of the effects of fasting and diabetes on the metabolism of pyruvate in the tricarboxylic acid cycle and of the utilization of pyruvate and ethanol in lipogenesis in perfused rat liver. Biochemistry 1987, 26 (2), 581-589. 134. Han, Q.; Kim, S. R.; Ding, H.; Li, J., Evolution of two alanine glyoxylate aminotransferases in mosquito. The Biochemical journal 2006, 397 (3), 473-481. 135. Cellini, B.; Bertoldi, M.; Montioli, R.; Paiardini, A.; Voltattorni, C. B., Human wild-type alanine: glyoxylate aminotransferase and its naturally occurring G82E variant: functional properties and physiological implications. Biochemical Journal 2007, 408 (1), 39-50. 136. Solon-Biet, S. M.; Walters, K. A.; Simanainen, U. K.; McMahon, A. C.; Ruohonen, K.; Ballard, J. W. O.; Raubenheimer, D.; Handelsman, D. J.; Le Couteur, D. G.; Simpson, S. J., Macronutrient balance, reproductive function, and lifespan in aging mice. Proceedings of the National Academy of Sciences 2015, 201422041. 137. Simpson, S. J.; Raubenheimer, D., The nature of nutrition: a unifying framework from animal adaptation to human obesity. Princeton university press: 2012. 138. Gührs, K.-H.; Weisshart, K.; Grosse, F., Lessons from nature — protein fibers. Reviews in Molecular Biotechnology 2000, 74 (2), 121-134.

139. Wright, S.; Goodacre, S. L., Evidence for antimicrobial activity associated with common house spider silk. BMC research notes 2012, 5 (1), 326. 140. Heidebrecht, A.; Eisoldt, L.; Diehl, J.; Schmidt, A.; Geffers, M.; Lang, G.; Scheibel, T., Biomimetic Fibers Made of Recombinant Spidroins with the Same Toughness as Natural Spider Silk. Advanced Materials 2015, 27 (13), 2189-2194. 141. Kaplan, D. L., Fibrous proteins—silk as a model system. Polymer Degradation and Stability 1998, 59 (1), 25-32. 142. Fu, C.; Shao, Z.; Fritz, V., Animal silks: their structures, properties and artificial production. Chem Commun 2009, (43), 6515-6529. 143. Hinman, M. B.; Lewis, R. V., Isolation of a clone encoding a second dragline silk fibroin. Nephila clavipes dragline silk is a two-protein fiber. J Biol Chem 1992, 267 (27), 19320-4. 144. Lefèvre, T.; Paquet‐Mercier, F.; Rioux‐Dubé, J. F.; Pézolet, M., Structure of silk by raman spectromicroscopy: From the spinning glands to the fibers. Biopolymers 2012, 97 (6), 322-336. 145. Sampath, S.; Isdebski, T.; Jenkins, J. E.; Ayon, J. V.; Henning, R. W.; Orgel, J. P. R. O.; Antipoa, O.; Yarger, J. L., X-ray diffraction study of nanocrystalline and amorphous structure within major and minor ampullate dragline spider silks. Soft Matter 2012, 8 (25), 6713-6722. 146. Jenkins, J. E.; Creager, M. S.; Lewis, R. V.; Holland, G. P.; Yarger, J. L., Quantitative Correlation Between the Protein Primary Sequences and Secondary Structures in Spider Dragline Silks. Biomacromolecules 2010, 11 (1), 192-200. 147. Asakura, T.; Suzuki, Y.; Nakazawa, Y.; Holland, G. P.; Yarger, J. L., Elucidating silk structure using solid-state NMR. Soft Matter 2013, 9 (48), 11440-11450. 148. Blamires, S. J.; Kasumovic, M. M.; Tso, I. M.; Martens, P. J.; Hook, J. M.; Rawal, A., Evidence of Decoupling Protein Structure from Spidroin Expression in Spider Dragline Silks. International Journal of Molecular Sciences 2016, 17 (8), 1294. 149. Shi, X.; Holland, G. P.; Yarger, J. L., Molecular Dynamics of Spider Dragline Silk Fiber Investigated by 2H MAS NMR. Biomacromolecules 2015, 16 (3), 852-859. 150. Blamires, S. J.; Nobbs, M.; Martens, P. J.; Tso, I. M.; Chuang, W.-T.; Chang, C.-K.; Sheu, H.-S., Multiscale mechanisms of nutritionally induced property variation in spider silks. PLoS One 2018, 13 (2), e0192005. 151. Blamires, S. J.; Blackledge, T. A.; Tso, I.-M., Physicochemical Property Variation in Spider Silk: Ecology, Evolution, and Synthetic Production. Annual Review of Entomology 2017, 62 (1), 443-460. 152. Lilly Thankamony, A. S.; Wittmann, J. J.; Kaushik, M.; Corzilius, B., Dynamic nuclear polarization for sensitivity enhancement in modern solid-state NMR. Prog Nucl Magn Reson Spectrosc 2017, 102-103, 120-195. 153. Ni, Q. Z.; Daviso, E.; Can, T. V.; Markhasin, E.; Jawla, S. K.; Swager, T. M.; Temkin, R. J.; Herzfeld, J.; Griffin, R. G., High frequency dynamic nuclear polarization. Acc Chem Res 2013, 46 (9), 1933-41. 154. Gupta, R.; Lu, M.; Hou, G.; Caporini, M. A.; Rosay, M.; Maas, W.; Struppe, J.; Suiter, C.; Ahn, J.; Byeon, I. J.; Franks, W. T.; Orwick-Rydmark, M.; Bertarello, A.; Oschkinat, H.; Lesage, A.; Pintacuda, G.; Gronenborn, A. M.; Polenova, T., Dynamic Nuclear Polarization Enhanced MAS NMR Spectroscopy for Structural Analysis of HIV-1 Protein Assemblies. J Phys Chem B 2016, 120 (2), 329- 39. 155. Perras, F. A.; Wang, Z. R.; Naik, P.; Slowing, I. I.; Pruski, M., Natural Abundance O-17 DNP NMR Provides Precise O-H Distances and Insights into the Bronsted Acidity of Heterogeneous Catalysts. Angewandte Chemie-International Edition 2017, 56 (31), 9165-9169. 156. Claire, S.; Melanie, R.; Gilles, C.; Fabien, A.; T., W. R.; Olivier, O.; Paul, T., Highly Efficient, Water-Soluble Polarizing Agents for Dynamic Nuclear Polarization at High Frequency. Angewandte Chemie 2013, 125 (41), 11058-11061. 157. Hartmann, S. R.; Hahn, E. L., Nuclear Double Resonance in the Rotating Frame. Physical Review 1962, 128 (5), 2042-2053.

158. Metz, G.; Wu, X. L.; Smith, S. O., Ramped-Amplitude Cross Polarization in Magic-Angle- Spinning NMR. Journal of Magnetic Resonance, Series A 1994, 110 (2), 219-227. 159. Holland, G. P.; Mou, Q.; Yarger, J. L., Determining hydrogen-bond interactions in spider silk with 1H-13C HETCOR fast MAS solid-state NMR and DFT proton chemical shift calculations. Chem Commun (Camb) 2013, 49 (59), 6680-2. 160. Wagner, G.; Pardi, A.; Wuethrich, K., Hydrogen bond length and proton NMR chemical shifts in proteins. Journal of the American Chemical Society 1983, 105 (18), 5948-5949. 161. Fujisawa, R.; Kuboki, Y., High-resolution solid-state nuclear magnetic resonance spectra of dentin collagen. Biochemical and Biophysical Research Communications 1990, 167 (2), 761-766. 162. Goldberga, I.; Li, R.; Duer, M. J., Collagen Structure–Function Relationships from Solid-State NMR Spectroscopy. Accounts of Chemical Research 2018, 51 (7), 1621-1629. 163. Gerecht, K.; Figueiredo, A. M.; Hansen, D. F., Determining rotational dynamics of the guanidino group of arginine side chains in proteins by carbon-detected NMR. Chem Commun (Camb) 2017, 53 (72), 10062-10065. 164. Mackenzie, H. W.; Hansen, D. F., A C-13-detected N-15 double-quantum NMR experiment to probe arginine side-chain guanidinium N-15(eta) chemical shifts. J Biomol Nmr 2017, 69 (3), 123-132. 165. Sahni, V.; Miyoshi, T.; Chen, K.; Jain, D.; Blamires, S. J.; Blackledge, T. A.; Dhinojwala, A., Direct solvation of glycoproteins by salts in spider silk glues enhances adhesion and helps to explain the evolution of modern spider orb webs. Biomacromolecules 2014, 15 (4), 1225-32. 166. Jain, D.; Zhang, C.; Cool, L. R.; Blackledge, T. A.; Wesdemiotis, C.; Miyoshi, T.; Dhinojwala, A., Composition and Function of Spider Glues Maintained During the Evolution of Cobwebs. Biomacromolecules 2015, 16 (10), 3373-80. 167. Weisman, S.; Trueman, H. E.; Mudie, S. T.; Church, J. S.; Sutherland, T. D.; Haritos, V. S., An unlikely silk: the composite material of green lacewing cocoons. Biomacromolecules 2008, 9 (11), 3065-3069. 168. Craig, C. L., Evolution of arthropod silks. Annual review of entomology 1997, 42 (1), 231-267. 169. Asakura, T.; Yao, J., 13C CP/MAS NMR study on structural heterogeneity in Bombyx mori silk fiber and their generation by stretching. Protein Science 2002, 11 (11), 2706-2713. 170. Asakura, T.; Yao, J.; Yamane, T.; Umemura, K.; Ulrich, A. S., Heterogeneous Structure of Silk Fibers from Bombyx M Ori Resolved by 13C Solid-State NMR Spectroscopy. Journal of the American Chemical Society 2002, 124 (30), 8794-8795. 171. Monti, P.; Freddi, G.; Bertoluzza, A.; Kasai, N.; Tsukada, M., Raman spectroscopic studies of silk fibroin from Bombyx mori. Journal of Raman Spectroscopy 1998, 29 (4), 297-304. 172. Asakura, T.; Yamaguchi, T., Proposal of new model for Silk I structure of Bombyx mori silk fibroin. The Journal of Sericultural Science of Japan 1987, 56 (4), 300-304. 173. Asakura, T.; Ashida, J.; Yamane, T.; Kameda, T.; Nakazawa, Y.; Ohgo, K.; Komatsu, K., A repeated β-turn structure in Poly (Ala-Gly) as a model for silk I of Bombyx mori silk fibroin studied with two-dimensional spin-diffusion NMR under off magic angle spinning and rotational echo double resonance1. Journal of Molecular Biology 2001, 306 (2), 291-305. 174. Asakura, T.; Yamane, T.; Nakazawa, Y.; Kameda, T.; Ando, K., Structure of Bombyx mori silk fibroin before spinning in solid state studied with wide angle x‐ray scattering and 13C cross‐ polarization/magic angle spinning NMR. Biopolymers: Original Research on Biomolecules 2001, 58 (5), 521-525. 175. Asakura, T.; Watanabe, Y.; Uchida, A.; Minagawa, H., NMR of silk fibroin. Carbon-13 NMR study of the chain dynamics and solution structure of Bombyx mori silk fibroin. Macromolecules 1984, 17 (5), 1075-1081. 176. Asakura, T.; Okushita, K.; Williamson, M. P., Analysis of the structure of Bombyx mori silk fibroin by NMR. Macromolecules 2015, 48 (8), 2345-2357. 177. Asakura, T.; Ohgo, K.; Komatsu, K.; Kanenari, M.; Okuyama, K., Refinement of repeated β- turn structure for silk I conformation of Bombyx mori silk fibroin using 13C solid-state NMR and X-ray diffraction methods. Macromolecules 2005, 38 (17), 7397-7403.

178. Ulrich, S.; Glišović, A.; Salditt, T.; Zippelius, A., Diffraction from the β-sheet crystallites in spider silk. The European Physical Journal E 2008, 27 (3), 229. 179. Du, N.; Liu, X. Y.; Narayanan, J.; Li, L.; Lim, M. L. M.; Li, D., Design of superior spider silk: from nanostructure to mechanical properties. Biophysical journal 2006, 91 (12), 4528-4535. 180. Keten, S.; Xu, Z.; Ihle, B.; Buehler, M. J., Nanoconfinement controls stiffness, strength and mechanical toughness of β-sheet crystals in silk. Nature Materials 2010, 9, 359. 181. Sezutsu, H.; Yukuhiro, K., The complete nucleotide sequence of the Eri-silkworm (Samia cynthia ricini) fibroin gene. Journal of Insect Biotechnology and Sericology 2014, 83 (3), 3_059-3_070. 182. Colomban, P.; Dinh, H. M., Origin of the variability of the mechanical properties of silk fibres: 2 The nanomechanics of single silkworm and spider fibres. Journal of Raman Spectroscopy 2012, 43 (8), 1035-1041. 183. Malay, A. D.; Sato, R.; Yazawa, K.; Watanabe, H.; Ifuku, N.; Masunaga, H.; Hikima, T.; Guan, J.; Mandal, B. B.; Damrongsakkul, S., Relationships between physical properties and sequence in silkworm silks. Scientific reports 2016, 6, 27573. 184. Rajkhowa, R.; Kaur, J.; Wang, X.; Batchelor, W., Intrinsic tensile properties of cocoon silk fibres can be estimated by removing flaws through repeated tensile tests. Journal of The Royal Society Interface 2015, 12 (107), 20150177. 185. Pawlak, A.; Galeski, A.; Rozanski, A., Cavitation during deformation of semicrystalline polymers. Progress in polymer science 2014, 39 (5), 921-958. 186. Kinloch, A. J., Fracture behaviour of polymers. Springer Science & Business Media: 2013. 187. Rajkhowa, R.; Gupta, V.; Kothari, V., Tensile stress–strain and recovery behavior of Indian silk fibers and their structural dependence. Journal of applied polymer science 2000, 77 (11), 2418-2429. 188. Mortimer, B.; Holland, C.; Vollrath, F., Forced reeling of Bombyx mori silk: separating behavior and processing conditions. Biomacromolecules 2013, 14 (10), 3653-3659. 189. Shao, Z.; Vollrath, F., Materials: Surprising strength of silkworm silk. Nature 2002, 418 (6899), 741. 190. Baranyi, J.; Le Marc, Y., Dmfit manual, version 2.0. Norwich, UK: Institute of Food Research 1996. 191. Abràmoff, M. D.; Magalhães, P. J.; Ram, S. J., Image processing with ImageJ. Biophotonics international 2004, 11 (7), 36-42. 192. Alrwashdeh, S. S., Modelling of Operating Conditions of Conduction Heat Transfer Mode Using Energy 2D Simulation. International Journal of Online Engineering (iJOE) 2018, 14 (09), 200- 207. 193. Robson, R. M., Microvoids in Bombyx mori silk. An electron microscope study. International journal of biological macromolecules 1999, 24 (2-3), 145-150. 194. Rozanski, A.; Galeski, A., Controlling cavitation of semicrystalline polymers during tensile drawing. Macromolecules 2011, 44 (18), 7273-7287. 195. Akai, H.; Nagashima, T.; Aoyagi, S., Ultrastructure of posterior silk gland cells and liquid silk in indian tasar silkworm, Antheraea mylitta Drury (Lepidoptera: Saturniidae). International Journal of Insect Morphology and Embryology 1993, 22 (5), 497-506. 196. Ma, L.; Wang, X.; Liu, Y.; Su, M.-Z.; Huang, G.-H., Temperature effects on development and fecundity of Brachmia macroscopa (Lepidoptera: Gelechiidae). Plos One 2017, 12 (3), e0173065. 197. Stevens, D. J., Pupal development temperature alters adult phenotype in the speckled wood butterfly, Pararge aegeria. Journal of Thermal Biology 2004, 29 (4-5), 205-210. 198. Zhang, W.; Chang, X.-Q.; Hoffmann, A.; Zhang, S.; Ma, C.-S., Impact of hot events at different developmental stages of a moth: the closer to adult stage, the less reproductive output. Scientific reports 2015, 5, 10436. 199. NARUMI, T.; KOBAYASHI, M., A method of morphometry for voids in Saturniidae cocoon filaments using image processing techniques. The Journal of Sericultural Science of Japan 1995, 64 (3), 203-208.

200. Lin, Y.; Liou, T.; Liu, C.; Liu, Y.; Wu, T.; Chang, Y., An Introduction to Taiwan Wild Silkworms: Wild Silkmoths' 92. International Society for Wild Silkmoths, Tsukuba. Japan 1993, 105-114. 201. Barrantes, G.; Eberhard, W. G., Ontogeny repeats phylogeny in Steatoda and Latrodectus spiders. J Arachnol 2010, 485-494. 202. Valtonen, T. M.; Kangassalo, K.; Pölkki, M.; Rantala, M. J., Transgenerational effects of parental larval diet on offspring development time, adult body size and pathogen resistance in Drosophila melanogaster. Plos One 2012, 7 (2), e31611. 203. Lillycrop, K. A., Effect of maternal diet on the epigenome: implications for human metabolic disease. Proceedings of the Nutrition Society 2011, 70 (1), 64-72.

Appendices

Phylogenetic comparative analysis search terms

SEARCH TERMS

• Silk • ((Amino Acids )OR (Protein)) Composition • Mechanical Properties

EXCLUSION TERMS ("NOT” TERMS)

• Gene • Wool • Brain • Tissue • Cell • Medical • Surgery • Suture • Delivery • Enzyme • Human • Scaffold • Electrospun • Nanoparticle • Film • Nano composite • Bio composite • Biomedical • Keratin • Degummed • Graft

List of references used in phylogenetic comparative analysis

Agnarsson I, Boutry C, Blackledge TA. Spider silk aging: Initial improvement in a high performance material followed by slow degradation. Journal of Experimental Zoology Part a-Ecological Genetics and Physiology. 2008;309a(8):494-504. Agnarsson I, Kuntner M, Blackledge TA. Bioprospecting Finds the Toughest Biological Material: Extraordinary Silk from a Giant Riverine Orb Spider. PLoS One. 2010;5(9). Andersen SO. Amino acid composition of spider silks. Comparative Biochemistry And Physiology. 1970;35(3):705-11. Benamu M, Lacava M, Garcia LF, Santana M, Fang J, Wang XG, et al. Nanostructural and mechanical property changes to spider silk as a consequence of insecticide exposure. Chemosphere. 2017;181:241-9. Blackledge TA, Hayashi CY. Silken toolkits: biomechanics of silk fibers spun by the orb web spider Argiope argentata (Fabricius 1775). J Exp Biol. 2006;209(13):2452-61. Blackledge TA, Summers AP, Hayashi CY. Gumfooted lines in black widow cobwebs and the mechanical properties of spider capture silk. Zoology. 2005;108(1):41-6. Blackledge TA, Swindeman JE, Hayashi CY. Quasistatic and continuous dynamic characterization of the mechanical properties of silk from the cobweb of the black widow spider Latrodectus hesperus. J Exp Biol. 2005;208(10):1937-49. Blamires SJ, Kasumovic MM, Tso IM, Martens PJ, Hook JM, Rawal A. Evidence of Decoupling Protein Structure from Spidroin Expression in Spider Dragline Silks. International Journal of Molecular Sciences. 2016;17(8). Blamires SJ, Liao C-P, Chang C-K, Chuang Y-C, Wu C-L, Blackledge TA, et al. Mechanical Performance of Spider Silk Is Robust to Nutrient-Mediated Changes in Protein Composition. Biomacromolecules. 2015;16(4):1218-25. Blamires SJ, Wu C-C, Wu C-L, Sheu H-S, Tso IM. Uncovering Spider Silk Nanocrystalline Variations That Facilitate Wind-Induced Mechanical Property Changes. Biomacromolecules. 2013;14(10):3484-90. Blamires SJ, Wu C-L, Blackledge TA, Tso IM. Post-secretion processing influences spider silk performance. Journal of the Royal Society Interface. 2012;9(75):2479-87. Blamires SJ, Wu C-L, Tso IM. Variation in Protein Intake Induces Variation in Spider Silk Expression. PLoS One. 2012;7(2). Boutry C, Blackledge TA. Biomechanical variation of silk links spinning plasticity to spider web function. Zoology. 2009;112(6):451-60. Boutry C, Blackledge TA. Wet webs work better: humidity, supercontraction and the performance of spider orb webs. J Exp Biol. 2013;216(19):3606-10. Boutry C, Rezac M, Blackledge TA. Plasticity in Major Ampullate Silk Production in Relation to Spider Phylogeny and Ecology. PLoS One. 2011;6(7). C. ZJ. A study of the mechanical behavior of spider silks Natick, MA: U.S. Army NatickLaboratories; 1968 [ Casem ML, Turner D, Houchin K. Protein and amino acid composition of silks from the cob weaver, Latrodectus hesperus (black widow). International Journal of Biological Macromolecules. 1999;24(2-3):103-8. Coddington JA, Chanzy HD, Jackson CL, Raty G, Gardner KH. The unique ribbon morphology of the major ampullate silk of spiders from the genus Loxosceles (recluse spiders). Biomacromolecules. 2002;3(1):5-8. Craig HC, Blamires, S. J. ,Sani, M., Kasumovic, M. M., Rawal, A., Hook, J. M. DNP NMR Spectroscopy Reveals New Structures, Residues and Interactions in Wild Spider Silks Chemical Communications. 2019;(In Review). Creager MS, Jenkins JE, Thagard-Yeaman LA, Brooks AE, Jones JA, Lewis RV, et al. Solid-State NMR Comparison of Various Spiders' Dragline Silk Fiber. Biomacromolecules. 2010;11(8):2039-43. Cunniff PM, Fossey SA, Auerbach MA, Song JW. Mechanical Properties of Major Ampulate Gland Silk Fibers Extracted from Nephila clavipes Spiders. Silk Polymers. ACS Symposium Series. 544: American Chemical Society; 1993. p. 234-51. Denny M. Physical-Properties of Spiders Silk and Their Role in Design of Orb-Webs. J Exp Biol. 1976;65(2):483-506. Dicko C, Knight D, Kenney JM, Vollrath F. Secondary structures and conformational changes in flagelliform, cylindrical, major, and minor ampullate silk proteins. Temperature and concentration effects. Biomacromolecules. 2004;5(6):2105-15.

Elices M, Perez-Rigueiro J, Plaza GR, Guinea GV. Finding inspiration in Argiope trifasciata spider silk fibers. Jom. 2005;57(2):60-6. Fischer FG, Brander J. Eine Analyse der Gespinste der Kreuzspinne. Hoppe-Seyler´s Zeitschrift für physiologische Chemie1960. p. 92. Guehrs KH, Schlott B, Grosse F, Weisshart K. Environmental conditions impinge on dragline silk protein composition. Insect Molecular Biology. 2008;17(5):553-64. Guinea GV, Elices M, Perez-Rigueiro J, Plaza GR. Stretching of supercontracted fibers: a link between spinning and the variability of spider silk. J Exp Biol. 2005;208(1):25-30. Guinea GV, Elices M, Plaza GR, Perea GB, Daza R, Riekel C, et al. Minor Ampullate Silks from Nephila and Argiope Spiders: Tensile Properties and Microstructural Characterization. Biomacromolecules. 2012;13(7):2087-98. Guinea GV, Elices M, Real JI, Gutierrez S, Perez-Rigueiro J. Reproducibility of the tensile properties of spider (Argiope trifasciata) silk obtained by forced silking. Journal of Experimental Zoology Part a-Comparative Experimental Biology. 2005;303A(1):37-44. Hayashi CY, Blackledge TA, Lewis RV. Molecular and mechanical characterization of aciniform silk: Uniformity of iterated sequence modules in a novel member of the spider silk fibroin gene family. Molecular Biology and Evolution. 2004;21(10):1950-9. Hesselberg T, Vollrath F. The mechanical properties of the non-sticky spiral in Nephila orb webs (Araneae, Nephilidae). J Exp Biol. 2012;215(19):3362-9. Hronska M, van Beek JD, Williamson PTF, Vollrath F, Meier BH. NMR characterization of native liquid spider dragline silk from Nephila edulis. Biomacromolecules. 2004;5(3):834-9. Hudspeth M, Nie X, Chen W, Lewis R. Effect of loading rate on mechanical properties and fracture morphology of spider silk. Biomacromolecules. 2012;13(8):2240-6. Jiang P, Guo C, Lv T, Xiao Y, Liao X, Zhou B. Structure, composition and mechanical properties of the silk fibres of the egg case of the Joro spider, Nephila clavata (Araneae, Nephilidae). Journal of Biosciences. 2011;36(5):897-910. Jiang P, Lv T-Y, Xiao Y-H, Wu M-L, Liao X-J, Zhou B, et al. Morphology, fibrous composition and tensile properties of drag-silk produced by two species of orb spider. International Journal of Materials Research. 2011;102(10):1261-9. Köhler T, Vollrath F. Thread biomechanics in the two orb-weaving spiders Araneus diadematus (Araneae, Araneidae) and Uloborus walckenaerius (Araneae, Uloboridae). Journal of Experimental Zoology. 1995;271(1):1-17. Lawrence BA, Vierra CA, Moore AMF. Molecular and mechanical properties of major ampullate silk of the black widow spider, Latrodectus hesperus. Biomacromolecules. 2004;5(3):689-95. Lewis RV. Spider Silk - the Unraveling of a Mystery. Accounts of Chemical Research. 1992;25(9):392-8. Liao C-P, Chi K-J, Tso IM. The effects of wind on trap structural and material properties of a sit-and- wait predator. Behavioral Ecology. 2009;20(6):1194-203. Liivak O, Flores A, Lewis R, Jelinski LW. Conformation of the Polyalanine Repeats in Minor Ampullate Gland Silk of the Spider Nephila clavipes. Macromolecules. 1997;30(23):7127-30. Liu Y, Shao ZZ, Vollrath F. Extended wet-spinning can modify spider silk properties. Chemical Communications. 2005(19):2489-91. Liu Y, Sponner A, Porter D, Vollrath F. Proline and processing of spider silks. Biomacromolecules. 2008;9(1):116-21. Lombardi SJ, Kaplan DL. The Amino-Acid-Composition of Major Ampullate Gland Silk (Dragline) of Nephila-Clavipes (Araneae, Tetragnathidae). Journal of Arachnology. 1990;18(3):297-306. Lucas F, Shaw JTB, Smith SG. Comparative studies of fibroins: I. The amino acid composition of various fibroins and its significance in relation to their crystal structure and taxonomy. Journal of Molecular Biology. 1960;2(6):339-49. Madsen B, Shao ZZ, Vollrath F. Variability in the mechanical properties of spider silks on three levels: interspecific, intraspecific and intraindividual. International Journal of Biological Macromolecules. 1999;24(2- 3):301-6. Madsen B, Vollrath F. Mechanics and morphology of silk drawn from anesthetized spiders. Naturwissenschaften. 2000;87(3):148-53. Michal CA, Jelinski LW. Rotational-echo Double-resonance in Complex Biopolymers: a Study of Nephila Clavipes Dragline Silk. J Biomol Nmr. 1998;12(2):231-41. Moore AM, Tran K. Material properties of cobweb silk from the black widow spider Latrodectus hesperus. Int J Biol Macromol. 1999;24(2-3):277-82.

Peakall DB. Composition, Function and Glandular Origin of the Silk Fibroins of the Spider Araneus Diadematus Cl. J Exp Zool. 1964;156:345-52. Pérez-Rigueiro J, Elices M, Plaza GR, Real JI, Guinea GV. The influence of anaesthesia on the tensile properties of spider silk. J Exp Biol. 2006;209(2):320-6. Perez-Riguero J, Elices M, Llorca J, Viney C. Tensile properties of Argiope trifasciata drag line silk obtained from the spider's web. Journal of Applied Polymer Science. 2001;82(9):2245-51. Savage KN, Gosline JM. The effect of proline on the network structure of major ampullate silks as inferred from their mechanical and optical properties. J Exp Biol. 2008;211(12):1937-47. Sensenig A, Agnarsson I, Blackledge TA. Behavioural and biomaterial coevolution in spider orb webs. J Evol Biol. 2010;23(9):1839-56. Shao ZZ, Vollrath F, Yang Y, Thogersen HC. Structure and behavior of regenerated spider silk. Macromolecules. 2003;36(4):1157-61. Shi X, Holland GP, Yarger JL. Amino acid analysis of spider dragline silk using 1H NMR. Analytical Biochemistry. 2013;440(2):150-7. Swanson BO, Blackledge TA, Beltran J, Hayashi CY. Variation in the material properties of spider dragline silk across species. Applied Physics a-Materials Science & Processing. 2006;82(2):213-8. Swanson BO, Blackledge TA, Summers AP, Hayashi CY. Spider dragline silk: Correlated and mosaic evolution in high-performance biological materials. Evolution. 2006;60(12):2539-51. Tillinghast EK. The Chemical Fractionation of the Orb Web of Argiope Spiders. Insect Biochemistry. 1984;14(1):115-&. Tillinghast EK, Christenson T. Observations on the Chemical-Composition of the Web of Nephila- Clavipes (Araneae, Araneidae). Journal of Arachnology. 1984;12(1):69-74. Tso IM, Chiang S-Y, Blackledge TA. Does the giant wood spider Nephila pilipes respond to prey variation by altering web or silk properties? Ethology. 2007;113(4):324-33. Tso IM, Wu HC, Hwang IR. Giant wood spider Nephila pilipes alters silk protein in response to prey variation. J Exp Biol. 2005;208(6):1053-61. Vollrath F, Köhler T. Mechanics of silk produced by loaded spiders. Proc R Soc Lond B Biol Sci. 1996;263(1369):387-91. Vollrath F, Madsen B, Shao Z. The effect of spinning conditions on the mechanics of a spider's dragline silk. Proceedings of the Royal Society B: Biological Sciences. 2001;268(1483):2339-46. Work RW, Young CT. THE AMINO-ACID COMPOSITIONS OF MAJOR AND MINOR AMPULLATE SILKS OF CERTAIN ORB-WEB-BUILDING SPIDERS (ARANEAE, ARANEIDAE). Journal of Arachnology. 1987;15(1):65-80. Zhao AC, Zhao TF, Nakagaki K, Zhang YS, SiMa YH, Miao YG, et al. Novel molecular and mechanical properties of egg case silk from wasp spider, Argiope bruennichi. Biochemistry. 2006;45(10):3348-56.

Figure 22. PRISMA flow chart showing the inclusion and screening process of papers used within this study. Blue numbers correspond to inclusion and red numbers correspond to removal after screening.

Figure 23. Results of using TeKPol in place of AmuPol on the silk of N. plumipes. Unirradiated sample in red Irradiated sample in blue .