UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Improved pulping efficiency in C4H-F5H transformed poplar Huntley, Shannon Kelly 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2003-0485.pdf [ 8.17MB ]
JSON: 831-1.0075020.json
JSON-LD: 831-1.0075020-ld.json
RDF/XML (Pretty): 831-1.0075020-rdf.xml
RDF/JSON: 831-1.0075020-rdf.json
Turtle: 831-1.0075020-turtle.txt
N-Triples: 831-1.0075020-rdf-ntriples.txt
Original Record: 831-1.0075020-source.json
Full Text

Full Text

IMPROVED PULPING EFFICIENCY IN C4H-F5H T R A N S F O R M E D POPLAR by Shannon Kelly Huntley B.Sc, University of British Columbia, 2000 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Wood Science) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A January 2003 © Shannon Kelly Huntley, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) Abstract Changes in wood chemistry could have significant impact on both environmental and economic aspects of the pulp and paper industry. Consequently, a considerable amount of effort has been devoted to altering lignin content and/or modifing lignin monomer composition, a cell wall component whose removal is a major part of the chemical pulping process. Analysis of poplar transformed with a cinnamate 4-hydroxylase (C4H):ferulate5-hydroxylase (F5H) construct confirmed significant increases in the mole percent syringyl lignin in transgenic lines. Further, this study demonstrated significant increases in pulping efficiency from greenhouse grown transgenic trees. Compared to wild-type pulp, decreases of 23 kappa units (residual lignin) and increases of >20 ISO brightness units were observed in tree lines exhibiting high syringyl monomer concentrations (93.5% mol S). These changes were associated with no significant change in total lignin content or observed phenotypic differences in the trees. Additionally, pulp yields were not affected by the enhanced removal of lignin.. Furthermore, transgenic lines exhibit reduced fibre coarseness and increased cellulose viscosity. These results suggest that C4H-F5H transformed trees could be used to produce pulp for paper with substantially less severe delignification conditions (lower chemical loading or less energy), and that the pulp produced is of comparable quality to that of the wild-type poplar. Consequently, the ecological footprint left on the environment, measured by the amount of deleterious pulping by-products released into the environment may be significantly reduced. ii Table of Contents Abstract ; ii Table of Contents iii List of Tables v List of Figures vi List of Abbreviations • viii Acknowledgements x Chapter 1 Introduction 1 1.1 Background 1 1.2 Wood Anatomy 3 1.2.1 Macrostructure of Wood 3 1.2.2 Wood Growth 3 1.2.3 Angiosperm Cell Types 5 1.2.3.i Longitudinal Cells 7 1.2.3.H Axial Cells 9 1.2.4 Cell Wall Ultrastructure 10 1.2.4.i Tracheid and Fibre Secondary Walls 11 1.2.4.ii Parenchyma Cell Wall Structure 13 1.2.5 Tension Wood 13 1.2.6 Juvenile Wood 14 1.3 Wood Chemistry 16 1.3.1 Carbohydrates 16 1.3.2 Extractives 20 1.3.3 Lignin 22 1.4 The Lignin Biosynthetic Pathway 26 1.4.1 A Review of the Literature 29 1.4.1 .i Phenylalanine Ammonia-Lyase (PAL) 29 1.4.1.ii Cinnamate 4-Hydroxylase (C4H) 30 1.4.1 .iii 4-Coumarate Coenzyme A Ligase (4CL) 30 1.4.1.iv Coumaroyl-CoA 3-Hydroxylase (CCoA3H) 31 L4.Lv Caffeoyl CoA O-Methyltransferase (CCoAOMT) 31 1.4.1 .vi Cinnamoyl-Coenzyme A Reductase (CCR) 32 1.4.1.vii Cinnamyl Alcohol Dehydrogenase (CAD) 33 1.4.1.viii Sinaple Alcohol Dehydrogenase (SAD) 34 1.4.1.ix Caffeate O-Methyltransferase (COMT) / 5-Hydroxyconiferayl Aldehyde O-Methyltransferase (AldOMT) 35 1.4.1.x Ferulate 5-Hydroxylase (F5H) / Coniferyl Aldehyde 5-Hydroxylase (CAld5H) 38 1.5 Objectives 40 Chapter 2 Materials and Methods 41 2.1 Plant Material 41 2.2 Moisture Content Determination 42 2.3 Wood Extractives Content 42 2.4 Wood and Pulp Compositional Analysis 43 2.4.1. Wood and Pulp Total Lignin Determination 43 iii 2.4.2. Wood and Pulp Carbohydrate Analysis 44 2.5 Determination of Wood and Pulp Lignin Monomer Ratio (S:G) 45 2.5.1. Determination of Wood Lignin Monomer Ratio (S:G) 45 2.5.2. Determination of Pulp Lignin Monomer Ratio (S:G) 46 2.6 Derivitization Followed by Reductive Cleavage (DFRC) 47 2.7 Fibre Quality Analysis of Wood 48 2.8 Fibre Quality Analysis of Pulp 50 2.9 Image Analysis 51 2.10 Specific Gravity of 2-Year Old Trees 51 2.11 Kraft Pulping Process 52 2.12 Kappa Number Determination 54 2.13 Pulp Viscosity 55 2.14 Pulp Bleaching 56 2.15 Molecular Weight of Black Liquor Lignin Determination 57 Chapter 3 Wood Analysis • 59 3.1 Wood Lignin Analysis 59 3.1.1 Determination of Lignin Monomer Composition 63 3.1.2 Determination of Lignin Content 67 3.2 Wood Carbohydrate Analysis 72 3.3 Extractives Analysis 74 3.4 Image Analysis 76 3.5 Wood Fibre Quality Analysis 81 3.6 Wood Specific Gravity Determination 81 Chapter 4 Pulp Analysis 84 4.1 Kraft Pulping 89 4.2 Pulp Yield 93 4.3 Determination of Pulp Lignin Content 95 4.4 Determination of Pulp Lignin Monomer Composition 98 4.5 Determination of Pulp Carbohydrate Composition 98 4.6 Determination of Pulp Viscosity 101 4.7 Pulp Fibre Quality Analysis 103 4.8 Pulp Bleaching Analysis 106 4.9 Black Liquor Lignin Analysis 106 Chapter 5 Conclusions 110 Chapter 6 Future Work 112 Bibliography 114 Appendix A 123 Appendix B 124 Appendix C 125 Appendix D 126 iv List of Tables Table 1. Kraft Pulping Cooking Conditions for 1-Year Old C4H-F5H Transgenic Poplar 53 Table 2. Kraft Pulping Cooking Conditions for 2-Year Old C4H-F5H Transgenic Poplar 53 Table 3. Syringyl and guaiacyl monomer content of wild-type and C4H-F5H transformed poplar as determined by thioacidolysis and derivatization followed by reductive cleavage (DFRC) 64 Table 4. Pulp viscosities (cP) of wild-type and C4H-F5H transformed poplar processed at different H-factors 102 Table 5. Initial and final ISO brightness values of wild-type and C4H-F5H transformed poplar processed at different H-factors 107 Table 6. Pulp coarseness (rag/m) of wild-type and C4H-F5H transformed poplar processed at different H-factors 125 v List of Figures Figure 1. Lateral section of the apical meristem 6 Figure 2. Major cell types 8 Figure 3. Stylized drawing of wood cell wall layers 12 Figure 4. Cross-sectional view of poplar tension wood fibres •. 15 Figure 5. Hydrogen bonding between cellulose strands : 17 Figure 6. Schematic representation of the two major hemicelluloses found in hardwoods 19 Figure 7. Structures of the three alcohol precursors in lignin monomer biosynthesis 23 Figure 8. Depiction of the possible structures of resonance-stabilized phenoxy radicals 23 Figure 9. Major structures organized into three categories of bonding found in the native lignin. .; 25 Figure 10. Pathway to angiosperm monolignols 28 Figure 11. Thioacidolysis mechanism 61 Figure 12. Derivitization followed by reductive cleavage mechanism 62 Figure 13. GC/MS spectrum of DFRC products from wild-type and C4H-F5H 64 68 Figure 14. Partial GC/MS spectrum of DFRC products 69 Figure 15. Percentage of lignin 70 Figure 16. Carbohydrate composition 73 Figure 17. Percentage of extractives 75 Figure 18. Relative proportions of the extractive classes 77 Figure 19. Average fibre lumen areas 78 Figure 20. Average vessel lumen areas 79 Figure 21. Average cell wall thickness 80 Figure 22. Average wood fibre coarsenesses 82 Figure 23. Average wood density 83 vi Figure 24. Kraft pulp mechanism 87 Figure 25. Residual lignin values (kappa number) 90 Figure 26. Residual l ignin values (kappa number) generated from pulp of 2-year-old 92 Figure 27. Pulp yields of one-year-old 94 Figure 28. Pulp yields o f two-year-old 96 Figure 29. Lignin content of pulp 97 Figure 30. Lignin monomer ratio in wood and pulp 99 Figure 31. Carbohydrate contents of pulp 100 Figure 32. Coarseness versus kappa number 104 Figure 33. Fibre lengths of pulp 105 Figure 34. Elution times (relative molecular weight) of black liquor lignin 109 Figure 35. L ignin content of pulp 123 Figure 36. Lignin content of pulp 123 Figure 37. Carbohydrate contents of pulp 124 Figure 38. Carbohydrate contents of pulp 124 Figure 39. Elution times (relative molecular weight) of black liquor lignin 126 Figure 40. Elution times (relative molecular weight) of black liquor lignin 126 vi i List of Abbreviations 3 C H coumaric acid 3-hydroxylase a alpha Ac acetyl Ar aryl Ara arabinose AU absorbance units p beta °C degrees Celsius C 4 H cinnamate 4-hydroxylase C 4 L 4-coumaroyl-CoA CAD cinnamyl alcohol dehydrogenase CCR cinnamoyl Coenzyme A reductase cm centimetre CoA coenzyme A COMT caffeic acid O-methyltransferase or caffeate (9-methyl transferase cp centipoises D- dextrorotatory Da dalton DFRC derivitization followed by reductive cleavage D 0 initial chlorine dioxide D- second chlorine dioxide DP degree of polymerization E alkaline extraction (NaOH) ER enzymatic reticulum Et20 ethanethiol F 5 H ferulate 5-hydroxylase y gamma g gram G guaiacyl Gal galactose GC gas chromatography GC/MS gas chromatography/mass spectrometry Glu glucose GPC gel permeation chromatography H p-hydroxyphenyl ha hectare hr(s) hour(s) HPLC high performance liquid chromatography IS internal standard ISO International Standards Organization kg kilogram L- levorotatory L litre M molar m metre m3 cubic metre Man mannose min. minute Mg milligram mL millilitre ML middle lamella mm millimetre mm2 square millimetre mol mole mPa-s metre x Pascales x seconds MWL milled wood lignin N normal nm nanometre 0 ortho P primary cell wall layer P para P pyranose PAL phenylalanine ammonia lyase psi pounds per square inch Rha rhamnose rpm revolutions per minute S syringyl s, first secondary cell wall layer s2 second secondary cell wall layer S3 third secondary cell wall layer So gelatinous secondary cell wall layer SAD sinapyl alcohol dehydrogenase SEC size exclusion chromatography SEM scanning electrom microscope Temp temperature microgram uL microlitre W warty layer Xyl xylose y year ix Acknowledgements I would like to extend my heartfelt thanks to my supervisor, Shawn Mansfield. He has facilitated my learning experience with his wisdom, energy, encouragement, and unique brand of humour. I am especially thankful for Shawn's dedication to his students, which is evidenced in the long hours he spends reading their papers and theses and giving his endless assistance in the lab. I would also like to say thank-you to Shawn for his 'open door' policy, which I have made use of many times. Furthermore, I would like to acknowledge Shawn for the time he takes to know his students, for making us feel like old friends, and for creating an environment that is a pleasure to learn and work in. To Dave El l i s , I also owe thanks for his indispensable insight, support, and valuable discussions not to mention, he provided the starting material that was the foundation for the success of this project. M y thanks are also extended to Jack Saddler for his backing and directed questions which have wonderfully supported the writing of this thesis. I would also like to thank the technical staff at the Pulp and Paper Centre and Paprican for kindly teaching me to use their pulp testing equipment and guiding me with their experience. O f course, the successes I have experienced pursuing this degree has largely been supported by my peers in the Wood Science Department. The other students that I have had the privilege of working beside have provided me with much needed research and emotional support and to them I am thankful. M y family has also played a pivotal role in my ability to complete the requirements for this degree by buoying me up when my spirit faltered. Therefore, I want to give my loving thanks to my son Alexei and husband Michael for their love, encouragement, and generous understanding. x Chapter 1 Introduction 1.1 Background The past decade has been marked by a radical shift in the way society views forest practices. Industrial utilization of natural forests has been paramount; however, a new paradigm has emerged that emphasises conservation of natural forestlands and recognizes the inherent value of forests as carbon sinks, wildlife habitats, scenic and recreational areas. Multiple use of forestland, whereby it is removed from the industrial land base, represents a growing global trend. However, as the human population continues to increase, pressures on the world's forests will escalate in order to meet demands for wood products, fuel and agricultural land. The United Nations estimates that the global population reached 6 billion in 2000 (Anon, 2001) and the present timber harvested is approximately 3.6 billion cubic meters annually (COFI, 2000). With the supposition that the per capita consumption of forest-based products continues at the current rate for a global population that is estimated to reach 9.4 billion in 2050, there will be a dramatic impact on the timber supply of the future. Countries having a vast forest resource have traditionally held an advantage in the global marketplace, however, this advantage is shifting toward those regions with the greatest productivity (Brown, 2000). Subsequently, high yield plantation forestry represents a practical means of supplying sufficient timber for future wood fibre demands. Currently, there is an estimated 187 million hectares of plantation forests worldwide, which represent only 5% of the global forest resource, yet plantation forests supply greater than 20% of the industrial roundwood (FAO, 2001). Furthermore, plantation forests provide approximately 23 percent of the global pulp resource (FAO, 2001). 1 The majority of species grown in tree farms are eucalypts, radiata pine, southern pine, and poplar; they are fast-growing, have short-rotations and provide high-yields. Traditional breeding programs have substantially increased the value of plantation trees, particularly Eucalyptus in the Southern hemisphere. Brazil provides an excellent case study; eucalypt fibre farms generate close to 100% of the country's wood fibre for pulp and paper production, -50% of this is supplied by Aracruz Celulose (Campinhos, 1999). In the Northern hemisphere, poplar has become an increasingly important species and is being utilized for a variety of wood products, including pulp and paper (Balatinecz and Kretschmann, 2001). Populus species are well suited for cultivation as they are native to northern climate, have rapid growth (~25m3/ha/yr), short generation time, and are amendable to vegetative propagation (Stanturf et al, 2001). They are also well suited for genetic modification (Dickmann, 2001); biotechnological advances in understanding and manipulating the poplar genome may provide a means of tailoring trees for specific industrial end uses, such as papermaking. Several areas of research are being pursued in order to transgenetically improve trees for specific end uses. These include reducing the duration of the tree's juvenile phase, increasing growth rates, insect and disease resistance, ensuring sterility, altering cellulose content, and modifying the lignin composition and content (Mullen and Bertrant, 1998; Pena and Seguin, 2001). The majority of tree transgenic research has focussed on lignin modification, largely because of its implications for the pulp and paper industry and for its value as an energy source (Baucher et al, 1998). Altered expression of almost every enzyme in the lignin biosynthetic pathway of several model plant species, including Arabidopsis, Tabacum, and Populus has been used as a means of understanding lignin biosynthesis and to modify lignin content and/or chemical ultrastructure (Baucher et al, 1998; Whetten et al, 1998; Grima-Pettenati and Goffner, 1999; Boudet, 2000; Dixon et al, 2001). The current research is a fundamental evaluation of the effects of over-expression of a 4-coumerate-ligase (C4H):ferulate 5-2 hydroxylase (F5H) construct in both the wood and kraft pulp of transgenic hybrid poplar (P.tremula x P.alba). As the chemical and physical properties of wood dictate its utilization an introductory chapter describing the unique characteristics of wood anatomy and chemistry is provided. This is followed by a review of the literature pertaining to the phenylpropanoid pathway and finally, the objectives of this study are outlined. The results and discussion section of this research project are addressed in two chapters, one dedicated to wood analysis, and the other to kraft pulp analysis. The results and discussion sections are followed by a summary of the conclusions that may be drawn from this work, and finally a discusion highlighting potential future work is provided. 1.2 Wood Anatomy 1.2.1 Macrostructure of Wood The xylem or wood is comprised of two distinct layers, the sapwood (outer layer) and heartwood (innermost layer). The sapwood is composed primarily of dead cells, but still continues to support a few living cells (parenchyma), and provides support, water conduction, and nutrient storage for the tree. The heartwood is usually darker in colour than the sapwood; composed generally of dead xylem cells, and functions mainly as support. Extractives are organic compounds that are concentrated mainly in the heartwood where they are often the cause of the dark colouration. At the very centre of the tree is the pith, a parenchymatous tissue. The formation of bark and characteristics of heartwood are unique identifiers of individual tree species (Panshin and de Zeeuw, 1980; Stenius, 2000). 1.2.2 Wood Growth The cambium is a layer of actively dividing cells between the xylem and the phloem, and as such, is responsible for the secondary growth of the tree. In northern climates, this active 3 layer becomes dormant in the winter, and therefore each year the cambium forms new xylem, to the inside, to produce an annual growth ring. However, annual rings may be difficult to discern, depending on the species, type of growing season, and climate. They are virtually non-existent in regions where continuous growth is possible, or may reflect rainy versus dry seasons, or loss of foliage rather than annual growth (Panshin and de Zeeuw, 1980; Stenius, 2000). There are two types of growth produced in the xylem, early wood and late wood. Early wood is formed early in the growing season (spring) and is characterized by cells with larger lumens and thinner cell walls to accommodate greater volumes o f liquid movement throughout the tree. Conversely, late wood cells (summer wood) have narrower lumens and thicker walls, and subsequently have greater density. The transition from early to late wood is indicative of different tree types. For example, there may be a very distinct boundary as seen in larch or gradual as with soft pines. The presence o f vessels, running longitudinally in the xylem o f angiosperms makes the distinction between early and late wood relatively clear. Vessels are incorporated in the xylem by particular patterns. For example, poplar, eucalyptus, birch, beech, maple, and aspen form a diffuse-porous pattern, with vessel elements of a fairly consistent size and distribution throughout the annual ring. A ring-porous pattern is formed when large vessel elements are laid-down in the early wood and much smaller vessels in the late wood; this pattern is typical of oak, ash, and elm. Semi-ring porous have gradually decreasing vessel diameters from early wood to late wood, and may have somewhat more vessels in the early wood, as is observed in cherry trees (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989; Stenius, 2000). Resin canals are a distinguishing feature of many softwood species; they are formed in both the radial and transverse planes of the wood. Epithelial parenchyma cells surround the resin canals. A l l trees however, produce rays, primarily comprised of ray parenchyma cells and oriented in the direction of pith to bark (Panshin and de Zeeuw, 1980; Stenius, 2000). 4 As discussed, the active division of cells in the cambium increases the girth of a tree. The lateral meristems or cambial initials are comprised of two cell types, fusiform initials and cambial ray initials. Fusiform initials are long cells with tapered ends and are oriented parallel to the pith, producing xylem and phloem cells. The ray cells in the tree are derived from cambial and fusiform ray initials. The circumference of the cambium increases along with the diameter of the tree; this is achieved through two types of cell divisions within the fusiform initials. Periclinal division gives rise to xylem and phloem. When new xylem is formed the initial cell divides to replace itself and to produce a xylem mother cell, the mother cell divides, then its daughter cells divide. Overall, much more xylem is produced compared to phloem. Anticlinal division results in the formation of new fusiform initials, thereby increasing the circumference of the cambial layer and consequently, the tree (Panshin and de Zeeuw, 1980). Of course, primary growth also occurs, increasing the height of the tree and lengthening its branches and roots. This growth occurs in the apical meristems located at the tips of roots and branches (Figure 1). The promeristem is the most distal cell layer (in angiosperms there may be several layers) that is actively undergoing anticlinal division to increase surface area. Behind this layer of tissue, central mother cells are dividing in a more random fashion, to increase volume. The cells begin to differentiate as this growth continues, protoderm becomes epidermis and procambium develops to give rise to the cambium, thereby remaining a meristematic tissue to form secondary wood. The ground meristem, the bulk of the most central cells, ultimately becomes the pith (Panshin and de Zeeuw, 1980). 1.2.3 Angiosperm Cell Types A thorough understanding of hardwood cell types is important when investigating the properties of genetically transformed trees; it will serve as a point of reference in the evaluation of any morphological changes, which may result from a transgenic modification. The anisotropy of wood is a direct reflection of its different cell types, their size (lumen diameter), their relative 5 Figure 1. Lateral section of the apical meristem. A triangle of growth is occurring at the promeristem (a), pith is being formed between the region (b-b), and the procambium is generating the vascular tissue indicated by (c) (Panshin and de Zeeuw, 1980 pg. 58). 6 proportions, and orientation within the wood. Additionally, the chemical composition and structure of the individual cell walls significantly impacts the efficacy of industrial processes on the wood. For example, larger lumen diameter increases permeability, while cell wall thickness affects compression strength, and microfibrillar angle influences the modulus of elasticity and tensile strength. Hardwoods are evolutionarily more advanced than softwoods and exhibit a greater complexity of cell types (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989; Stenius, 2000), as shown in Figure 2. The following two sections summarize the cell types oriented in the longitudinal and radial direction of angiosperms. 1.2.3.i Longitudinal Cells Hardwoods have two types of fibres that run longitudinally; fibre tracheids and libriform fibres. Both provide strength to the wood. Compared to the other cell types, fibres have narrow lumens and thick cell walls. The type of pits, or openings between cells, is what delineates a fibre tracheid (having boardered pits) from a libriform fibre (having simple pits). Fibres make up on average 50% of the wood volume in hardwoods, with wide variations between species. Populus fibre volume has been estimated to be close to 54% (Panshin and de Zeeuw, 1980). Two types of tracheids are found in hardwoods, vascular tracheids and vasicentric tracheids. Vascular tracheids are associated with vessels in the latewood and share many similarities with this cell type. Except for the lack of perforated ends, they are very similar to vessel elements; sharing the same dimensions, vascular tracheids are indiscernible from vessels in a transverse section. Vasicentric tracheids are mainly associated with earlywood vessels and with axial parenchyma in some ring-porous species. They have closed ends and many lateral bordered pits, and an irregular shape and distribution within the wood. The average lengths of Populus fibres and tracheids range between 0.7-1.6 mm (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989; Stenius, 2000). 7 softwood hardwood "* bordered p i t s tracheld / perforation £3 ^7 vessel element trachei to wood fiber fusiform initials axia l parenchyma cells axial parenchyma cells D ray parenchyma eel ray tracheid ray X ray parenchyma eel1 initials -—=:—— ' Figure 2. Major cell types derived from fusiform and ray initials in wood (Fujita and Harada, 2001 page 3). 8 Vessels or pores are the key feature of hardwood xylem distinguishing it visually from the wood of conifers. In Populus, vessels make up approximately 33% of the total wood volume. Vessels vary in length from several centimetres to several metres. On average, poplar vessels are 500 mm long (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989). Individual cells called vessel elements stack end to end, primarily in a longitudinal orientation to form a vessel. Vessel elements are derived from fusiform initials. The newly formed vessel element initially increases its diameter to its full width, and then forms pits and spaces on the ends called perforations, prior to secondary wall deposition. An enzyme called cellulase dissolves the perforations between two vessel elements and opens the pits to accommodate the flow of water up the stem (Panshin and de Zeeuw, 1980). The region where perforations occur on the vessel element is called a perforation plate. There are several styles of perforations plates: simple, as seen in Populus, with one perforation; scalariform, with several parallel perforations; and reticulate or foraminate perforation plates with several random shaped perforations (Panshin and de Zeeuw, 1980). 1.2.3.H Axial Cells Rays are a transverse organization of entirely paranchyma cells in hardwoods, with the exception of aggregate rays, which are narrow rays separated by fibres and/or vessels. Rays in hardwoods form a complexity of arrangements by variations in their size, shape, spacing, and groupings (Panshin and de Zeeuw, 1980). Homocellular rays are comprised of cells of one type exhibiting the same size and shape, while heterocellular rays have both types of ray parenchyma cells. Procumbent cells have their long axis along the radial plane, while upright cells have a transverse arrangement. Ray parenchyma cells may have either simple or bordered pits, where the size and distribution reflects the pit style of neighbouring cells. In Populus, the rays are homocellular and form a uniseriate arrangement, that is, a single row of upright cells stacked vertically. A multiseriate 9 arrangement is formed when two or more vertical rows of procumbent and/or upright cells form the ray. The proportion of rays contained in the wood varies between species from 10-20% of total wood volume; in Populus approximately 12% of the wood volume is taken up by rays (Panshin and de Zeeuw, 1980). 1.2.4 Cell Wall Ultrastructure The different cell wall layers are discernable with light microscopy, however, with the aid of an electron microscope even structural details of the cell wall are detectable. Harada and Cote (1985) proposed an architecture to explain the interaction between cellulose, hemicellulose, and lignin in the cell wall. The primary component of all cell walls is the microfibril, a long strand of crystalline cellulose surrounded by short-chain hemicelluloses. As microfibrils are deposited in an essentially parallel fashion, they are bound closely together in a matrix with lignin to form rigid sheets called lamellae. Microcapillaries (void spaces) can form between incongruous microfibrils that lack complete lignification, thereby creating cell wall porosity (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989; Fujita and Harada, 2001). Cells are separated from each other by the middle lamella (or true middle lamella). The middle lamella flanked by two primary walls of adjoining cells is called the compound middle lamella (Figure 3). The middle lamella is initially composed of pectin, which forms a plastic layer to accommodate cellular growth, then later, lignin containing both syringyl and guaiacyl residues is deposited (Panshin and de Zeeuw, 1980; Stenius, 2000; Fujita and Harada, 2001). The primary cell wall is associated with meristematic tissue and cellular enlargement. Obliquely oriented cellulose microfibrils form several criss-crossing layers within a loose matrix of approximately 70% water, 9% cellulose, and the remainder comprised of pectin, lignin, and hemicelluloses. As the cell matures it will contain a greater portion of lignin (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989). 10 1.2.4.i Tracheid and Fibre Secondary Walls Inside the fully enlarged primary cell wall, the three secondary wall layers begin to thicken with the deposition of cellulose (Figure 3). Secondary wall lamellae formation occurs from the centre and radiates out toward the tips of the cell. Lignification begins during the development of the Si layer, first being deposited into the middle lamella and primary cell wall (I-lignification). Lignin deposition begins in the cell corners and moves toward the centre, then inward with the formation of successive layers (S-lignification). Concurrently, with I-lignification, hemicellulose deposition is occurring in the enlarging secondary walls. Lignification marks the end of cell growth by producing cell wall rigidity. Cell death occurs a short time after lignification is complete; autolysis dissolves the cytoplasm contents, which are deposited on the inner surface of the S3 layer. (Panshin and de Zeeuw, 1980; Stenius, 2000; Fujita and Harada, 2001). The relatively thin Si layer is formed just inside the primary cell wall. The cellulose within lamellae is oriented in a moderate helix configuration. Several lamellae, having oppositely oriented cellulose helices termed S (oriented to the left) and Z (oriented to the right), form the Si layer. With each addition of lamellae, the helical angle from the longitudinal axis decreases. The microfibril orientation within the lamellae of the Si layer is largely responsible for tensile strength in the transverse direction, and compression resistance in the longitudinal direction. (Panshin and de Zeeuw, 1980; Fujita and Harada, 2001). The characteristics of the S2 layer are largely responsible for the overall strength of the cell wall, as it is the thickest layer. In normal wood, the angle of the microfibrils from the longitudinal axis is shallow, approximately 10-30°. Syringyl residues predominate in the lignin of this cell wall layer in hardwood fibres (Panshin and de Zeeuw, 1980; Stenius, 2000; Fujita and Harada, 2001). 11 Figure 3. Stylized drawing of wood cell wall layers depicting the middle lamella (ML), primary cell wall (P), secondary cell wall layers (Si, S 2 , S 3 ) , and warty layer (W) (Sjostrom, 1993 page 14). 12 The S 3 layer is the thinnest section, having several lamellae. The microfibrillar angle varies widely, but usually falls between 60-90°. The innermost surface facing the lumen may be a warty layer; the bumps are comprised mainly of lignin. The warty layer is not present in more advanced hardwoods. Vessel cell wall development is similar to that of fibres except that lignification primarily involves guaiacyl residues. Furthermore, the Si and S 3 layers of vessel elements are thicker than the other cell type, a form necessary to withstand the tension of water conduction (Panshin and de Zeeuw, 1980; Fujita and Harada, 2001). 1.2.4.ii Parenchyma Cell Wall Structure Parenchyma cells also have primary and secondary walls that are called protective layers. These layers have a structure very similar to the primary wall, but are thicker. Typically, a protective layer has two lamellar sheets with microfibrils arranged similar to the secondary walls of tracheids or fibres. A plastic layer (isotropic) separates the lamellar sheets; initially, it is rich in hemicelluloses and later becomes encrusted with lignin deposits. There may be up to six of these protective layers in a given parenchyma cell. Cellular death does not follow lignification in parenchyma cells, as they remain alive until they are converted into heartwood (Panshin and de Zeeuw, 1980; Fujita and Harada, 2001). 1.2.5 Tension Wood Reaction wood forms in trees experiencing mechanical stress, and is developed to counter the effects of gravity on branches and leaning stems. This type of wood differs from normal wood in both anatomical and physiological properties, which will subsequently produce changes to the physical properties of the wood. Regions where tension wood may have been present, such as the bases of the stem and knots, were avoided in this research to reduce aberrant findings. 13 Tension wood is the type of reaction wood found in most angiosperms, although the more primitive species produce reaction wood resembling conifer compression wood. Usually tension wood forms opposite to the curvature of a lean; the radius of the tension wood is greater and experiences faster growth than the opposite side (Panshin and de Zeeuw, 1980). Microscopically, tension wood fibres will often occur in earlywood, have more rounded corners, are longer, and have thicker walls. However, the most obvious feature of tension fibres is the presence of a gelatinous wall (Figure 4), which may occur adjacent to the S3 layer, or replace it, or potentially replace all of the cell wall layers. The woolly appearance of lumber cut from tension wood is caused by the loose adhesion of the G-layer (So) to the other cell walls, which causes extrusion of this layer when the wood is sawn. Chemically, tension wood is rich in highly crystalline cellulose (40-50% more cellulose content), and has increased galactan content, while lignin and xylan contents are significantly reduced in comparison to normal wood (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989). 1.2.6 Juvenile Wood An understanding of juvenile wood is important to this research as all the analyses were performed on wood falling into this category. The following section discusses the unique characteristics associated with this type of wood. Juvenile wood is xylem formed closest to the pith, and is often referred to as pith wood. It is associated with the ascent of the apical meristems as the crown grows. The formation of juvenile wood varies between species, but is generally produced during the first 5-20 years of growth. Typically, juvenile wood contains shorter cells, which have lower cellulose, higher lignin and hemicelluloses contents than mature wood, although, in hardwoods the total lignin content remains fairly constant (Panshin and de Zeeuw, 1980; Fengel and Wegener, 1989). As well, juvenile wood has greater microfibrillar angle and thinner cell walls. Spiral grain is also typical of juvenile wood, and results from the manner in which fusiform initials divide (Panshin and de Zeeuw, 1980). 14 Figure 4. Cross-sectional view of poplar tension wood fibres showing the gelatinous layers that have extruded as the wood dried (SEM 3800x) (Panshin and de Zeeuw, 1980 pg. 317). 15 1.3 Wood Chemistry Cellulose is the most abundant molecule in wood (40%), followed by hemicelluloses (30-35%) and lignin (20-25%), and a variety of other components present in small amounts including extractives, proteins, pectin and starchs. (Stenius, 2000). This section will include a general discussion of the basic chemistry of carbohydrates, then of cellulose, hemicellulose, and extractives, followed by a more detailed overview of lignin chemistry which is the primary focus of the current investigation. 1.3.1 Carbohydrates Not only is cellulose the most abundant molecule in wood, it is the most prevalent biopolymer on Earth, and it has been estimated that some 1011 tons is produced and recycled annually (Stenius, 2000). Cellulose is an intramolecular, linear /3-D-glucopyranose homopolysaccharide with repeating (l-» 4)-glycosidic bonds. Hydrogen bonding between adjacent cellulose polymers produces highly crystalline microfibrils, punctuated by amorphous regions. However, the manner in which microfibrils interact with the other components, hemicellulose and lignin, to form fibrils has not been fully elucidated (Sjostrom, 1993; Stenius, 2000). Cellulose chains are highly stable, characterized by the chair (4Ci) conformation of the pyranose ring and equatorial orientation of all its substituents. Cellobiose is the repeating disaccharide, 4-0-(/3-D-glucopyranosyl)-D-glucopyranose, which makes up the cellulose chain. Microfibrils are oriented parallel to each other, with 180° angle of separation between glucose residues. Figure 5 shows the hydrogen bonding that occurs between adjacent microfibrils. Intermolecular hydrogen bonding is possible between the hydrogen of the hydroxyl group on C6 of one glucose to the oxygen of the hydroxyl group on C3 of the other molecule. Two types of 16 Figure 5. Hydrogen bonding between cellulose strands running parallel to one another forming a layer along the a-c plane. Two intramolecular H-bonds are possible and one intermolecular bond (please refer to the text for details) (Sjostrom, 1993 pg. 56). 17 intramolecular hydrogen bonding are possible; between the ring oxygen and the adjoining hydrogen from the hydroxyl group on C3 and between hydrogen of the hydroxyl group on C6 and the oxygen of the hydroxyl group on C2 of the neighbouring residue (Sjostrom, 1993; Stenius, 2000). The crystalline nature of cellulose restricts its ability to be solubilized, however, both cupriethylenediamine and cadmiumethylenediamine have been shown to effectively solubilize cellulose. It has been determined that wood cellulose has a degree of polymerization of approximately 10,000 glucose residues, with an estimated total molecular mass between 1.6 and 2.4 million Da (Sjostrom, 1993; Stenius, 2000). Hemicellulose is comprised of different carbohydrate combinations, including hexoses: D-glucose, D-mannose, and D-galactose; and pentoses: D-xylose, L-arabinose, and D-arabinose; deoxyhexoses: L-rhamnose, 6-deoxy-L-mannose, and only rarely are L-fucose and 6-deoxy-L-galactose found in hemicellulose. Uronic acids may also be found in hemicelluloses, such as 4-O-methyl-D-glucuronic acid, D-galacturonic acid, and D-glucuronic acid. It follows that hemicelluloses are heterogeneous polysaccharides, which may be branched, and are much shorter than cellulose, with a degree of polymerization of approximately 200 residues. Having these characteristics mean that hemicelluloses are much more readily solubilized than cellulose (Sjostrom, 1993; Stenius, 2000). Hardwood hemicelluloses are found in two major forms: glucuronoxylan makes up between 20-30% dry wood mass, while glucomannan comprises <5% of dry wood mass (Figure 6). Glucuronoxylan or 0-acetyl-4-0-methylglucurono-/3-D-xylan has a backbone of repeating /3-D-xylopyranose units with 4 Ci bonds. The xylose component of glucuronoxylan contains substituents, generally 7 in 10 having an O-acetyl group on the C2 or C3 and approximately 1 in 10 having a lC2 bonded 4-O-methyl-a-D-glucuronic acid residue. The 18 4)-p-D-Xylp-(1 f * 4)-p-D-Xylp-(1 1 > 4)-p-D-Xylp-(1 r \ ^ 9 4-O-Me-a-D-GlucpU Glucuronoxylan 4)-p-D-Glcp-(1 *~ 4)-p-D-Manp-(1 4)-p-D-Manp-(1 Glucomannan Figure 6. Schematic representation of the two major hemicelluloses found in hardwoods: glucuronoxylan and glucomannan (Stenius, 2000 pg. 37). 19 uronic acid bonds are highly resistant to acid cleavage, whereas the 4 Ci bonds are readily hydrolysed. Alkaline treatment such as kraft pulping readily cleaves the acetyl groups to form acetate. In contrast, glucomannan exhibits 4 Q bonds, as observed in cellulose, between /3-D-glucopyranose and /?-D-mannopyranose forming a highly linear molecule. The molecule exists in a 1:1-2 glucose to mannose ratio. The 4 Ci bonds between mannose monomers are readily broken by acid hydrolysis. Trace amounts of other hemicelluloses also exist in hardwoods, but will not be discussed here (Sjdstrom, 1993; Stenius, 2000). 1.3.2 Extractives Although extractives make up <5% of dry wood weight, they represent a significant diversity of small, non-structural wood components. They are removed from wood via extraction with neutral solvents or water, depending on whether the molecule is lipophillic or hydrophilic. Extractives content is primarily analyzed by gas chromatography, which maybe coupled with mass spectrometry. Tree species attain their unique wood colour, odour, and taste from extractives. Essential biological functions are performed by extractives; for example, fats and waxes supply energy, and phenolics, resin acids, and some terpenoids thwart microbial and insect damage. Extractives are important from an industrial perspective as well. For example, kraft pulping yields tall oil and turpentine, both are value-added by-products produced from pulping industry. Furthermore, extractives can impact the pulping process by corroding metals, causing colour reversion in mechanical pulp, and by forming pitch deposits (Sjostrom, 1993; Stenius, 2000). Essential oils are volatile materials primarily found in softwoods; they include terpenes and terpenoids, aliphatic hydrocarbons, phenols, alcohols, ethers, aldehydes, and lactones. The basic structural component of terpenoids and steroids is the isoprene unit, 2-methyl-l,3-20 butadiene. Terpenoids are categorized into subgroups based on the number of isoprene units and ring structures. Compounds that are formed from isoprene units usually follow the isoprene rule, with head to tail bonding, but not always. Tail to tail dimerization of isoprene units forms squalene, which in turn is used to produce farnesyl pyrophosphate, a triterpenoid. Monoterpenes, monoterpenoids, sesqiterpenes-terpenoids, and diterpenes-terpenoids are primarily found in softwoods and rarely in tropical hardwoods. However, triterpenes/terpenoids are ubiquitous in the kingdom Plantae. These compounds are derived from the acyclic squalene precursor, and differ only in the presence of one or two methyl groups on tetracyclic triterpenoids. These compounds are further categorized into tetracyclic lanostane, pentacyclic lupane, and pentacyclic oleanane. Barring the hydroxyl group, sterols are very lipophilic. Pulp and paper processing is hampered by triterpenoids and steroids, which occur mainly as fatty acid esters (waxes) and glycosides. Although a wide variety of triterpenes and steroids are found in hardwoods, in only very minor amounts, sitosterol is the most profuse. White birch bark attains its colour from the lupane triterpenoid, betulinol. Rubber, gutta percha, and balata are polyterpenoids and each is comprised of many isoprene units, but differ in either their cis or trans stereochemistry (Sjostrom, 1993; Stenius, 2000). Parenchyma resin is the primary location of fats and waxes in hardwoods, because of its location in the wood it may be difficult to liberate during the pulping process. However, once fats and waxes are hydrolyzed they may be recovered from the black liquor. In wood, triglycerides are the predominant form of fats (glycerol esters of fatty acids). Waxes are long chains of fatty alcohols, terpene alcohols or sterols (Sjostrom, 1993; Stenius, 2000). There are a vast array of aromatic extractives in wood, particularly in the bark and heartwood. They are often coloured molecules derived from the phenylpropanoid pathway. Stilbenes are extractives found in the heartwood of pines. Lignans are common in both softwood and hardwood species, and are the primary cause of heartwood colouration. Oxidative 21 coupling of two phenylpropane units at C 6 C 3 forms the basic lignan structure. Hydrolyzable tannins are uncommon in wood. When present they are usually esters of D-glucose bound to at least one polyphenol carboxylic acid (gallic, digallic, and ellagic). Flavonoids are also found in deciduous and coniferous trees and their basic structure is a tricyclic, diphenylpropane (C6C3C6). Polymers of 3-8 flavonoid units form condensed tannins, which are found in many tree species (Sjostrom, 1993; Stenius, 2000). 1.3.3 Lignin After cellulose, lignin is the most abundant biopolymer on Earth. As with extractives, the basic building blocks of lignin are phenylpropane molecules, which are fashioned into aromatics. Lignin is unlike any of the other biopolymers addressed thus far in that it is an amorphous substance, with seemingly random molecular linkages. Lignin is essential for plants as evidenced in its prevalence in cells providing mechanical support and water conduction. Additionally, lignin may be spontaneously synthesized in response to injury (Sjostrom, 1993; Stenius, 2000; Sakakibara and Sano, 2001). Lignin composition in different plant types allows for a broad classification; grass lignin is comprised of derivatives from trans-p-coumaryl alcohol (the precursor of p-hydroxyphenyl monomers), trans-coniferyl alcohol (the precursor of guaiacyl monomers), and trans-sinapyl alcohol (the precursor of syringyl monomers) (Figure 7). Softwood lignin may contain greater than 95% guaiacyl residues (G); usually the remainder is comprised of p-hydroxyphenyl monomers (H), although some species will also contain traces of syringyl residues (S). Hardwood lignins exhibit various ratios of both G and S monolignols depending on the species, with an average of 60% S and 40% G units (Sjostrom, 1993; Baucher et al., 1998; Stenius, 2000; Sakakibara and Sano, 2001). Lignin undergoes complex intramolecular and intermolecular bonding within the cellulose-hemicellulose matrix, and as such an absolute determination of its native structure 22 ^rans-conifery! a l coho l trans-smzpyl a l cohol trans-/?-coumaryl a lcohol Figure 7. Structures of the three alcohol precursors in lignin monomer biosynthesis. Trans-coniferyl alcohol becomes the guaiacyl monolignol dominant in softwoods, rrara-sinapyl alcohol becomes the syringyl monolignol primarily formed in hardwoods, and trans-p-coumaryl alcohol becomes p-hydroxyphenyl the monolignol found in grasses (Fengel and Wegener, 1989; Sjostrom, 1993 page 76 and 136, respectively). coniferyl a l coho l Figure 8. Depiction of the possible structures of resonance-stabilized phenoxy radicals. Structure III does not participate in bonding of sinapyl alcohol and V does not participate in lignin biosynthesis at all, because of thermodynamic and steric hindrance produced by the methoxyl group. (Fengel and Wegener, 1989; Sjostrom, 1993 page 78 and 136, respectively). 23 has not been possible. Furthermore, as both bonding type and lignin monomer content significantly impact the industrial delignification process, a great deal of research has focussed on determining the frequencies of these bond types, and the different moieties present in the lignin of various trees (Stenius, 2000; Sakakibara and Sano, 2001). The formation of lignin macromolecules begins with the dehydrogenation of the precursor alcohol moieties to form resonance-stabilized phenoxy radicals (Figure 8). This is a reaction catalyzed by peroxidase and laccase activity (Fengel and Wegener, 1989; Sitbon et al, 1999). For syringyl moieties, both phenoxy radicals III and IV would be sterically and thermodynamically unfavourable. Three types of bonding are possible to form lignin macromolecules. Ether linkages (C-O-C) are the most prevalent form of bonding between lignin monolignols, while carbon-carbon and ester bonds also occur (Figure 9) (Stenius, 2000). Uncondensed guaiacyl moieties have an unbound C5, whereas condensed monomers are formed with C-C or ether bonds at the C5 position (Sakakibara and Sano, 2001). The aryl ether linkage, B-O-4 is the most common, comprising 40-60% of all possible lignin bond types in hardwoods and softwoods. In descending order of frequency, the following bonds are also prevalent: 8-5, B-l, 5-5, and a-O-4, respectively (Fengel and Wegener, 1989; Stenius, 2000). Initially, the presence of B-O-4 bonds was determined by alcoholysis, which cleaves the bond to form ketones. A later study of MWL using the acidolysis method, which supplants the /3-0-4 bond with a C-methyl group, was used to estimate the frequency of this bond to be between 25-30%. These results were corroborated with subsequent research using 'H NMR to estimate B-O-4 bond frequency in birch to be between 40-50%. Further investigation of lignin oxidation products suggests the occurrence to be even higher, at 62% (Sakakibara and Sano, 2001). 24 o Figure 9. Major structures organized into three categories of bonding found in the native lignin of hardwoods and softwoods (Stenius, 2000 page 41). 25 Acidolysis has been used to convert both dehydrodiconiferyl alcohol and dihydrodehydrodiconiferyl alcohol and its phenyl methyl ether into the phenylcoumaran structure. Phenylcoumaran represents the B-5 bonding type, and has been isolated via hydrolysis with dioxane and water. Diazomethane methylation of spruce MWL and subsequent acidolysis indicates, via its absorption spectrum, that there are approximately 0.08 phenylcoumaran moieties per methoxyl (Sakakibara and Sano, 2001). Overall, for both hard-and softwoods, /3-5 bonding represents between 5-10% of possible lignin bond types (Sakakibara and Sano, 2001). It has been speculated that B-l bonding occurs through the coupling of radicals II and IV (Figure 8). Acidolysis yields a compound with /3-1 bonding and a glyceraldehyde-2-aryl ether. The frequency of this ether has been used to determine the frequency of the B-l bonding type in lignin to be 3 % (Sakakibara and Sano, 2001). The most frequent condensed lignin moieties contain 5-5 bonding. The degree of condensed lignin formation was estimated by default. Oxidation, with Fremy's salt, of the uncondensed -^substituted monolignols forms o-quinones. The percentage of uncondensed monolignols was determined with spectophotometry (the difference between these and the total phenolics represents the condensed lignin portion) estimated at approximately 40-50% (Sakakibara and Sano, 2001). 1.4 The Lignin Biosynthetic Pathway D-glucose is the antecedent of all the lignin biosynthetic pathway precursors. The sugar enters the shikimic acid pathway, where a series of complex steps converts linear D-glucose into the amino acids L-phenylalanine and L-tyrosine. Not only are these aromatic molecules the precursors of the cinnamic acid pathway (L-tyrosine is the precursor of the lignin biosynthetic pathway in grasses and L-phenylalanine is the molecule from which woody plant monolignols 26 are derived) but they also feed into the metabolism of various extractives. Figure 10 is the most recently accepted version of the lignin biosynthetic pathway (Harding et al, 2002). Phenylalanine ammonia lyase (PAL) initiates the conversion by deaminating phenylalanine, after which cinnamate 4-hydroxylase (C4H) adds a hydroxyl group to the para position to produce 4-coumarate. Following this, coumaric acid 3-hyroxylase (3CH) catalyzes the hydroxylation at the C3 position to produce caffeate; establishing two separate potential routes. Hydroxycinnamoyl-CoA esters are formed by the activity of 4-coumarate ligase (C4L) to give 4-coumaroyl-CoA and caffeoyl CoA. 4-coumaroyl-CoA is then directed toward flavonoid production, or in grasses, p-hydroxyphenyl synthesis, while caffeate proceeds toward monolignol synthesis. Caffeoyl CoA O-methyltransferase (CCoAOMT) methylates the 3-hydroxyl of Caffeoyl-CoA to produce feruloyl CoA, which is then reduced by CCR (cinnamoyl Coenzyme A reductase) to give coniferaldehyde. Coniferaldehyde represents a branch point in monolignol biosynthesis, whereby coniferaldehyde may continue toward the production of guaiacyl monolignols by the action of cinnamyl alcohol dehydrogenase (CAD), which converts the coniferaldehyde to an alcohol. The coniferyl alcohol ultimately becomes a guaiacyl monolignol. Conversely, coniferaldehyde may be shunted away from the guaiacyl pathway via a hydroxylation step of the aldehyde by ferulate 5-hydroxylase (F5H). 5-hydroxyconiferaldehyde is then methylated by caffeic acid O-methyltransferase (COMT) into sinapaldehyde. Sinapyl alcohol dehydrogenase (SAD) reduces the sinapaldehyde to sinapyl alcohol, which then becomes a syringyl monolignol. (Campbell and Sederoff, 1996; Baucher et al, 1998; Boudet, 1998; Whetten et al, 1998; Grima-Pettenati and Goffner, 1999; Boudet, 2000; Dixon et al, 2001; Harding et al, 2002). Although the largely accepted pathway to monolignol biosynthesis is represented as a fairly simple metabolic grid, another theory suggesting separate routes (metabolic channels) 27 - % PAL C4H C3H phenylalanine Flavonoids 4-coumarate I4CL2 4-coumaroyl-CoA P-hydroxyphenyl CCoAOMT caffeoyl-CoA OH feruloyl-CoA IcCR HO F5H/CAId5H O H conifeiyl aldehyde I CAD COMT/AldOMT 5-hydroxyconiferaldehyde Guaiacyl Lignin Syringyl Lignin Figure 10. Pathway to angiosperm monolignols (Harding et al, 2002). PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; C3H, 4-coumarate 3-hydroxylase; 4CL, 4-coumarate:coenzyme A ligase; CCoAOMT, caffeoyl-CoA O-methyltransferase; CCR, dinnamoyl CoA reductase; F5H, ferulate 5-hydroxylase; OMT, caffeic acid O-methyltransferase; C A D cinnamyl alcohol dehydrogenase; SAD, sinapyl alcohol dehydrogenase. 28 resulting in either the formation of syringyl or guaiacyl moieties, is also a very attractive alternative (Dixon et al, 2001; Harding et al, 2002). This theory is supported by the presence of two essential, ER bound enzymes, C4H and F5H (Pierrel et al, 1994; Meyer et al, 1996). Although a number of enzymes, COMT, CCOMT, 4CL, CCR, and CAD are clearly soluble in the cytoplasm, research has yet to rule out the possibility that they have membrane associations; these close associations would allow for the channelling of intermediates between enzymes. Furthermore, phenylpropanoid enzyme modification, in some cases, has outcomes that are inconsistent with the concept of a linear pathway mediated by diffusion of the transitional molecules (Sewalt et al, 1997a; Sewalt et al, 1997b). The results of recent research indicate that the pathways to syringyl and guaiacyl lignin are largely independent and controlled by negative feedback mechanisms (Osakabe et al, 1999; Li et al, 2000; Li et al, 2001; Harding et al, 2002). 1.4.1A Review of the Literature A significant and ever increasing body of research has focused on altering the expression of key enzymes in the monolignol pathway in a variety of plants, including Arabidopsis, alfalfa, tobacco, and poplar. The resulting transgenic plants exhibit either altered lignin content and/or chemical ultrastructure (reviews in: Campbell and Sederoff, 1996; Baucher et al, 1998; Boudet, 1998; Whetten et al, 1998; Grima-Pettenati and Goffner, 1999; Boudet, 2000; Dixon et al, 2001). 1.4.1.i Phenylalanine Ammonia-Lyase (PAL) PAL catalyzes the deamination of phenylalanine to produce cinnamate, the first step in the production of monolignols. However, the products derived from PAL activity also form the precursors of other phenolic compounds. Consequently, alterations in PAL activity may induce 29 effects beyond lignin biosynthesis. This suspicion has been illustrated in the poor growth exhibited by tobacco expressing reduced levels of PAL and, as expected, these plants also have reduced total lignin content (Elkind et al, 1990; Bate et al, 1994). Interestingly, PAL-suppression also increased the syringyl :guaiacyl ratio (S:G), which is suggestive of monolignol channelling early in the biosynthetic pathway (Sewalt et al, 1997a). 1.4.1.H Cinnamate 4-Hydroxylase (C4H) C4H catalyzes the hydroxylation of the C4 on the cinnamate molecule to produce 4-coumarate. C4H is a cytochrome P450-linked monooxygenase bound to the outer membrane of the endoplasmic reticulum (Teutsch et al, 1993; Chappie, 1998), that has been shown to be expressed in both the xylem and sclerenchyma of poplar (Grand, 1984). Transgenic tobacco with reduced C4H activity has also been shown to exhibit reduced S:G ratios and lower total lignin levels (Sewalt et al, 1997a). 1.4.1.iii 4-Coumarate Coenzyme A Ligase (4CL) 4-coumaroyl CoA is one of a group of thioesters produced by the catalytic activity of 4CL on hydroxylated cinnamic acids. 4CL is present in plants as a number of isozymes. In vitro studies have shown that 4CL is capable of utilizing multiple substrates: 4-coumaric acid, caffeate, ferulic acid and sinapic acid, and as such, the specific activity of 4CL in lignin metabolism has been difficult to elucidate. However, research has demonstrated that the reaction of sinapic acid to sinapoyl CoA is not catalyzed by 4CL in either gymnosperms or most angiosperms (Kutsuki et al, 1982). The substrate affinities of recombinant 4CL proteins from Arabidopsis, tobacco and poplar have been compared and the results corroborate the earlier findings. Overall, these results suggest that sinapate is not a component of monolignol biosynthesis (Lee and Douglas, 1996; Lee et al, 1997; Ehlting et al, 1999). Recent work on 30 4CL kinetics suggests that the kinetic properties of 4CL indicate neither ferulate nor 5-hydroxyferulate could be in vivo substrates for these enzymes (Harding et al, 2002). Two unique 4CL genes have been identified in P. tremuloides; Pt4CLl is expressed in nascent xylem of the stem for lignin biosynthesis, and Pt4CL2 is expressed in epidermal tissues for the production of phenylpropanoids other than lignin (Hu et al, 1998). Recent findings have suggested that caffeate competitively inhibits the other hydroxycinnamic acids in the developing xylem, thereby increasing the carbon flux, via 4CL1, into monolignol biosynthesis. Conversely, 4-coumarate inhibits other hydroxycinnamic acids in the developing epidermis, directing carbon flow toward flavonoid production (Harding et al, 2002). Significantly reduced expression of 4CL in both Arabidopsis and tobacco produced plants with only slightly reduced total lignin (Kajita et al, 1997; Lee et al, 1997). However, transgenic Populus tremoloides, with severely reduced 4CL expression, have significantly less total lignin and as,a result demonstrate substantially higher levels of cellulose, and notably improved growth (Hu et al, 1999). 1.4.1 .iv Coumaroyl-CoA 3-HydroxyIase (CCoA3H) A shift in pH was used as an elicitor to induce CCoA3H expression in both parsley and Zinnia, and has been shown to catalyze the hydroxylation of the C3 on 4-coumaroyl CoA to produce caffeoyl CoA (Kneusel et al, 1989; Ye et al, 1994). However, to date, the enzyme has not been cloned and its substrate affinity has yet to be determined (Baucher et al, 1998). Therefore, it is not included in the current view of lignin biosynthesis. 1.4.1 .v Caffeoyl CoA O-Methyltransferase (CCoAOMT) CCoAOMT is responsible for catalyzing the methylation of caffeoyl-CoA to form feruloyl-CoA. Induction of CCoAOMT occurs during lignification and is primarily expressed in the developing xylem of many dicot plants. As the plant matures, CCoAOMT is expressed in 31 both the developing xylem and in fibres of the phloem (Ye et al, 1994; Ye and Varner, 1995; Ye, 1997). Similarly, loblolly pine appears to express primarily CCoAOMT in developing secondary xylem of the mature stem (Li et al, 1999). Further work in Populus has shown that the expression of CCoAOMT normally occurs in the vessel elements and rays adjoining vessels, not in storage rays or fibres; the specificity of this expression is lost upon both mechanical and biotic stress, with expression occurring in all xylem cell types (Chen et al, 2000). An antisense method was used to reduce the expression of CCoAOMT in transgenic tobacco and alfalfa, and these plants were shown to possess significantly reduced total lignin. While total units were decreased, the guaiacyl monomer content was preferentially reduced such that the S:G ratio increased (Zhong et al, 1998; Guo et al, 2001). Transgenic tobacco plants did not exhibit unusual growth, although an analysis of the vessel anatomy revealed that they, in fact, had collapsed (Zhong et al, 1998). More recent research investigating the down-regulation of CCoAOMT in poplar concurs with the findings covering lignin content, however, while the xylem exhibited a pinkish discolouration, cell wall structure was not affected (Meyermans et al, 2000). 1.4.1.vi Cinnamoyl-Coenzyme A Reductase (CCR) CCR has been found in a variety of plants and is capable of using multiple substrates, such as caffeoyl-CoA, sinapoyl-CoA, and 4-coumaroyl-CoA (Kc a t of 4.9, 3.7, and 3.3 min"1, respectively). However, it has been shown to achieve the best efficiency (Kcat of 12 min"1) with the reduction of feruloyl-CoA to coniferaldehyde (Sarni et al, 1984; Goffher et al, 1994). CCR has been severely suppressed in tobacco, and the resulting transgenic plants exhibited reduced lignin content, increased S:G ratio, poor growth, and abnormal phenotype. In contrast, plants only slightly depressed in CCR activity appear to grow well (Piquemal et al, 1998; Chabannes et al, 2001a). Microanalytical techniques were use to elucidate lignin deposition at the cellular and subcellular level of CCR down-regulated tobacco. It was shown 32 that xylem cells contained less lignin in the cell walls when compared to wild-type; in these cells, the S2 sublayer exhibits significantly reduced lignification, while the Si layer appears to be unaltered (Chabannes et al, 2001b). Kraft pulping of the highly suppressed transgenic tobacco achieved slightly greater delignification (lower kappa number) when compared to wild-type. Furthermore, CCR suppressed tobacco pulp subjected to bleaching produced pulp with improved brightness values. However, it has been suggested that severely reducing CCR will not likely produce ideal candidate plants for improved delignification during pulping because of their poor growth (O'Connelle/fl/., 2002). 1.4.1.vii Cinnamyl Alcohol Dehydrogenase (CAD) The CAD enzymes are responsible for the conversion of cinnamaldehydes to cinnamyl alcohols. There are reportedly two isozymes of CAD, with CAD2 being found in a variety of plants and CADI present in only a few species (Baucher et al, 1998). In Eucalyptus, it has been shown that CADI in unable to catalyze the reduction of sinapaldehyde to sinapyl alcohol, and has low affinity for both 4-coumaraldehyde and coniferaldehyde (Goffner et al, 1992). In contrast, CAD2 readily catalyzes these reactions in angiosperms (Sarni et al, 1984; Goffner et al, 1992; Baucher et al, 1998). Only one form of CAD exists in gymnosperms, and sinapaldehyde has been shown not to be a substrate for this enzyme (O'Malley et al, 1992; Galliano et al, 1993). In a number of plant species, CAD expression is fairly ubiquitous (Baucher et al, 1998). However, in poplar CAD expression is specifically associated with regions of active lignification (xylem and phloem) (Feuillet et al, 1995; Roth et al, 1997). The successful down-regulation of CAD has been performed on a number of plant species and has been shown to result in increased incorporation of cinnamaldehydes into lignin, and although growth appears to be normal, the xylem is a reddish colour (Halpin et al, 1994; Higuchi et al, 1994; Stewart et al, 1997; Chabannes et al, 2001a). Similarly, reduced 33 expression of CAD in transgenic alfalfa has been shown to reduce the incorporation of syringyl units into lignin (Baucher et al, 1999). A loblolly pine mutant with significantly decreased CAD expression and normal growth has been characterized and shown to exhibit lower levels of total lignin with increased coniferaldehyde monomer content (MacKay et al, 1997; Ralph and MacKay, 1997). In poplar, the down regulation of CAD does not appear to affect growth, total lignin content, or composition. However, the xylem of this poplar is also red in colour. Trees expressing the red colouration also exhibit increased incorporation of vanillin and syringaldehyde upon NaOH extraction (Baucher et al, 1996). Further research has shown that hydroxycinnamyl aldehyde is also a component of the lignin in these transgenic poplar (Ralph et al, 2001). Kraft pulping has been performed on down-regulated CAD trees and tobacco, and both transgenic species show lower residual kappa numbers (residual lignin) when compared to wild-type pulps (Baucher et al, 1996; Lapierre et al, 2000; O'Connell et al, 2002). Transgenic tobacco plants double transformed for the down-regulation of both CCR and CAD have also been characterized. These transgenic plants demonstrated a 50% reduction in total lignin, that was somewhat enriched in syringyl monomers. These plants also exhibit the unusual red coloration typical of CAD down-regulation, however, they appear to exhibit normal growth (Chabannes et al, 2001a). 1.4.1.viii Sinaple Alcohol Dehydrogenase (SAD) Recently, SAD, an enzyme likely only found in Angiosperm trees, has been characterized and shown to share very similar sequence homology with CAD (Li et al, 2001). Kinetic analysis of this enzyme has shown that SAD preferentially reduces sinapaldehyde to sinaple alcohol, although p-coumaraldehyde, caffealdehyde, coniferaldehyde, and 5-hydroxyconiferaldehyde represent useable substrates for this enzyme. This study further confirmed that CAD favours the catalysis of coniferaldehyde to coniferyl alcohol (Li et al, 34 2001). SAD and CAD expression appears to be spatially and temporally controlled, as CAD expression was detected in the protoxylem and metaxylem of vessel elements, concurrently with the presence of guaiacyl lignin in the primary xylem. In contrast, SAD was expressed in the protophloem parenchyma cells, which, upon maturation, become fibres with lignin primarily comprised of syringyl monomers. As these phloem fibre cells mature, CAD expression is detected and concurs with the late incorporation of guaiacyl monomers in this cell type. However, in the vessels of the developing secondary xylem, expression of SAD is almost negligible, and is consistent with the expression of CAD and presence of guaiacyl monomers in the lignin of these cells (Li et al, 2001). 1.4.1.ix Caffeate 0-Methyltransferase (COMT) / 5-Hydroxyconiferayl Aldehyde O-Methyltransferase (AldOMT) Initial research on COMT in angiosperms found that it could catalyze the formation of ferulate and sinapate (Bugos et al, 1991), while in gymnosperms caffeate was the only substrate used by this enzyme (Shimada et al, 1972; Kuroda et al, 1975). More recent research has shown that not only are these substrates not preferred for catalysis by COMT, but that the whole region of the pathway is non-existent (Humphreys et al, 1999; Maury et al, 1999; Osakabe et al, 1999; Li et al, 2000). It has been established that COMT in Populus preferentially catalyzes the formation of sinapaldehyde from 5-hydroxyconiferaldehyde, and is better suited for 5-methylation than 3-methylation. Furthermore, 5-hydroxyconiferyl aldehyde regulates COMT activity in a mixture of potential substrates. In light of these findings, it has been suggested that caffeate O-methyl transferase be known as 5-hydroxyconiferaldehyde O-methyltransferase (AldOMT) (Li et al, 2000). A novel, multifunctional conifer COMT, AEOMT (hydroxycinnamic acids/hydroxycinnamoyl CoA esters OMT) has also been found. This enzyme is capable of 35 using caffeic acid, 5-hydroxyferulic acid, caffeoyl CoA, and 5-hydroxyferuloyl CoA as substrates, while exhibiting similar substrate affinities for each (Li et al, 1997). An extensive amount of work has been performed in transgenic studies focussing on COMT.. Down-regulation of COMT in transgenic tobacco plants results in decreased total lignin, relatively larger xylem cells, and normal growth. In one study, the composition of lignin appeared to be unaffected (S:G), however, upon further analysis, was shown to exhibit altered S:G ratios (Ni et al, 1994; Sewalt et al, 1997b). In contrast, other research found only slightly reduced total lignin, coupled with reductions in the incorporation of syringyl monomers in the lignin (Dwivedi et al, 1994). Severely reduced COMT expression in transgenic tobacco and alfalfa produces plants with total lignin comparable to the control, however, markedly reduced S:G ratio. Furthermore, these results show the incorporation of a novel 5-OH guaiacyl unit in the lignin of transgenic plants and slightly increased guaiacyl monomers, hence more condensed lignin (Atanassova et al, 1995; Guo et al, 2001; Pincon et al, 2001). Similar results were found in transgenic poplar with severely reduced COMT activity, however, the xylem was a pale rose colour (Van Doorsselaere et al, 1995). More recent research indicates that the odd colouration was caused by the incorporation of coniferyl aldehyde residues into the xylem lignin (Tsai et al, 1998). Further, it has also been shown that transgenic COMT poplar incorporates 5-hydroxyconiferyl alcohol in the lignin (Ralph et al, 2001). Crosses between transgenic tobacco with reduced CCoAOMT and COMT expression have also been produced. The double transformed plants produced significantly less total lignin (close to 50%), with substantially increased S:G ratio. These transgenic plants grew as well as corresponding wild-type plants, however, the vessel elements appeared to exhibit crushed walls (Zhong et al, 1998). Similarly, transgenic tobacco double transformed for reduced expression of both CCR and COMT have been produced and analyzed. Some double transformed plants are derived from crosses of a significantly decreased COMT expressing line with a moderately 36 reduced CCR line, while others are from a significantly decreased CCR line with a moderately reduced COMT line. In the double transformed crosses, severely reduced expression of COMT accentuated the reduction of CCR expression, however, the reverse was not observed. An analysis of the ensuing lignin indicates that the lignin composition is dependent on the level of expression of each enzyme. For example, the cross with the most severe reduction in CCR activity displayed a significant decrease in total lignin, yet it was highly condensed. Interestingly, the S:G ratio was increased, and the novel 5-OH guaiacyl monomer was absent. In contrast, the other cross which had somewhat higher expression of CCR, produced plants expressing a lignin content and composition comparable to the controls (Pincon et al, 2001). Pulping experiments have subsequently been performed on COMT down regulated poplar trees. Depressed COMT expression in poplar produces lignin with a greater amount of guaiacyl lignin, and subsequently more condensed lignin, albeit at a lower total amount. Consequently, these trees were more resistant to kraft delignification and bleaching (Lapierre et al, 1999; Lapierre et al, 2000). COMT silencing has also been performed and shown to cause a significant reduction in total, yet highly condensed lignin. Pulp yields for these trees were shown to increase during the kraft pulping trials, yet this increase was likely a reflection of the higher residual lignin (kappa number) achieved with this pulp (Jouanin et al, 2000). CAD and COMT double transformed poplar have also been tested for their efficacy of delignification in the chemical pulping process, and these results indicated that delignification is dependent on the relative down-regulation of the individual enzymes. For example, double transformed trees with more severely depressed COMT than CAD expression were difficult to pulp, while those with more CAD reduction were easier to pulp, pointing to the benefit of CAD down-regulation for efficacious delignification (Lapierre et al, 1999; Lapierre et al, 2000; Pilate et al, 2002). 37 1.4.1.x Ferulate 5-HydroxyIase (F5H) / Coni fery l A ldehyde 5-Hydroxylase (CA ld5H) CAld5H/F5H is a multifunctional cytochrome P450-linked monooxygenase (Grand, 1984) that is capable of using ferulate, coniferyl alcohol, and coniferaldehyde as substrates for the hydroxylation of the C5 position (Humphreys et al, 1999). CAld5H was first characterized in poplar by its ability to catalyze the reaction from ferulate to 5-hydroxyferulate, hence its original name F5H (Grand, 1984). Subsequent, pivotal research has shown that, based on enzyme kinetics and substrate inhibition, CAld5H/F5H produces 5-hydroxyconiferaldehyde from coniferylaldehyde at a rate 140 times greater than for ferulate to 5-hydroxyferulate. Furthermore, when both coniferyl aldehyde and ferulate are available for CAld5H/F5H catalysis, 5-hydroxyferulate is not produced (Humphreys et al, 1999). With the construction of a yeast CAld5H/COMT system, and later with the same construct in transgenic poplar, it has been shown that these two enzymes are responsible for the catalysis of coniferyl aldehyde to sinapaldehyde, confirming a direct pathway from the synthesis of guaiacyl lignin to that of syringyl lignin residues (Osakabe et al, 1999; Li et al, 2000). These results are highly significant as they have redirected the theory of lignification and helped establish the syringyl lignin pathway, which is based on catalysis of free aldehydes, not cinnamic acids (Humphreys et al, 1999; Osakabe et al, 1999). In wild-type Arabidopsis, deposition of syringyl monomers initially occurs in the sclerified parenchyma surrounding the vascular bundles, and increases as these cells mature and produce secondary thickening. CAW5H/F5H was over-expressed, using a C4H promoter, as this promoter's expression occurs early in plant development and would likely not restrict expression of a C4H-CAld5H/F5H construct in early lignification. Over-expression resulted in lignin comprised almost entirely of syringyl monomers. Furthermore, this over-expression eliminated syringyl residue tissue specificity, and elevated expression of CAld5H/F5H meant syringyl monomers were deposited in the vascular bundle cells as well as in the sclerified 38 parenchyma. Equally significant is that morphologically, Arabidopsis plants with enhanced expression of CAld5H/F5H are indistinguishable from their wild-type counterpart (Meyer et al, 1998) . More recently, the CAld5H/F5H construct has been over-expressed in woody plants, tobacco and poplar. The construct was evidenced in the xylem and predominantly in the ray parenchyma. Additionally, no morphological differences relative to wild-type were apparent. The same predominance of syringyl monomers observed in the lignin of transgenic Arabidopsis was evident in these woody transgenic plants. Interestingly, the total lignin content remained comparable to wild-type, and delignification occurred more readily as evidenced by TGA (thioglycolic acid) derivatization and the Klason lignin method, which showed comparatively lower amounts (Franke et al, 2000). Analysis of the lignin produced by CAld5H/F5H transgenic Arabidopsis confirmed the predominate incorporation of syringyl monomers in the lignin. Furthermore, it revealed that 8-aryl ether linkages and erythro-stereochemistry predominates in this type of lignin (Marita et al., 1999) . Lignin monomer composition, the ratio of syringyl to guaiacyl units (S:G), has a dramatic affect on lignin chemical solubility. For example, hardwoods, which have a high S:G ratio are more readily delignified than softwoods, which have lignin comprised almost entirely of guaiacyl monomers (Chiang et al, 1988; Chiang and Funaoka, 1990). Although all the work performed on CAld5H/F5H suggests that woody plants genetically modified with this construct would result in greater delignification efficacy from an industrial perspective, by virtue of the significant increases of syringyl lignin monomers possible with this construct, to date no pulping trials have been attempted. 39 1.5 Objectives There were two primary goals for this research: to evaluate the wood of transgenic poplar over-expressing the C4H-F5H transgene and to evaluate the kraft pulp derived from this wood. Wood Analysis: • thoroughly investigate the effects of C4H-F5H over-expression on poplar wood chemistry • to establish whether the morphology of the same wood has been altered by the over-expression of the C4H-F5H transgene Pulp Analysis: • several pulping conditions were employed to evaluate the relative efficacy of kraft delignification of C4H-F5H over-expressing poplar wood, by analyzing pulp yield, residual lignin content and monomer composition • to evaluate the quality of the resultant pulp by determining the carbohydrate composition and viscosity, the fibre coarseness and length • to assess the ease with which pulp derived from transgenic lines responds to bleaching • and fiinally, to elucidate the relative molecular weight differences between the lignin liberated during the kraft pulping in the black liquor. 40 Chapter 2 Materials and Methods 2.1 Plant Material The generation of nine lines of C4H-F5H transgenic hybrid poplar (P. tremula x P. alba) were previously described (Franke et al. 2000), and together with wild-type control plants, were maintained as shoot cultures on MS (Murashige and Skoog Basal) medium with a 16 hour photoperiod. Shoots were multiplied from each line by excising nodal segments and allowing axillary buds to elongate. Prior to planting into the greenhouse, 5-8 cm long tips from actively growing shoots were excised and placed on MS medium supplemented with 0.01 uM NAA (a-Naphthalene-acetic acid) for two weeks to initiate root formation. Shoots were then transplanted directly into potting soil, acclimated 2 weeks in a high humidity environment, and grown and transplanted into successively larger pots over the next two years. Growth conditions in the greenhouse were approximately 25°C (March - November) and 14°C (December - February) without supplemental light. Watering was done at least twice .-a- week during the growing season with bi-weekly fertilization. One year old plants were harvested prior to dormancy in the fall. Two year-old plants were top-pruned at four feet above the soil and allowed to over-winter in 14°C greenhouse. This two year-old material was then harvested approximately 2 months after flushing in the spring. Wood designated as two years-old was therefore the bottom 4 foot section of each two year old tree. Leaves and bark were removed from the harvested stems, and these stems were then left to air dry at ambient temperatures in the laboratory. No significant differences in height were observed between the non-transformed wild type and the transformed material with the exception of line C4H-F5H 21 (Ellis, unpublished). Line C4H-F5H 21 grew slower and had noticable brown necrotic leasons on the leaves. With 41 the exception of line C4H-F5H 21, no differences in the timing of bud break, leaf morphology, or leaf size were observed between the transgenic and wild types 2.2 Moisture Content Determination The moisture content of wood chips, ground extractive-free wood, and pulp was determined following TAPPI standard method T 264 cm-97 for basic density and moisture content of pulp and wood. Approximately 4 g of air-dry wood chips or 2 g of extractive-free wood were dried for 24 hours at 105 °C, removed to a desiccator to cool, and the oven-dry weight was rapidly recorded to the nearest O.OOOlg. The moisture content was then determined using the following equation: Percent Moisture Content = (I-F/I) • 100 where: I = Air-dry weight of sample F = Oven-dry weight of sample 2.3 Wood Extractives Content The determination of total extractives in the wood samples was performed according to a variation of TAPPI Method T 280 pm-99. Wood samples (4-5 g of air-dry wood, ground in a Wiley mill to pass through a 0.4 mm screen) were extracted with reagent grade acetone (150 mL) in a Soxhlet apparatus heated to -55 °C for 12 hrs. The weight difference between non-extracted and extracted wood was used to determine percent of extractive in the dry wood. Although both 1-year and 2-year old wood samples were prepared as above, only the 2-year old wood was analysed by GC. The acetone was rotary-evaporated under reduced pressure at 40 °C to dryness. Extractives were re-solublized with 5 mL of HPLC grade acetone (Fisher Scientific) and 5 mL 42 deionized water, and stored in 15 mL scintillation vials. One mL of the extractives solution was filtered (through a Kimwipe packed pasteur pipette) then added to a GC vial, and their composition analyzed by GC using betulin, tetracosane, tripalmitin, and cholesteryl palmitate as standards. A Hewlett Packard (HP) 5890 Series II GC fitted with a HP 6890 Series injector and a 10m DB-XLB column (J&W Scientific) was used to perform the extractives analysis. The GC method employed a 1.0 fiL injection volume, with an initial injector temperature of 320 °C and a detector temperature of 330 °C. The initial oven temperature was set at 50 °C for 3 minutes, with a ramping rate of 10 °C/minute to 240 °C, were it was held for 3 minutes. The temperature was again increased by 10 °C/minute to 310 °C and held for 3 minutes, then increased once more by 10 °C/minute to 350 °C and held for 25 minutes. 2.4 Wood and Pulp Compositional Analysis 2.4.1. Wood and Pulp Total Lignin Determination Klason analysis was performed on duplicate or triplicate samples of 0.4 mm (40-mesh) screened Wiley-milled, extractive-free wood and pulp. Total wood and pulp lignin contents were found using a modified version of TAPPI standard method T222 om-98 for Acid insoluble Lignin in Wood and Pulp. Extractive-free, oven-dried wood equivalent to 0.2 g and 0.15 g of oven-dried pulp was treated with 3 mL of cold (4 °C), 72% H 2 S O 4 (Fisher Scientific) at an ambient temperature of 20 °C. The mixture was initially macerated continuously for 2 minutes, and then stirred every 10 minutes for 2 hours. Acid hydrolysis was stopped with the addition of cold de-ionized water, whereby the wood mixture was diluted with 112 mL of deionized water and the pulp with 112.5 mL of de-ionized water, to achieve a final acid concentration of 4% (w/w) H 2 S O 4 . The mixture was then transferred to a serum bottle, which was sealed with a septa cap and autoclaved (Castle Thermatic 60) at 121 °C for 1 hour. 43 Klason lignin was determined gravimetrically with the hydrolysates filtered through pre-weighed, medium coarseness, sintered-glass crucibles. The filtrate was added back to the septa bottles and re-filtered to ensure recovery of all solids, and the filtrate retained. The solids were then washed with 100 mL of 40 °C deionized water and oven-dried at 105 °C for 12 hours. The oven-dried crucibles with acid-insoluble lignin were then weighed. The Klason lignin filtrate was further analysed using TAPPI Useful Method UM-250 to determine the portion of acid soluble lignin. 30 fxL of wood hydrolysate samples were diluted with 970 \xL of 4% H2SO4 in a test-tube and mixed thoroughly, such that spectophotometer (Milton Roy Spectronic 1001 Plus) absorbance readings were between 0.2 and 0.7 absorbance units (AU) at 205 nm. In contrast, the pulp filtrate was diluted by 50% with acid to achieve readings between 0.2-0.7 AU. The solution was transferred to a quartz cuvette, 4% H2SO4 was used to calibrate the spectrophotometer and then absorbance values for each sample were taken. An Expression of Beer's Law is used to calculate the percent of acid-soluble lignin as follows: Acid-soluble lignin % = B-V-100/1000-W Where: B - Absorbance • volume of diluted filtrate /110 • volume of original filtrate V = Total volume of filtrate W = Oven-dry weight of wood 2.4.2. Wood and Pulp Carbohydrate Analysis The Klason lignin filtrate was filtered through 0.45 mm HV filters (Millipore, MA, USA) prior to injection of a 20 fiL sample volume. The HPLC system (Dionex DX-600, Dionex, CA, USA) was equipped with an ion exchange PA1, 4 x 250 mm column (Dionex), an ED50 Electrochemical detector (Dionex), and an AS 50 autosampler (Dionex). The column was equilibrated with 250 mM NaOH (BDH), eluted with deionized water at a flow rate of 1.0 44 mL/min., and post-column wash of 200 mM NaOH at a flow rate of 0.5 mL/min. Fucose (Sigma) (5 mg/mL) was used as an internal standard. 2.5 Determination of Wood and Pulp Lignin Monomer Ratio (S:G) The lignin monomer composition of the starting wood material and that of the ensuing pulp was determined with duplicate samples following the thioacidolysis procedure (Rolando et al, 1992). 2.5.1. Determination of Wood Lignin Monomer Ratio (S:G) Briefly, the solvolysis step began with extract-free wood ground through a 40-mesh screen using a Wiley mill and extracted for 12 hours with acetone in a Soxhlet apparatus. A 10 mg sample was weighed into a 25 mL Kimax test tube fitted with a Teflon coated cap. A 10 mL aliquot of freshly prepared reaction mixture (1:4:40 ratio of boron trifluoride diethyl etherate (Sigma), ethanethiol (Sigma), and dioxane (Fisher Scientific), respectively) was added to the wood. Nitrogen gas was used to displace the air, and the cap was closed tightly. The reaction was allowed to proceed in a 100 °C silicon oil (Aldrich) bath for exactly 4 hours and was mixed by shaking every 30 minutes. Afterwards, the mixture was removed to an ice bath for 5 minutes to stop the reaction. 3 mL of deionized water and 2 mL of internal standard composed of 0.25 mg/mL tetracosane (Aldrich) in dichloromethane (Fisher Scientific), were then added to a separation funnel followed by the addition of the reaction mixture. The tubes were rinsed twice with 10 mL of deionized water and the contents emptied into the separation funnel. Approximately 4 mL of 0.4 M sodium bicarbonate in water was added to attain a pH of 3-4. Next, 30 mL of dichloromethane was added and the mixture in the separation vial was shaken well, left to separate for 5 minutes and then the lower fraction collected into an Erlenmeyer flask. The addition of 30 mL of dichloromethane was repeated twice more and the lower fractions pooled. The mixture was dried over excess Na2SC>4 (Fisher Scientific), and filtered 45 through Whatman #4 filter paper into a 250 mL round bottom flask. The Erlenmeyer flask was rinsed twice with 15mL dichloromethane and the contents of the filter paper rinsed once with 30 mL dichloromethane. The total pooled dichloromethane mixture was evaporated at ~40 °C under reduced pressure (using a Biichi RE III (Switzerland) rotovap) until approximately 3 mL remained. 4 mL of methanol (Fisher Scientific) was added to the 3 mL mixture and evaporated (as described) to dryness. The residue in the round bottom flask was re-suspended in 2 mL dichloromethane, transferred to a 5 mL vial fitted with a Teflon cap, wrapped with aluminium foil and stored at 4 °C. After all of the samples had been prepared, the samples were silylated. In short, 10 /xL of pyridine (Fisher Scientific), 10 /*L of sample, and 50 J M L of N,0- bis(trimethylsilyl)acetamide (Sigma-Aldrich) were added to a gas chromatography vial insert, and briefly agitated by hand. The reaction was allowed to proceed for a minimum of 1.5 hours before being injected into the GC for analysis. Gas chromatography (GC) analysis was performed on a HP 5890 Series II, using a HP 6890 Series injector and a 15 m x 0.25 /mi DB-5 column (J&W Scientific). The GC method used a 2.0 /xL injection volume, an initial injector temperature of 250 °C, and a detector temperature of 270 °C. The initial oven temperature was set at 130 °C for 3 min., with a ramping rate of 3 °C/min. to 260 °C at which temperature it was held for an additional 5 min. The concentration of syringyl and guiacyl lignin was determined from the response factor for each, as the ratio of relative concentration to the relative peak area of the internal standard. 2.5.2. Determination of Pulp Lignin Monomer Ratio (S:G) Pulp thioacidolysis was performed as described above with the exception of the amount of substrate initially used. As the pulp samples contain minimal amounts of lignin, 40 mg of 46 moisture-free, 0.4 mm mesh screened pulp was employed for thioacidolysis determination of pulp lignins, while all other reaction solvents and conditions remained the same. 2.6 Derivitization Followed by Reductive Cleavage (DFRC) Monolignol analysis of wood was performed using the DFRC method developed by Lu and Ralph (1997a). Briefly, the extract free wood was ground to pass a 40-mesh screen using a Wiley mill. The initial acetyl bromide step employed 50 mg of sample in a 100 mL round bottom flask. To this, 4 mL of glacial acetic acid (BDFf), 0.5 mL of internal standard (0.05 M tetracosane (Aldrich) in dioxane (Fisher Scientific), and 1.0 mL acetyl bromide (Sigma) was added; the flask was stopped and sealed with parafilm. This mixture was stirred (magnetic stirrer) in a 50 °C water bath for exactly 3 hours. The solvent was removed using rotary evaporation under reduced pressure (Rotovap) at 45 °C to near dryness. For the reductive cleavage step the residue in the flask was solublized in 5.0 mL of 5:4:1 dioxane: acetic acid: deionized water. Then 250 mg zinc metal dust (<10 microns 98+% Fischer) was added, and left to stir for 30 minutes at room temperature (-20 °C). When the reaction was complete, the solution was filtered through Whatman #4 filter paper into a 50 mL separation funnel. The round bottom flask was washed twice with 10 mL methylene chloride (Fisher Scientific) and this solution was added to the separatory funnel. Further, 10 mL of saturated NH4CI (Nichols Chemical) solution was added to the separatory funnel and the resulting mixture was well shaken, and the lower fraction collected into an Erlenmeyer. The aqueous layer was extracted two more times with 5 mL of methylene chloride, collected and then pooled. The pooled fractions were added to a separatory funnel with 10 mL of deionized water, shaken and the lower fraction collected in a flask. This lower fraction was dried with excess NaSC>4 (Fisher) and left at -20 °C for 30 minutes. This solution was then filtered through 47 Whatman #4 filter paper into a round bottom flask, the residual sodium sulphate was washed twice with 10 mL of C H 2 C I 2 , and the solution added to the filter. The sodium sulphate in the filter paper was then washed with 5 mL C H 2 C I 2 . The pooled filtrate was evaporated under reduced pressure (Rotovap) at -40 °C to dryness. The acetylation step followed with a 1 mL addition each of acetic anhydride (Fisher Scientific) and pyridine (Fisher Scientific) being added to the residue in the round bottom flask, and mixed for 40 minutes at 20 °C. Once again, the solution was evaporated under reduced pressure at ~40 °C to near dryness, and resolublized in 5 mL ethanol (Fisher Scientific HPLC grade) to co-evaporate the volatile compounds. This process was repeated three times and finally brought to dryness. The residue was solublized in 1 mL of dichloromethane transferred to a GC vial insert and wrapped in aluminium foil and stored at 4 °C until GC/MS analysis was performed to quantitatively evaluate the degraded lignin monomers. The GC-MS analysis was performed by GC (Varian 3800), with a Varian 1079 injector, a 15 m x 0.25 11m DB-1 column (J&W Scientific), and He as a carrier gas (1.0 mL/min). 1.0 J U L injection volume with 1:10 split was injected, with an injection temperature of 250 °C, and detector temperature of 300 °C. The initial temperature was set at 140 °C and held for 1 minute, with a ramping rate of 3 °C/min. to 240 °C, which was held for 30 seconds then the temperature was increased by 30 °C/min. to 310 and finally held for 10 minutes. The mass spectrometry data was collected using a Saturn 2000 MS/MS detector and the same column and temperature program described. Syringyl and guaiacyl moieties were identified using standards kindly provided by John Ralph. 2.7 Fibre Quality Analysis of Wood Air-dry 2-year old wood stems were cut into slivers such that the dimensions were approximately 2 mm x 2 mm x 30 mm (tangential plane). Several slivers of wood from each 48 sample were reacted in Franklin solution (1:1 mixture of 30% peroxide (Fisher Scientific) and glacial acetic acid (BDH)) for 48 hours at 70 °C. Following the reaction, the solution was decanted off and the dissolved wood samples were removed to coarse crucibles with deionized water. The wood was suction filtered with multiple washes of deionized water until a neutral pH was achieved. Deionized water was used to wash the wood from the crucibles into pre-weighed, 20 mL scintillation vials and the samples dried in a 105 °C oven for 12 hours and their weight determined. The samples were re-dissolved, by shaking, in 10 mL of deionized water and stored at 4 °C for analysis. The entire dissolved wood sample was blended with 1 L of distilled water such that the wood fibres were freely suspended. The sample was swirled rapidly by hand and three 50 mL scoops were removed to a 5 L beaker and the weight recorded. This sample was then diluted to a final volume of 4 L and the weight recorded. Again, the solution was agitated by hand-mixing, and approximately 150 mL was transferred to a 700 mL beaker and the weight recorded, this sample was auto-diluted by the FQA to a 600 mL volume. Triplicate samples were tested. The final dilution was adjusted to achieve between 25-40 fibres per second in the Fibre Quality Analyzer (Code LDA96 OpTest Equipment Inc). Fibre length was measured optically by the FQA and coarseness for the wood samples was determined using the following equation: Coarseness (mg/m) = Mass of oven dried fibre tested (mg) / L T (m) Where, L T (m) (total fibre length) = Fibre Total • Ln [mean arithmetic length (mm)] • 1 m/1000 mm Mean arithmetic length (mm) = E njLj/ E nj n = fibre count L — contour length 49 2.8 Fibre Quality Analysis of Pulp Fibre Quality Analysis of pulp samples was performed using a modified version of the method outlined in the FQ Analyzer operation manual (Code LDA96, OpTest Equipment Inc., Hawkesbury, ON, 1996), with pulp hand-sheets prepared for testing as described by Seth (1997). Two grams of representative pulp from each samples was carefully teased away from an air-dry pulp pad and measured to an accuracy of 0.001 g. The pulp was blended with 1 L of distilled water in a blender, until the pulp was thoroughly dispersed in a slurry. The fibres were then added to a British hand-sheet machine filled with distilled water and fitted with six sample rings (27 mm inner diameter, 31 mm outer diameter and 15 mm high) on the couch screen. Pulp was dispersed and then drained. The fibre was removed to blotting paper. The hand-sheet was allowed to air-dry to near dryness, then three of the mini-sheets were removed, stored in 20 mL scintillation vials and oven-dried for 12 hours to determine the average oven-dry weight of each mini-sheet. The other three mini-sheets were removed to three beakers and re-suspended in 500 g (recorded to a 0.01 g) of distilled water. This slurry was thoroughly mixed and between 50-200 g (recorded to a 0.01 g) removed to a 700 mL beaker. This aliquot of pulp slurry was then auto-diluted to approximately 600 mL volume and auto-sampled by the FQ Analyzer. The amount of the final aliquot of pulp slurry is dependent on the fibre content of individual samples, such that the EPS (events per second) is between 25-45 during the FQ analysis. Fibre lengths were expressed as arithmetic means by the following equation: /„ = E,«,7//I;,«/ The FQA groups fibres into length classes therefore, Hi = the number of fibres in the length class /, 50 2.9 Image Analysis Image analysis of 2-year old poplar was performed using the method described by Wang and Aitken (2001). Whole sections of wood stems were saturated with distilled water and 18.5 /mi thick cross-sections prepared using a microtome (Spencer, Buffalo) avoiding knots and tension wood. Sections were stained with 5 g/L safranin (Fisher Scientific) in distilled water for approximately 30 minutes, rinsed with distilled water and set in a drop of glycerine (Fisher Scientific) on a glass microscope slide. Cytosine glue was used to seal a glass coverslip over the prepared cross-section. Slides prepared this way were stored at 4 °C for later analysis. A Meiji microscope (ML9000) equipped with a video camera linked to a PC running Sigma Pro Image Analysis (version 2.0) software was used to capture images of the cross sections. A 4x magnification was used for the objective lens and 35x for the video camera to provide optimal resolution. A micrometer having 9.0 pixels equivalent to 10 fun was used to calibrate the horizontal scale, while 7.6 pixels was equivalent to 10 um on the vertical scale. Ten images of each cross section were made to cover approximately 50% of each cross section. Roughly 2500 cells were measured to determine cell wall thickness and average lumen area per sample. 2.10 Specific Gravity of 2-Year Old Trees The specific gravity of the 2-year old tree stems was determined using ASTM Standard Test Methods for Specific Gravity of Wood and Wood-Base Materials (D 2395-93 re-approved 1997) Method B - Volume by Water Immersion, Mode IV. Three different 2-year old poplar stems (free of bark) from each sample were cut into 3-4 cm lengths and oven-dried for 24 hours at 105 °C and the weight determined. A 50 ± 0.4 mL graduated cylinder was filled to 30 mL and the pre-weighed oven-dry specimens were 51 submerged, without touching the sides of the cylinder by using a thin-straight pin. The difference between the initial and final water volume was recorded and is equivalent to the volume of the wood sample. Specific gravity was then determined using the following equation: Specific Gravity = W/V where: W = weight of oven-dry specimen (g) V = volume of oven-dry specimen cm 2.11 Kraft Pulping Process Initially, the branches and bark were removed from the 1 year old poplar, and the stems chopped into uniform dimensions (approximately 2 x 0.5 x 0.5 cm) and left to air-dry (20 °C) for approximately 5 days. Three different laboratory scale Kraft pulping conditions were performed on each of the nine poplar samples (1 wild-type, 8 transgenic), with the exception of F5H 21, which had significantly less substrate available for analysis than the other samples, and was therefore not cooked at the lowest H-factor. The Kraft pulping conditions were varied with respect to H-factor, a numerical value that represents the cooking variables, time and temperature. H-factors of 1547 (time @ Temp. = 90 min.), 1096 (time @ Temp. = 60 min.), and 856 (time @ Temp. = 45 min.) were used to attain the different Kraft pulp. Table 1 summarizes the Kraft pulp cooking conditions that were used. Pulping was performed in a 500 mL pressurized reactor in a circulating oil-thermostatic bath. Two-year old trees were prepared for pulping in a similar fashion. However, a different pulping reactor was used, which provided slightly different H-factors. Only two cooks were performed, one with an H-factor of 1483 and the other of 796. The cooking conditions are 52 9' ft " b Table 1. Kraft Pulping Cooking Conditions for 1-Year Old C 4 H - F 5 H Transgenic Poplar. Wood A i r Dry Weight Wood Oven Dry Weight Liquor Volume - N a O H ( B D H ) - N a 2 S (Fisher Scientific) Active A lka l i Effective A l k a l i Sulfidity Liquor: Wood Temperature Time to Temperature Time at Temperature Pressure 31-33 g 30 g 300 m L 18.3 g/L 16.0 g/L 25% 10 170 °C 30 minutes 45, 60, 90 minutes for transgenic lines and 45, 60, 90, 120 and 150 minutes for wild-type. 160 psi Table 2. Kraft Pulping Cooking Conditions for 2-Year Old C 4 H - F 5 H Transgenic Poplar. Wood A i r Dry Weight 21-22 g Wood Oven Dry Weight 20 g Liquor Volume 200 m L Active A lka l i 18.3 g/L Effective A l k a l i 16.0 g/L Sulfidity 25% Liquor: Wood 10 Temperature 170 °C Time to Temperature 30 minutes Time at Temperature 45 and 90 minutes Pressure 160 psi 53 summarized in Table 2. Laboratory pulping was performed using a 400 mL pressurized reactor in a circulating oil-thermostatic bath. For both series of Kraft cooks, the reactor was removed from the oil bath at the appropriate time and cooled in an ice-water bath. The spent liquor was reserved and the pulp was washed with water, and blended in a British Disintegrator for 15 minutes. The resulting pulp was filtered and washed until the filtered wash water was clear. Pulps were dried over night at 50 °C left to achieve ambient air moisture and weighed for total pulp yield and moisture content determination. The pulp pads were kept in airtight packaging until further analysis. 2.12 Kappa Number Determination The residual lignin content of the pulp was determined using TAPPI Useful Method (UM246) for Micro Kappa Number. Kappa number determination was performed in triplicate using air-dried pulp samples. Pulp was weighed and dispersed in 50 mL of deionized water, then transferred to a beaker and all the pulp was removed by rinsing with water to achieve a total volume of 80 mL. The mixture was continuously stirred to maintain the pulp suspension. A mixture of 10 mL 4 N sulphuric acid (Fisher Scientific) and 10 mL of 0.1 N potassium permanganate (Aldrich) was made and immediately added to the stirring pulp suspension. After 5 minutes the temperature was recorded and the reaction allowed to proceed for a total of 10 minutes. The reaction was stopped with 2 mL of potassium iodide (BDH). A titration, with 0.1 N sodium thiosulphate (Sigma), was performed and the volume of sodium thiosulphate consumed by the pulp mixture was recorded. Soluble starch for iodometry (Fischer) was used as the indicator. The kappa number calculation may be modified to account for variations in permanganate consumption varying from 10-70%, however, the amount of pulp used was varied for samples in order to achieve a pulp consumption of sodium thiosulphate equivalent to 50% of 54 that consumed by the blank (same method described above without the addition of pulp) to increase accuracy. Accuracy is also improved by maintaining the temperature at 25 °C, and therefore all the chemicals used in this experiment were kept in a 25 °C water bath. However, the calculation may also correct for temperature variations of 20-30 °C. Kappa Number was calculated as follows: K = (100F/WN) [1+0.013(25-T)] Where K = kappa number F = factor for correction to 50% permanganate consumption W = weight of moisture - free pulp (mg) N = volume of 0.1 N permanganate solution consumed (mL) T = reaction temperature at 5 min. (°C) 2.13 Pulp Viscosity TAPPI Method 230 om-94 was used to compare the viscosity (degree of polymerization) of the cellulose polymers derived from the pulp samples produced from the various cooks. Air-dry pulp samples were weighed to 0.2500 ± 0.0005 g of moisture-free pulp in 118 mL dissolving bottles fitted with a stopper. Five 6 mm diameter glass beads were placed in each bottle and 25 mL of deionized water added. The bottle was capped and shaken vigorously to disperse the pulp fibres (approximately 10 minutes) and left to stand 2 min. Then 25 mL of 1.0 ± 0.02 M cupriethylenediamine (CED) (Fisher Scientific) solution was added to achieve a 0.5% pulp in 0.5% CED solution. The air was evacuated with nitrogen gas for 5 minutes, the stopper replaced, and sealed with Parafilm and shaken with a Burrell wrist action shaker for 25 minutes to completely dissolve the pulp. The sample was then left in a horizontal position to facilitate degassing. A pair of capillary viscometers (size number 300) was secured to within 1° 55 of vertical in a 25 ± 0.5 °C constant temperature water bath. Duplicate samples were measured simultaneously by adding 10 mL of each pulp solution to the wide diameter arm of the viscometer and allowed to equilibrate to temperature. The solution was then drawn up the narrow arm of the viscometer beyond the upper measuring mark and allowed to drain down to wet the walls of the viscometer. The solution was again drawn up and the time recorded for the meniscus to pass between the upper and lower marks; this measure was performed in triplicate for each pulp sample. The viscometers were calibrated using standard viscosity oils (Brookfield Engineering Lab, Stoughton, Ma) and pulp viscosity determined by the following equation: V=CTD Where: V = viscosity of CED solution at 25 °C, mPa-s C = viscometer constant (mPa-cm3/g) T = average efflux time (s) D = pulp solution density (1.052 g/ cm3) 2.14 Pulp Bleaching A standard DED sequence (D = chlorine dioxide and E = sodium hydroxide) was used to bleach 10 grams of oven-dry weight pulp. A 10% consistency was used for all stages. Briefly, for the Do stage, the pulp sample was placed in a double-layered polyethylene bag with distilled water and thoroughly mixed by hand to disperse the fibres. A charge of 0.5% chlorine dioxide and 0.05% sodium hydroxide (BDH) was applied to the pulp, and the mixture was then kneaded for 1.5 minutes to provide adequate dispersion of the chemicals. The sealed bag was weighted down in an 80 °C water bath and the reaction allowed to proceed for 78 minutes. The pulp mixture was then cooled with cold water and the pulp placed on a polymer filter in a Buchner 56 funnel and washed with tap water and vacuum filtration until filtrate was clear. The wet pulp pad ws then weighed. For sodium hydroxide extraction, the wet pulp was placed in a double sealed bag, with distilled water and mixed to disperse the pulp, and then a 1.3% NaOH charge was added to achieve a 10% consistency. Again, the sample was kneaded for 1.5 minutes, and then submerged in a 78 °C water bath for 42 minutes, after which the pulp was removed and washed as previously described. For the final chlorine dioxide stage, the pulp was allowed to achieve ambient moisture content which was determined with the method described in section 2.1, except that triplicate pulp samples of 0.4 g each were used. The leftover pulp, approximately 9 g, was then used for the final bleaching stage. The concentration of chlorine dioxide was determined by titration, then the pulp was again placed in a double polyethylene bag with distilled water to disperse the pulp, and a sufficient amount of 0.3% CIO2 was added. The D] stage was allowed to proceed for 194 minutes at 70 °C. Four-gram brightness pads were made from' the bleached material (maintained at 23 °C and 50% humidity) using the procedure outlined in C.P.P.A. Standard C.5 Approved Method, June 1973 - Forming Handsheets for Optical Tests of Pulp. A Technobrite Mirco TB-1C was used to perform brightness measurements at 457 nm. 2.15 Molecular Weight of Black Liquor Lignin Determination Kraft lignin was isolated via acidification of the black liquor collected from the various cooks. Briefly, cold (4 °C) black liquor was shaken vigorously to thoroughly mix the lignin solution and then poured into 600 mL beaker containing a magnetic stir bar. The sample was stirred and heated to 60 °C then filtered through a Whatman #3 filter. 100 g of filtrated was weighed into a 500 mL centrifuge tube then stirred with a magnetic stir bar while dilute hydrochloric acid was added drop-wise to reduce the pH to 2.0. The acidified black liquor was 57 then centrifuged (Sorvel RC 24 DuPont) using a GS-3 rotor at 7000 rpm for 20 minutes at 20 °C. The liquid portion was discarded and the pellet resolubilized with deionized water, the solution was acidified with dilute hydrochloric acid to attain a pH of 2.0 and centrifuged as described. These washing processes were repeated four times or until the liquid portion was clear. The thoroughly washed pellet was left to dry at 50 °C over night then dried over sodium pentoxide for two weeks. 50 mg of dry kraft lignin was weighed into scintillation vials and solubilized with 5 mL of 0.1 M sodium hydroxide (NaOH) and 0.1 M lithium chloride (LiCl) and ~1 mL of this solution was added to a 1.5 mL vial fitted with a screw cap having a septum top. Samples were analyzed by size exclusion chromatography with a Dionex 500 fitted with a column filled with 20-40 /xm size exclusion packing material (Toyopearl® HW-40 S by Supelco, Bellefonte, PA) and eluted with 0.1 M LiCl and 0.1 M NaOH solution at 0.3 mL/min. and an AD 20 absorbance detector at 280 nm. 58 Chapter 3 Wood Analysis Wood from C 4 H - F 5 H transformed poplar and wild-type trees was analyzed chemically and morphologically. Because of the limitations associated with individual techniques, several different methods were used to evaluate the lignin composition and quantity. The next section (Wood Lignin Analysis) w i l l provide an overview of the attributes and constraints of the methods that have been employed for lignin analysis. A comparison has also been made o f the wood carbohydrate and extractives content in the transgenic trees and the wild-type. Concurrently, the cellular morphological characteristics were determined. A microscopic evaluation of the cell morphology in cross-section has been performed, coupled with the determination of fibre lengths for all tree lines. The purpose of these analyses was to determine to what extent over expression of C 4 H - F 5 H in transgenic poplar impacts their lignin content and to identify any effects to the cell wall , which may have occurred as a result of this over expression. 3.1 Wood Lignin Analysis In order to determine the ratio of syringyl to guaiacyl monomers in the lignin both thioacidolysis and derivitization followed by reductive cleavage ( D F R C ) procedures were employed. D F R C is a very sensitive analytical tool, and is capable of providing additional information regarding the monolignol composition of wood. The total amount of lignin in both the transgenic lines and the wild-type was evaluated using the Klason lignin method. Klason lignin is the acid insoluble portion of the lignin, while the Klason lignin filtrate contains the acid-soluble component of the total lignin. Lignin degradative methods, which reduce the lignin macromolecule into lower molecular weight products, have provided a wealth of information regarding bonds formed in lignins. However, they are subject to lower yields and the formation o f side reactions. For 59 example, lignin side chains are truncated during nitrobenzene oxidation (NBO), thereby preventing characterization of these substituents (Lapierre et al, 1995). Furthermore, nitrobenzene oxidation, which preferentially cleaves 8-0-4 bonds, achieves relatively low yields, approximately 35% of Klason lignin for spruce (Chen, 1992). Thioacidolysis is also a lignin degradative procedure. In this analysis, lignin is reduced to smaller products via a reaction catalyzed by an anhydrous mixture of boron trifluoride etherate, a hard Lewis acid and ethanethiol a soft nucleophile, which cleaves the arylglycerol-/?-aryl ether linkages (Figure 11) (Rolando et al, 1992). The thioacidolyis method is based on the acidolysis procedure; although both selectively cleave arylglycerol- /3-aryl ethers the thioacidolysis silylation products are more readily produced and are more stable.. A comparison between acidolysis and thioacidolysis indicates that thioacidolysis is also a more sensitive method (Lapierre and Monties, 1986). Although thioacidolysis has several obvious advantages over other degradative techniques, it is not ideal. It has been noted that one of the reactants, ethane thiol, is unpleasant to work with, as well, some substituent degradation and incomplete cleavage may occur during the thioacidolysis reaction (Ralph and Grabber, 1996; Lu and Ralph, 1997a, 1997b, 1998). A more recent alternative is derivitization followed by reductive cleavage (DFRC) protocol (Lu and Ralph, 1997a). Figure 12 is a proposed mechanism for the DFRC cleavage of ethers. The major DFRC derived monomers for hardwoods are 4-acetoxy-3-methoxcinnamyl acetate (coniferyl peracetate, G) and 4-acetoxy-3,5-dimethoxycinnamyl acetate (sinapyl peracetate, S) (Lu and Ralph, 1998). The chromatograms obtained with this method are cleaner than those for both acidolysis and thioacidolyis, suggesting that DFRC produces fewer side reactions (Lu and Ralph, 1998). Furthermore, the end groups coupled to monomers with 4-0-/3 ether bonds, a-60 HO H2 \ / 7 *f p C — O R C — O R R = Ar, H R., = H or OCH 3 HO H2 \ / C I C — O R E t 2 0 + - B F 3 R1 - R1 EtSH (-ROH) Et Q0 +--B-F, HO H2 \ / C / V C — O — R EtS CH B f , EtSH (-ROH) EtS. .H2 C I E t S — C H E t S — C H Et 20 +--B"F 3 EtSH (-H 20) Products MeO CHR 1 C H R r -CHR, R1 = SC2H5; R2 = H (guaiacyl units); R2 = OCH3 (syringyl units) Figure 11. Thioacidolysis mechanism proposed by (Rolando et al, 1992). The initial reaction is the B F 3 substitution at the benzyl alcohol or benzyl ether of Coc. This is followed by thioester participation at CP and substitution resulting in ether cleavage. The hydroxyl group at Cy proceeds in the same fashion. The reaction produces erythro and threo isomers. 61 AcBr Derivatization and Solubilization OMe R1 | R2 ,0 [Ac] Zn Reductive Cleavage OAc R1 | R2 ,OAc Ac 20/Py Acetylation OAc R1 R2 ,0 [Ac] R = H or Aryl Guaiacyl: R1 = H, R2 = OMe Syringyl: R1 = R2 = OMe Figure 12. Derivitization followed by reductive cleavage mechanism proposed by (Lu and Ralph, 1997a). P-0-4 alkyl-aryl-ether bonds are specifically cleaved. Bromination of Ccc and acetylation of available hydroxyl group occurs. This is followed by zinc reduction of the bromine and cleavage of p-0-4, and subsequent acetylation of remaining hydroxyls. Products include coniferyl diacetate that represents guaiacyl and sinapyl diacetate represents syringyl monomers. 62 carbonyl units (keto), and other constituents present in small amounts may all be evaluated by DFRC (Lu and Ralph, 1998). These are in addition to the primary products derived from syringyl and guaiacyl monolignols, which are available from a and /3-aryl ether cleavage (Lu and Ralph, 1997b, 1998). The yield for DFRC products is also very high, 92-97% of model lignin compounds, 52-75% yield occurs with thioacidolysis, and 32-69% for acidolysis (Lu and Ralph, 1997b). The deriviatives of thioacidolysis and DFRC may be analyzed using either gas chromatography (GC) or GC mass spectrometry (GC/MS). Acid-insoluble lignin or Klason lignin is isolated through the acid hydrolysis of wood carbohydrates. Therefore, the lignin is collected as a residue and weighed. In softwoods, Klason lignin closely approximates the total weight of lignin in the wood. In contrast, hardwoods .have a large amount of the acid soluble lignin that is not accounted for in the gravimetric Klason determination, and therefore, it is necessary to analyze the Klason lignin filtrate for acid-soluble lignin (Dence, 1992). Klason lignin is often cited as the standard for the determination of total lignin upon which other analyses' percent yields are based. Because of the condensation reactions that occur with acid hydrolysis, Klason lignin does not offer structural information (Stenius, 2000). 3.1.1 Determination of Lignin Monomer Composition As described above, thioacidolyis and DFRC are complimentary methods for the determination of lignin monomer composition. Using the thioacidolysis protocol (Rolando et al, 1992), it was apparent that the one-year old trees contain syringyl and guaiacyl lignin ratios ranging from 1.9 for the wild-type to a maximum of 14.2 in line F5H-64 (Table 3). These findings were confirmed by the DFRC results (Table 3). An example of a GC chromatogram of the DFRC products is provided in Figure 13. The thioacidolysis results differ from those obtained in previous work performed initially on the 3-6 month old C4H-F5H lines (Franke et a/., 2000). However, these authors used nitrobenzene oxidation analysis which indicated that 63 Table 3. Syringyl and guaiacyl monomer content of wild-type and C 4 H - F 5 H transformed poplar as determined by thioacidolysis and derivatization followed by reductive cleavage (DFRC). Tree Line Total G Units (/xmol/g)* Total S Units (itmol/g)* S:G* M o l % Syringyl Lignin Thioacidolysis D F R C Wild-type 254.4 484.2 1.9 65.6 64.7 C 4 H - F 5 H 85 179.0 429.9 2.4 70.7 69.8 C 4 H - F 5 H 2 1 139.7 540.6 3.9 79.5 78.8 C 4 H - F 5 H 82 114.0 538.1 4.7 82.5 81.5 C 4 H - F 5 H 41 110.3 561.0 5.1 83.6 82.7 C 4 H - F 5 H 37 73.9 587.7 8.0 88.8 , 87.8 C 4 H - F 5 H 65 59.8 492.2 8.2 89.1 88.0 C 4 H - F 5 H 26 74.2 612.6 8.3 89.2 88.8 C 4 H - F 5 H 64 58.9 836.5 14.2 93.4 92.6 * Values determined from thioacidolysis analysis Results are the average of two replicates falling within a range o f 0.004-0.5. 64 wild-type trees have a mole percentage of syringyl (mol % S) of 59 and transgenic lines with mol % S ranging from 55-85, while our results have a wild-type mol % S of 65 and a range of 71-93.5 for transgenic lines. This overall rise in S:G ratio with increasing age of trees has previously been reported (Lapierre et al, 1999; Pilate et al, 2002). Furthermore, the value obtained for our wild-type sample is comparable to results obtained by others analyzing poplar by thioacidolysis (Lapierre et al, 1995). To date, over-expression of a phenylpropanoid pathway enzyme as a means of altering lignin content has only been attempted with the F5H gene, all other investigations of this type have focused on gene suppression or down-regulation in an effort to elucidate which enzymes are pathway constituents or to reduce lignin content. However, in many instances enzyme down-regulation has caused alterations to the composition of lignin produced by these transgenic plants. A comparison of stem lignin composition between wild-type and anti-sense suppressed 4CL transgenic Arabidopsis plants shows that S:G ratio could be increased by a factor of nearly 2.5 (Lee et al, 1997). However, in transgenic poplar with 4CL suppression the S:G does not appear to be altered from wild-type (Hu et al, 1999). Reduced expression of CCR in tobacco and of CCoAOMT in tobacco, alfalfa, and poplar has also resulted in S:G ratio increases. However, in all instances the guaiacyl monomer content was decreased to a greater extent than were the syringyl monomers. Double transformed transgenic tobacco with severely reduced CCR expression and moderately reduced COMT expression has also been characterized, and shown to exhibit increased S:G ratio (Pincon et al, 2001). However, these results suggest that a greater proportion of the guaiacyl monomers are participating in condensed bonds, such that the increased S:G ratio reflects the inaccessibility of these condensed guaiacyl units by the thioacidolysis method employed in this experiment (Pincon et al, 2001). Consequently, these changes in S:G ratio do not reflect true increases in syringyl units as has been observed in the 65 C4H-F5H over-expressing poplar (Piquemal et al, 1998; Zhong et al, 1998; Meyermans et al, 2000; Chabannes et al, 2001a; Guo et al, 2001). A significant amount of transgenic research focussing on altering the expression of other enzymes in the phenylpropanoid pathway has also resulted in decreased S:G ratios. However, the degree to which the ratio is changed is largely dependent on the extent of gene suppression and possibly growth conditions, as various investigations of transgenic plants with altered COMT expression illustrate. Differing lignin contents are reported for various lines of transgenic tobacco. Early work on anti-sense COMT transgenic plants suggests that reducing COMT expression by approximately 29% leads to reductions in syringyl monomer content and a lower S:G ratio (from 0.97 in wild-type to 0.77 in transformants) (Dwivedi et al, 1994). Yet, further studies suggest that COMT expression in tobacco and poplar must be reduced by more than 50% before decreases in syringyl monomer content are observed, and that between 5-12% residual enzyme activity significantly reduces the S:G ratio (Atanassova et al, 1995; Van Doorsselaere et al, 1995). Although Atanassova et al (1995), successfully over-expressed COMT no analysis of the lignin content or composition was made. Conversely, a later study of reduced COMT activity (by 32%) found the S:G ratio had increased from 1.1 to 1.5 (Sewalt et al, 1997b). Enzymes expressed early in monolignol formation have also been studied. Substantial increases in the S:G ratio of transgenic tobacco have been achieved in PAL down-regulated transgenic tobacco, where the S:G ratio has been increased by approximately 50% (Sewalt et al, 1997a). C4H has also been down-regulated in tobacco, to show a reduction in S:G ratio from approximately 6.8 in wild-type to 0.05 in transgenic lines (99.3%) (Sewalt et al, 1997a). These results are very intriguing as they are highly suggestive of a control mechanism very early in monolignol production. 66 D F R C analysis was also employed in an attempt to identify 5-hydroxyconiferaldehyde, the direct product of coniferyl aldehyde conversion by F 5 H in the lignin of wild-type and transgenic poplar. This intermediate in the monolignol biosynthetic pathway was not identified in the G C / M S chromatogram of the wild-type sample (Figure 13). Qualitative D F R C results indicate that this product is present in C 4 H - F 5 H transgenic lines expressing greater than 85 mole % syringyl lignin. Figure 14 (the chromatogram from Figure 13 focussed at 36-41.5 minutes) shows C 4 H - F 5 H 64 (93.4 mol % S) and the wild-type for comparison (the 5-hydroxyconiferaldehyde peak is highlighted in the rectangle). Since, over-expression of F 5 H in some of the transgenic lines resulted in the accumulation of 5-hydroxyconiferaldehyde, this would suggest that C O M T may be a limiting enzyme in the phenylpropanoid pathway. This data suggests that i f C O M T and possibly S A D were over-expressed, in addition to F 5 H , transgenic poplar may be pushed to express closer to 100 mol % syringyl monomer lignin content. 3.1.2 Determination of Lignin Content A modified version of the Klason lignin method was employed to determine total lignin content (Figure 15). Total wild-type lignin content was found to be approximately 30%, while transgenic lines ranged from 27-32% of dry weight. There does not appear to be an appreciable difference in total lignin values. However, a notable trend appears with insoluble and soluble lignin ratios. Wild-type trees exhibit approximately 25% insoluble lignin, while the transgenic lines range from 25-21%, displaying a general decrease in insoluble lignin as mol % S increases. The difference in soluble lignin values is significant; wild-type has about 4% soluble lignin, while the transgenic lines range from 5-9%. A s the work previously performed by Chiang and Funaoka (1990) would predict, these results clearly show that lignin solubility concomitantly increases with increased mole % syringyl monomers. 67 15 20 25 30 Time (min) 35 40 Figure 13. GC spectrum of DFRC products from wild-type poplar. The region indicated with a rectangle is shown in a close up view in Figure 14. 68 30 5 I—'—'—i—'—i—i—i—•—i—i—i—i i • i i i • i . i . l 36.0 36.5 37.0 37.5 38.0 38.5 39.0 39.5 40.0 40.5 41.0 41.5 T i m e (min) Figure 14. Partial GC/MS spectrum of DFRC products from wild-type and C4H-F5H 64 (93.4 mol % S) poplar focussed at 36 to 41.5 minutes. The 5-hydroxyconiferaldehyde peak is highlighted in the rectangle and was derived from the MS fragmentation pattern. The molecular structure of 5-hydroxyconiferaldehyde is also shown. 69 45 40 35 30 25 20. 10" [ Total Lignin -Y///X Insoluble Lignin Soluble Lignin 0 i ft H 5 2^ 65.6 70.7 79.5 82.5 83.6 86.4 Mol % S 88.8 89.2 93.5 ure 15. Percentage of lignin in the wild-type and transgenic C 4 H - F 5 H lines of poplar. Values represent the average of two replicates while error bars represent a range of 0.6-3.2 for insoluble, 0.1-0.8 for soluble, and 0.07-1.8 for total lignin. 70 In an effort to achieve improved efficacy of delignification, many other researchers have focused on decreasing the total lignin content in their transgenic lines. For example, PAL suppressed transgenic tobacco and those with severely depressed CCR expression have shown Klason lignin contents reduced by approximately 50%. However, these decreases unfortunately are accompanied by significant growth deficits (Sewalt et al, 1997a; Piquemal et al, 1998; Chabannes et al, 2001a). CCoAOMT transgenic plants also have reduced total Klason lignin values with transgenic alfalfa exhibiting a 14.5% reduction, while both transgenic tobacco and poplar display decreases of approximately 50%. Although the significant reduction in total lignin was accompanied by collapsed vessel walls in the tobacco, transgenic poplar appear to be morphologically unaffected (Zhong et al, 1998; Meyermans et al, 2000; Guo et al, 2001). While some research suggests that COMT anti sense transgenic tobacco plants have total lignin values reduced by 11-26% (Dwivedi et al, 1994; Sewalt et al, 1997b; Guo et al, 2001), others have found that total Klason lignin values are unchanged by even substantially reduced expression of COMT in tobacco (Atanassova et al, 1995). In transgenic tobacco double transformed for severely reduced CCR expression and moderately decreased COMT expression, Klason lignin values decreased by approximately 50% (Pincon et al, 2001). Reduced expression of both COMT and CAD-COMT in poplar generates reduced Klason lignin content of 17 and 16%, respectively (Lapierre et al, 1999; Jouanin et al, 2000). CAD expression alone has also been shown to affect total lignin content by as much as 10% (Lapierre et al, 1999). However, field grown transgenic poplar with depressed CAD expression achieved Klason lignin values decreased by only 1-2% (Pilate et al, 2002). The total lignin content in 4CL suppressed Arabidopsis was slightly decreased relative to the wild-type (Lee et al, 1997). The positive effects of 4CL suppression appear to translate well in transgenic poplar, where lignin was reduced by approximately 50% and the cellulose 71 content increased by up to 15%. Furthermore, young transgenic trees experience enhanced growth (Hu etal, 1999). Clearly, transgenic research has demonstrated the potential of engineering trees with tailored properties, which have the potential to significantly affect wood processes and quality. 3.2 Wood Carbohydrate Analysis High pressure or high performance liquid chromatography ( H P L C ) is an excellent separation method providing quantitative and qualitative information. Individual compounds in a mixture w i l l elute at a unique time under the same conditions. The area under the molecule's peak on the chromatogram is indicative of the amount of that substance present in the mixture (Meyer, 1998). H P L C has been used in this study to analyse the Klason lignin filtrate for carbohydrate content of the C 4 H - F 5 H over-expressing poplar. Other researchers have observed secondary effects to the cell wal l caused by the transgenic manipulation of particular phenylpropanoid pathway enzymes (Hu et al, 1999; Chabannes et al, 2001a). Therefore, to determine i f the over-expression of C 4 H - F 5 H affects carbohydrate content an analysis of this major wood component was performed (Figure 16). A l l transgenic lines showed carbohydrate levels comparable to wild-type, ranging from 68-71%. These results suggest that F5H-over expression, resulting in up to 93 mol % S, does not impact carbohydrate metabolism in these transgenic trees. This is however not the case for down-regulated 4 C L poplars, which were shown to significantly reduced total lignin content, as previously discussed. In severely reduced lines, the carbon flux appears to be redirected toward cellulose metabolism, and demonstrates a 15% increase. Interestingly, these chemical changes are accompanied by increased growth (Hu et al, 1999). Transgenic poplars having reduced expression of C O M T also exhibit altered lignin content and composition. Further characterization of these transformants indicated a concurrent 72 70 Arabinose KWN Glucose 60 f- V77A Galactose Xylose \ \ s \ \ ' s s N > s > rN> '"•>•_, ^>: A& r — w F;>H 1 > 1 h / s > : § 1 3 Mannose 1 Rhamnose 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol% S gure 16. Carbohydrate composition of wild-type and C 4 H - F 5 H transgenic poplar lines. Values represent the average of two replicates while error bars represent a range of 0.02-0.08 (Ara), 0.003-0.1 (Gal), 0.5-3.0 (Glu), 0.5-1.9 (Xyl) , 0.001-0.1 (Man), and 0.002-0.06 (Rha). 73 increase in the cellulosic fraction by approximately 6% (Jouanin et al, 2000). Enhanced carbohydrate content has also been observed in transgenic tobacco having reduced expression of C C R and in those double transformed for reduced C C R and C A D , where carbohydrate content may increase by 20% in the C C R under-expressing line, and by approximately 10% in the double transformed line (Chabannes et al, 2001a). As these studies demonstrate, reduced expression of several enzymes in the lignin biosynthetic pathway may impact carbohydrate content. However, alteration to the carbohydrate content and composition has not been observed in this study of poplar over-expressing C 4 H - F 5 H . 3.3 Extractives Analysis D-glucose is the precursor molecule of a wide variety of compounds found in wood. Phosphoenol pyruvic acid, a glucose derivative, may be accepted into the shikimate pathway through which aromatic amino acids are formed, including phenylalanine, tyrosine, and tryptophan. The cinnamic acid pathway accepts phenylalanine and in a series of enzyme catalyzed steps forms various hydroxycinnimate acids. In turn, these acids are the precursors of a vast array of compounds including lignin, carbohydrates, lipids, terpenoids flavonoids, and stelbenes (Sjostrom, 1993; Baucher et al, 1998). A s the metabolism o f all o f these compounds is intimately linked, it is plausible that a transgenic alteration in one branch of this biosynthetic pathway may cause changes in the synthesis of the other compounds. A n analysis of the total extractive content (Figure 17) suggests that over-expression of C 4 H - F 5 H may be associated with slight increases in overall extractives content. The percentage of total extractives in dry wood is 0.35% for wild-type with a range of 0.5 and transgenic lines are 0.45-1.1% with a range of 0.05 - 0.6, however, the degree of error associated with this analysis means that no definite 74 65.5 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 17. Percentage of extractives in dry weight wood of C 4 H - F 5 H transgenic and wild-type poplar. Values represent the average of two replicates while the error bars represent a range of 0.45-1.1. 75 statement may be made regarding this increase. Therefore, this study also included an analysis of four broad extractives groups present in the 2-year-old trees in an attempt to determine i f the transgenic trees over-expressing C 4 H - F 5 H produce different amounts of their major extractive constituents in comparison to the wild-type trees. The results indicate that there is no observable difference in any of the extractive classes studied between all of the tree lines (Figure 18). Based on total extractives, fatty acids comprise between 5.6-7.0%, steroids between 32.7-34.7%, steryl esters between 59.5-61.3%, and triglycerides make up a very small proportion, approximately 0.03% of the total amount of extractives present in the wood. 3.4 Image Analysis To rule out any influence of cell morphological differences for altered delignification efficacy in the transgenic trees, image analysis of 2-year-old stem cross sections was performed. This technique was used to determine fibre lumen area (Figure 19) and demonstrates that the average fibre lumen area for all samples appears to be consistent, 0.001mm 2 (0.02mm 2 standard deviations). The average vessel lumen areas for all samples also fall within a similarly wide standard deviation; wild-type average vessel lumen area is approximately 0.05 mm 2 , and transgenic lines range between 0.03-0.06 m m 2 (Figure 20). The average cell wal l thicknesses are also fairly consistent (approximately 2 /mi) across all samples and standard deviations from 0.1-0.4 /xm (Figure 21). Collectively, these results indicate that over-expressing C 4 H - F 5 H transgenic trees do not have larger or more vessel elements that may have aided pulping efficacy. A s well , F 5 H over-expression does not appear to affect fibre lumen area, number, or cell wall thickness. Furthermore, no morphological anomalies were observed in the gross anatomical structure of the vessel and cell walls, corroborating earlier work with this construct (Meyer et al, 1998; Franke et al, 2000). These results compare well with those of other 76 CO CD > "-(—< O CO v_ -(—> X LU 100 90 80 70 60 50 40 30 20 10 Wk%\\\ Triglycerides V777\ Fatty Acids Steroids R s ^ s l Steryl Esters 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.5 Mole % S Figure 18. Relative proportions of the extractive classes in the wood of C 4 H - F 5 H transgenic and wild-type poplar. 77 0.05 0.04 E 0.03 0.02 CD i _ ^ 0.01 c CD E o.oo 13 I CD -0.01 .g ^ -0.02 0 0 0 0 0 0 0 0 0 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S lire 19. Average fibre lumen areas of C 4 H - F 5 H transgenic and of the wild-type poplar. Values represent the average of -3000 measurements and standard deviations of 0.02 mm 2 . 78 2.0 r-1.8 -CN E 1.6 -E 1.4 -1.2 -1.0 -"ea 0.8 -< 0.6 -c 0.4 -CD E 0.2 -0.0 -_ i -0.2 -CD 00 -0.4 -00 CD -0.6 -> 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 20. Average vessel lumen areas of C 4 H - F 5 H transgenic and wild-type poplar. Values are the average o f - 4 0 measurements and standard deviation of 0.01 mm 2 . 79 E C O C O C D o r— 0 3 CD O 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.5 % Mol Syringyl Lignin Figure 21. Average cell wall thickness of both fibres and vessels in a cross-sections wood from wild-type and transgenic C 4 H - F 5 H poplar. Values represent the average of ~ 760 measurements and standard deviation between 0.1-0.4 fim. 80 transgenic poplar having either reduced C C o A O M T activity or reduced C O M T expression, both of which exhibit normal vessel walls (Jouanin et al, 2000; Meyermans et al, 2000). This is not the case with some investigations of transgenic tobacco such as those down-regulated in C O M T and C C o A O M T activity, which appear to grow well but exhibit larger cell diameters (Ni et al, 1994) and abnormal vessel walls, which appear to have collapsed in both the primary and secondary xylem (Zhong et al, 1998). Furthermore, severe down-regulation of C C R expression in transgenic tobacco causes the same type of vessel abnormality observed in the transgenic C C o A O M T tobacco lines (Piquemal et al, 1998; Chabannes et al, 2001a). 3.5 Wood Fibre Quality Analysis Fibre Quality Analysis was performed on two-year-old stem wood of the C 4 H - F 5 H transformed poplar. Fibre coarseness, defined as the mass of the fibres per unit of length, does not show a trend with regard to mole % syringyl units in the wood of poplar over expressing C 4 H - F 5 H (Figure 22). Wild-type trees showed an average coarseness value of 0.31 mg/m, while transformed lines range from 0.30-0.47 mg/m. This result further supports the image analysis that was performed and suggests that the up regulation of C 4 H - F 5 H does not influence the size of wood fibres. 3.6 Wood Specific Gravity Determination The specific gravity was determined for two-year old stem sections of the C 4 H - F 5 H transformed poplar (Figure 23). There does not appear to be a trend regarding wood density and mole % syringyl monomers in poplar over-expressing C 4 H - F 5 H . The wild-type trees show an average density of 451 kg/m 3 , while transformed lines range from 398-464 kg/m 3 , all samples measured are within similarly wide ranges. However, lines exhibiting greater than 83.6 mole % syringyl lignin seem to display a slightly lower specific, but as sample ranges are overlapping, there does not appear to be a true difference. 81 65.5 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 22. Average wood fibre coarsenesses of the wild-type and C 4 H - F 5 H transgenic poplar. Values represent the average of duplicate experiments where approximately 10,000 measurements were made. The error bars represent a range o f 0.0-0.01 mg/m. 82 600 550 C O CD 500 4 ^ 450 00 c C D Q " O o o 400 350 300 0 0 0 0 -r 0 0 • a 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S ure 23. Average wood density of wild-type and C 4 H - F 5 H transgenic poplar. Values represent the average of six measurements made from three 2-year-old wood cross-sections while error bars represent a range of 17.9-72.4 kg/m 3 . 83 Chapter 4 Pulp Analysis Kraft pulping was performed on the transgenic poplar over-expressing C 4 H - F 5 H in order to determine i f delignification efficacy is altered by this transgenic modification. Furthermore, subsequent analyses of the resulting pulp were undertaken to evaluate the quality of the pulp derived from the transgenic lines in comparison to the wild-type poplar. The kraft pulping process has been chosen for this investigation, as it is the most widely used method for pulp production; kraft pulp comprises 56% of the pulp manufacture globally (Stenius, 2000). Effective pulping occurs when maximum delignification is achieved with minimal degradation of cellulose (Stenius, 2000). Sodium hydroxide (NaOH) and sodium sulfide (Na 2S) make up the active alkali ( A A ) portion of the cooking liquor as follows: N a O H -> N a + + OH" Na 2 S -» 2 N a + + S 2" S2" + H 2 0 SH" + OH". However, because the sulfide ion is completely hydrolysed to hydrosulfide and hydroxide, as shown, the term effective alkali (EA) (NaOH + lA Na 2 S) more accurately represents the alkali concentration of the white liquor (Adams et al, 1983). The hydroxyl ions (OH") and hydrogen sulfide ions (SH") present in the liquor react with the lignin to reduce it to small soluble molecules (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000), whereby approximately 90% of the lignin is removed during the pulping reaction. However, the carbohydrate portion of the wood is also subjected to partial solubilization during sulfate pulping through reactions with the hydroxyl ions (Mimms et al, 1989; Stenius, 2000). Overall, as much as 15% of the cellulose and 60% of the hemicelluloses may be depolymerized during the kraft pulp process (Stenius, 2000). 84 Kraft delignification occurs generally in 3 phases. During the initial or extraction phase the cooking temperature is low and very little delignification occurs, where between 15-20% of the lignin is removed, however, a large portion (40%) of the hemicellulose is removed (Stenius, 2000). A s well , the majority of extractives enter solution in this initial phase of the cook (Adams et al, 1983). The relative reaction rate of delignification is subsequently very low during this stage and therefore has little effect on lignin removal (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). This portion of the kraft cook generally facilitates liquor penetration into the wood chips and diffusion of the liquor throughout the chips in the reactor (Mimms et al, 1989). Bulk delignification occurs in the second phase and results in removal of a further 70% of the total lignin (Mimms et al, 1989; Stenius, 2000). A s this stage o f the cook occurs at high temperatures (140-180 °C) the relative reaction rate significantly increases, and therefore greatly affects the total delignification process (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). During the final stage (residual delignification phase) very little lignin is solubilized and the lignin remaining may largely represent condensation products formed during the second phase. Consequently, as few gains are made with lignin removal, the alkaline hydrolysis of cellulose becomes a significant factor during the final stage of cooking, because as the cellulose polymer is cleaved, viscosity decreases and the strength of the pulp is reduced (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). There are a number of factors that may be modified to control the cooking conditions, including: alkali charge, liquor-to-wood ratio, time and temperature (Mimms et al, 1989). The alkali charge and liquor-to-wood ratio are set before the cook begins, while the time and temperature are factors that may be modified during the cook. Collectively, time and temperature are known as the H-factor, a numerical value, which represents the area under the relative reaction rate curve (Adams et al, 1983; Mimms et al, 1989). Usually, the H-factor is adjusted to achieve target kappa numbers (residual lignin), as small increases in temperature can 85 substantially increase the relative reaction rate, as can increases in cooking time at high temperatures (Mimms et al, 1989). Alpha and /3-aryl ether linkages, in both the free and etherified phenolic structures, represent the majority o f the bond types targeted during kraft l ignin depolymerization (Stenius, 2000). Although carbon-carbon bonds may also be cleaved during pulping, they are more resistant, particularly at temperatures below 140 °C (Adams et al, 1983). Hydrosulfide ions are stronger nucleophiles than the hydroxyl ions present in the kraft pulping liquor, thereby accelerating the delignification process without detriment to pulp yields (Adams'e^ al, 1983). Figure 24 represents the proposed mechanisms for lignin depolymerization during kraft pulping. As shown in mechanism A , the hydrogen sulphide ions facilitate the rapid cleavage of the phenolic /3-aryl ether bond (I) by attacking the quinone methide (II) breaking the /3-0-4 bond to form a mercaptide structure (III) and liberating an ether group (Adams et al, 1983; Stenius, 2000). A 1,4-dithiane structure (IV) ultimately decomposes to yield elemental sulphur and a styrene (V) (Stenius, 2000). Hydrogen sulphide ions do not participate in the cleavage of non-phenolic /3-aryl ether bonds, which subsequently occurs more slowly (Figure 20B) through an oxirane intermediate (VI), also yielding a styrene (VII) (Stenius, 2000). Demethylation also occurs most often through the activity of hydrogen sulfide ions as shown (Figure 20C) forming methyl mercaptan (VIII) (Stenius, 2000). It is possible for condensation reactions to occur between the liberated phenolics, which consequently causes the formation of highly resistant dimmers that remain associated with the pulp. Multiple conjugations produce chromophores, which are responsible for the browning of pulps (Stenius, 2000). Two methods, Klason lignin and kappa number, have been used to determine the amount of residual lignin in the kraft pulp samples. The kappa number represents the amount of permanganate that is consumed in its reaction with lignin in the pulp sample. The kappa number when multiplyed by a factor of 0.147 quite accurately corresponds to the percentage of Klason lignin (Mimms et al, 1989). 86 Figure 24. Kraft pulp mechanism adapted from Adams et al. (1983) page 41 and Stenius (2000) page 67. Mechanism A represents the rapid cleavage o f phenolyic P-aryl ether bonds by hydrogen sulphide ions. Non-phenolyic P-aryl ether bonds are cleaved by hydroxide ions and is a slower reaction. Mechanism C represents the formation of methyl mercaptan by hydrogen sulfide ions. 87 A reaction called peeling causes the degradation of carbohydrate polymers one monosaccharide at a time. A hemiacetal end group (reducing end) of the polysaccharide may be isomerised causing the removal of a -3-alkoxy, thereby solublizing a carbohydrate monomer. This process is known as primary peeling (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). This same mechanism is facilitated at higher temperatures (secondary peeling) when alkaline hydrolysis randomly cleaves glycosidic bonds to expose more reducing end groups (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). On average, peeling reactions remove between 65-70 glucose monomers from cellulose before the reaction stops (Adams et al, 1983; Stenius, 2000). The stopping reaction occurs when the hydroxyl group is cleaved from the polysaccharide without subsequent isomerization. This reaction results in the formation of a metasaccharinic acid endgroup which is alkali-stable and the cessation of the peeling reaction (Adams et al, 1983; Mimms et al, 1989; Stenius, 2000). Physical stopping may also occur, more often in cellulose when the polysaccharide simply becomes inaccessible (Adams et al, 1983). The extent of alkaline hydrolysis in a pulp sample may be evaluated by measuring the pulp viscosity (Mimms et al, 1989; Stenius, 2000). As discussed previously, hemicellulose is highly soluble during kraft pulping. In hardwoods the major hemicellulose is glucuronoxylan, hence a great deal of xylan is liberated in this process. Studies have shown that kraft pulp fibres retain xylan, likely because xylan solubility decreases as alkali is consumed during the cook and when their uronic acid groups are removed (Adams et al, 1983; Stenius, 2000). The uronic acid group on xylan may also be converted to hexenuronic acid groups (HexA and HexU) during pulping. When present, these acids consume bleaching chemicals and react with permanganate, which may contribute to falsely high kappa numbers (Stenius, 2000). Pulp bleaching has also been performed to determine how amendable the residual lignin in the pulp derived from C 4 H - F 5 H over-expressing poplar is to brightening. Brightness refers 88 to the reflectance o f visible light at 457 nm by pulp when compared to the standard (absolute reflectance) of magnesium oxide (Stenius, 2000). There are several methods for determining pulp brightness, which are commonly reported as ISO (International Standards Organization) brightness units (Adams et al, 1983). Chromophores cause pulp browning and are present in the residual lignin; they therefore are the targets of the bleaching sequence. Bleaching may involve either lignin extraction or chromophore conversion (Stenius, 2000). In this study, a bleaching series of chlorine dioxide, sodium hydroxide, followed by a second chlorine dioxide stage (DED) was employed. Chlorine dioxide is a radical scavenger, which brightens the pulp by oxidizing and solubilizing the lignin, while the sodium hydroxide removes the lignin and hydrolyses the chlorinated lignin (Stenius, 2000). Mult iple stages are required to achieve sufficient brightness without damaging pulp quality (Adams et al, 1983; Stenius, 2000). 4.1 Kraft Pulping Various kraft pulping cook times were used to determine the optimum cooking cycle for sufficient delignification of the transgenic lines. A s expected from the investigation of the transgenic wood samples, residual lignin values for the pulps show appreciable decreases with increasing mol % S in the transgenic lines, under all pulping conditions (Figure 25). With each cook employed, the transgenic lines achieved much greater delignification than was observed for the wild-type. A kappa number of 36 was determined for the wild-type under the least severe cook (H-factor 856) having a 45 minute cooking time, while transgenic line F 5 H 6 4 with 93 mol % S has a kappa of 13 at the same conditions. A s shown, a similar trend was observed for the other two cooking conditions. The greatest residual lignin reductions reported in the literature are those achieved for the pulps derived from C A D deficient transgenic trees, for which lignin reductions by up to 5.8 89 40 J — i — • i • i i i . i i i . i i i i i" 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.5 % Mol Syringyl Lignin Figure 25. Residual lignin values (kappa number) generated from pulp of wild-type and C 4 H -F 5 H transgenic poplar cooked at different H-factors. Only the wild-type trees were pulped at higher H-factors (1784 and 2110). Values represent the average of three replicates while the error bars represent a range of 0.01-3.0 kappa units. 90 kappa units have been reported (Baucher et al, 1996; Lapierre et al, 1999; Lapierre et al, 2000; Pilate et al, 2002). Lapierre et al. (2000) reported significant reductions in kappa number (25.5) for a 1-year-old, down-regulated C A D line, however, further analysis of this transgenic poplar line at 2-years-old indicates that this significant reduction is not maintained in the older trees where kappa is reduced by only 3.1 units. In the C 4 H - F 5 H transgenic lines the reduction in kappa value is maintained from the 1-year-old trees to the 2-year-old trees (Figure 26). Transgenic line C 4 H - F 5 H 64 (93.4 mol % S) exhibits a 22.7 reduction in kappa units from the wild-type at a low H-factor of 796. In other cases, residual lignin values, such as those of down-regulated C O M T transgenic poplar, have been observed to increase by up to 18 kappa units (Jouanin et al, 2000; Lapierre et al, 2000; Pilate et al, 2002). In light of these previous findings, the results attained for the C 4 H - F 5 H poplar have set a new standard. Pulp derived from trees having this genetic modification have residual lignin content reduced by up to 23 kappa units. Furthermore, to attain a kappa value (14) in the wild-type poplar that could be compared to the C 4 H - F 5 H transgenic lines, a much greater severity (H-factor of 2110) with a 2.5 hour cook was required. The difference in the relative ease of delignification for the transgenic poplar equates to significant savings in the necessary reaction time and/or the chemical load requirement to attain a target kappa. Additionally, trees with >83.6 mol % S (F5H 37, 65, 26, and 64) showed smaller decreases in residual lignin with increases in pulping severity. A s arylglycerol-/3-aryl ether bonds are formed more often between syringyl moieties, and as this bond type is the key target of the kraft delignification reaction (Chiang and Funaoka, 1990), it is highly plausible that near maximum delignification has already occurred at much lower severities in lines with high syringyl lignin content. These data suggest that C 4 H - F 5 H over-expressing lines with >83.6 mol % S could achieve the same amount of delignification as the native poplar with a cooking severity decreased by more than half. 91 ;ure 26. Residual lignin values (kappa number) generated from pulp of 2-year-old wild-type and C 4 H - F 5 H transgenic poplar cooked at different H-factors. Values represent the average of two replicates while the error bars represent a range of 0.03-2.0 kappa units. 92 Other research has shown transgenic lines with reduced total lignin containing an increased proportion of G units (Jouanin et al, 2000), which are known to form condensed bonding types (Chiang and Funaoka, 1990). Therefore, as expected, the resulting pulp exhibited a greater proportion of highly condensed lignin, which was very resistant to delignification (Jouanin et al, 2000). These results combined with those for the C 4 H - F 5 H transgenic poplar further supports the research conducted by Chiang et al. (1988), where they suggest that ease of delignification is more greatly attributed to the reactivity of the lignin, not to the amount of lignin. 4.2 Pulp Yield Pulp yield appears to be lower in the transgenic lines (Figure 27), however this decrease directly corresponds to their decreased residual lignin (Figure 25). Considering that a kappa number of 6.67 is equivalent to 1% lignin; using the least severe H-factor (856) as an example, there is a difference of approximately 3.45 % lignin removed which accounts for the 3.52 % difference in yield between wild-type and C 4 H - F 5 H 64 (93 mol % S). Furthermore, wild-type kappa values determined for pulps derived from the three conditions used for all the pulp samples were too high to compare to the pulps of the transgenic lines, therefore wild-type samples were cooked to two additional H-factors (2110 and 1784). The pulp yields and kappa values measured for wild-type cooked at the greatest severity are similar to those of the transgenic lines (Figure 27 and 25, respectively). The residual lignin and pulp yield trends observed in the one-year old trees are maintained in the 2-year old trees. Again, comparing the least severe cook (H-factor = 796), wild-type has a residual kappa of approximately 37, while 93 mol % S (C4H-F5H 64) has a kappa number of about 14 (Figure 26). Furthermore, the 2-year-old trees exhibit corollary values for yield, with the reduced yield in the transgenic lines being 93 50 49 48 Q 47 P 46 CD Q_ 45 Q_ 44 \ \ \ h >\ \ > N: 2110 H-factor 1^^1784 H-factor 1547 H-factor F^11096 H-factor F = l 856 H-factor V \ \ 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 27. Pulp yields of one-year-old wild-type and transgenic C 4 H - F 5 H poplar Pulps are derived from a single wood sample of between 10-20 pooled trees. 94 approximately equivalent to the lower residual lignin values (Figure 28). Other researchers have been able to show increased pulp yields for C O M T down-regulated lines that are likely reflective of the higher residual lignin content (Lapierre et al, 1999; Jouanin et al, 2000). However, other studies, which report decreased kappa numbers, have shown that pulp yields may decrease (Baucher et al, 1996), remain comparable to the control (Lapierre et al, 1999; Lapierre et al, 2000), or slightly increase (Pilate et al, 2002). 4.3 Determination of Pulp Lignin Content The klason lignin method was used to determine lignin content of the pulps. In contrast to the total lignin values observed for the transgenic wood samples, there is a general trend with regard to pulp lignin. A s the mole % syringyl units increased there is a corresponding decrease in total lignin (Figure 29). Only the results for samples pulped at an H-factor of 1096 are shown, but, the same trend is observed for the other severities that were employed (Appendix A ) . The wild-type has approximately 3.7% of dry weight insoluble lignin, and the transgenic samples range from 3.7-1.3%. This substantially decreased insoluble lignin content is responsible for the corresponding reduction in total residual lignin of the pulp, as the soluble lignin content is comparable across all samples, including wild-type (approximately 1%). The plateau effect that is observed for the soluble lignin content is l ikely because the kraft pulp process has removed almost all o f the accessible soluble lignin from the tree samples having greater than 83.6 mole % syringyl monomers. These results support those determined by the kappa analysis, whereby residual lignin values change little with increased pulping severity in the transgenic lines expressing greater than 83.6 mole % syringyl units. The results obtained for acid soluble lignin compare well to those observed by other researchers, however, the C 4 H - F 5 H transgenic poplar have shown a greater reduction in the insoluble lignin portion (Tamura et al, 2001). 95 60 ]1483 H-factor 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 28. Pulp yields of two-year-old wild-type and transgenic C 4 H - F 5 H poplar cooked at different H-factors. 96 4 CD 'CD „ Q 2 1 0 Total Lignin Y///X Insoluble Lignin Soluble Lignin $5 ft 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 29. L ignin content of pulp generated from the wild-type and C 4 H - F 5 H transgenic poplar cooked at an H-factor of 1096. Values represent the average of two replicates while the error bars represent a range of 0.1-1.2 (insoluble), 0.03-0.2 (soluble), and 0.1-1.3 (total). 97 4.4 Determination of Pulp Lignin Monomer Composition Thioacidolysis was performed on one pulp sample series (H-factor 1096). The pulp S:G ratio increases with samples having a correspondingly higher S:G ratio in the wood (Figure 30). For example, the wild-type wood has an S:G ratio o f 1.9 and a pulp S:G ratio of 0.8, while the transgenic line with a wood S:G ratio of 14.2 has a corresponding pulp S:G ratio of 3.4. A s Franke et al. (2000) have shown, the entire course of cell wal l lignification is impacted by the C 4 H - F 5 H construct. Therefore, these results are as expected; transgenic lines should have pulps with residual lignin monomers (those contained within the cell wall) that reflect the increased S:G ratio of the starting material. Furthermore, condensation reactions of the syringyl monomer B-B and B-l bonding types w i l l likely occur between solvated lignin monomers (Lapierre et al., 1995), which are reflective of the proportionality of syringyl monomers present in the starting material. 4.5 Determination of Pulp Carbohydrate Composition Pulp carbohydrate content was determined by H P L C analysis of the klason lignin filtrate. For clarity, and as they are representative of the other pulps (Appendix B) , only the results of the 1096 H-factor are shown (Figure 31). Generally, the pulps from transgenic lines have slightly increased carbohydrates. A l l of the samples have approximately 20% xylose, while wild-type has a glucose content o f - 7 2 % , and transgenic.lines range from approximately 73-75%. This increased carbohydrate content in the transgenic lines is to be expected, as they have reduced residual lignin values. Furthermore, the results obtained for kraft pulp carbohydrate content are comparable to the values determined by others (Tamura et al, 2001). The higher glucose value observed in the pulp derived from transgenic lines indicates that a greater proportion o f this pulp is composed of cellulose, and further suggests that these samples may be made into higher quality paper than wild-type, a supposition supported by the viscosity 98 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.5 Mol % S ure 30. Lignin monomer ratio in wood and pulp produced from wild-type and transgenic C 4 H - F 5 H poplar lines pulped at an H-factor of 1096. Values represent the average of two replicates while the error bars represent the range of 0.004-0.2 (wood) and 0.02-0.5 (pulp). 99 90 j I I Glucose E^^l Xylose 85 -80 - T _ T T T T T T £ 75 -CD 65.6 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mol % S Figure 31. Carbohydrate contents of pulp produced from wild-type and C 4 H - F 5 H transgenic poplar pulped at an H-factor of 1096.' Values represent the average of two replicates while error bars represent the range 0.06-5.7 (Glu) and 0.06-1.0 (Xyl) . 100 data (Section 4.6). The greater proportion of cellulose, and lower lignin content may in turn contribute to enhanced bleaching (Section 4.8). 4.6 Determination of Pulp Viscosity Pulp viscosity, which may be used to estimate the degree of polymerization (DP) of the pulp carbohydrates, was determined for all pulp samples. Generally, greater cooking severity or increased physical accessibility of the liquor to the polysaccharides w i l l cause a decrease in cellulose D P (Mimms et al, 1989; Stenius, 2000). For the most part, this trend is observed in the samples tested in this study (Table 4), as the viscosity values are lower at higher H-factors. There is also an observable trend with regard to increased viscosity with increased mole % syringyl units; for example, under the least severe cook (H-factor 856) wild-type has a viscosity of 45.1 cP and C 4 H - F 5 H 64 has a value of 66.7 cP. The viscosity values of this study appear to surpass studies reporting reduced kappa values and cellulose D P values that are slightly improved or comparable to wild-type (Baucher et al, 1996; Lapierre et al, 1999; Jouanin et al, 2000; Lapierre et al, 2000). Studies have shown that during papermaking and use, the tensile load is borne by cellulose within the cell wall (Gurnagul et al, 1992). A decrease in cellulose DP , which occurs during the final stage o f kraft pulping and during bleaching, appears to be associated with lower fibre strength and subsequently weaker pulp and paper (Gurnagul et al., 1992). It follows, that pulp with greater cellulose D P (as indicated by viscosity testing) w i l l have improved strength characteristics over pulp with lower cellulose D P , as long as all other morphological characteristics are similar. Therefore, the results of the current study suggest that pulp derived from the transgenic lines w i l l l ikely have equal or improved strength characteristics over wild-type, as the cellulose chain length is greater. 101 Table 4. Pulp viscosities (cP) of wild-type and C 4 H - F 5 H transformed poplar processed at different H-factors. Tree Line S:G Ratio H-Factor 2110 1784 1547 1096 856 Wild-type 1.90 36.6 43.42 45.98 42.85 45.07 C 4 H - F 5 H 85 2.41 49.80 56.10 44.89 C 4 H - F 5 H 2 1 3.87 52.16 50.86 N D C 4 H - F 5 H 82 4.72 51.44 62.86 60.41 C 4 H - F 5 H 41 5.09 47.19 49.25 52.37 C 4 H - F 5 H 37 7.95 55.04 59.61 66.48 C 4 H - F 5 H 65 8.21 47.60 59.17 64.40 C 4 H - F 5 H 26 8.26 60.08 52.88 57.59 C 4 H - F 5 H 64 14.17 56.85 74.90 66.74 Viscosity values are the average of three replicates with a range o f 0.01-7.1 cP. 102 4.7 Pulp Fibre Quality Analysis Fibre quality analysis was performed on the pulp samples to determine fibre coarseness and fibre length. A comparison o f residual lignin values and pulp coarseness indicates that, in the majority of samples, lower kappa values correspond to lower coarseness measurements (Figure 32). However, there does not appear to be a direct relationship between pulp coarseness and mole % syringyl monomers. Some pulps derived from lower mole % syringyl units are finer than those with higher mole % syringyl units. For example, the sample with 89.1 mol % S (C4H-F5H 65) has coarseness values ranging from 0.069 - 0.070 mg/m while 93.5 mol % S (C4H-F5H 64) has coarsenesses of 0.097 - 0.108 mg/m. Although, compared to the wild-type, all the transgenic lines demonstrate lower coarseness values for a given H-factor (all sample coarseness values are provided in Appendix C). Fibre lengths have also been determined for all of the pulp samples. Generally, as Figure 33 indicates, fibres are shorter for pulps produced under more severe cooking conditions. Furthermore, fibre lengths appear to decrease with increased mole % syringyl units. However, when the wild-type sample (65.6 M o l % S) is subjected to pulping conditions (H-factor 1784 and 2110), which reduce its kappa number to a value comparable to those of the higher syringyl content samples, the fibres are notably shorter and similar in length to those observed in the transgenic lines. Fibre coarseness and lengths are important pulp attributes affecting paper sheet strength. Increasing the surface area contact between individual fibres in the sheet increases the sheet's strength and stretch characteristics, as it is fibre cohesiveness that fails during paper formation, not fibre breakage (Seth, 1995). Longer pulp fibres produce paper with greater tensile strength, while coarse fibres decrease paper strength; these characteristics have a significant impact on the throughput of the papermaking machine (Seth, 1995). It has been noted that fibre coarseness affects paper strength much the way short fibres do (Clark, 1962). It may be surmised that although the 103 0.45 0.40 0.35 ^ 0.30 CO 0.25 co CD g 020 CO CO 0.15 O O 0.10 0.05 • 65.6 O 83.6 V 89.1 O 93.4 0 L 10 15 20 25 30 Kappa 35 40 [lire 32. Coarseness versus kappa number of the pulp derived from wild-type and a selection of C 4 H - F 5 H transgenic poplar cooked at different H-factors. Coarseness measurements are the average of three replicates where 15,000-20,000 individual fibres were measured per pulp sample, and error bars represent the range of 0.0004-0.07 mg/m. 104 E E 0.50 0.45 [• 0.40 r -C D 0.35 h C D -Q 0.30 r-0.25 Mi r [ > : N >->->: >->: >->->: >->: >->->: >->: >->->->: >->-> >-2110 H-factor Y77A 1784 H-factor 1547 H-factor 1096 H-factor 856 H-factor W~ T m >->-_ Kb WT 70.7 79.5 82.5 83.6 88.8 89.1 89.2 93.4 Mole % Syringyl Units Figure 33. Fibre lengths of pulp derived from wild-type and C 4 H - F 5 H transgenic poplar cooked at different H-factors. Length measurements are the average of three replicates where between 15,000-20,000 individual fibres were measured per pulp sample, and error bars represent a range of 0-0.05 mm. 105 transgenic C 4 H - F 5 H lines have comparable lengths to those o f the wild-type trees, at a given residual kappa, they are finer and have a greater glucose content and higher degree of polymerization. Taken together, the pulp characteristics exhibited by the transgenic lines w i l l likely result in the production of higher quality paper relative to the wild-type. 4.8 Pulp Bleaching Analysis Bleaching was performed on pulp derived from the 2-year-old wild-type and C 4 H - F 5 H transgenic poplar samples using a standard D E D sequence (chlorine dioxide; sodium hydroxide; chlorine dioxide). The results from the bleaching analysis are provided in Table 5. A l l o f the transgenic lines, for both pulp conditions analyzed, achieved higher initial and final ISO brightness values than were observed for the wild-type. For example, at an H-factor of 796 the wild-type has a ISO brightness of 49.5, while the transgenic line C 4 H - F 5 H 64 has a brightness value of 81.1 - a difference of greater than 30 ISO brightness units. A t a more severe digestion (1483 H-factor) a brightness difference of approximately 20 ISO brightness units is evident. Previously, researchers have reported improved bleached pulp brightness of less than 1.0 ISO for transgenic poplar lines (Lapierre et al, 2000; Pilate et al, 2002). In light o f these findings, the results obtained for the C 4 H - F 5 H transgenic poplar have once more set a new precedent. Furthermore, these results strongly suggest that target brightness may be achieved for transgenic lines using significantly lower chemical loading compared to wild-type, and subsequently, fewer toxic chemicals being released to the environment. 4.9 Black Liquor Lignin Analysis Lignin degradation products may be precipitated from black liquor with acidification, purified, and analyzed by gel permeation chromatography (Gellerstedt and Lindfors, 1984; Sjostrom, 1993; Stenius, 2000). Lignin was isolated from the kraft pulping liquor of wild-type 106 Table 5. Initial and final ISO brightness values of wild-type processed at" different H-factors and C 4 H - F 5 H transformed poplar Tree Line S:G Ratio H-factor 796 H-factor 1483 Initial '• Brightness D E D Brightness Initial Brightness D E D Brightness W T 1.90 38.5 49.5 43.4 69.1 C 4 H - F 5 H 85 2.41 39.6 50.1 43.5 69.3 C 4 H - F 5 H 2 1 3.87 43.1 64.1 46.9 77.8 C 4 H - F 5 H 82 4.72 • 44.3 62.4 45.7 76.3 C 4 H - F 5 H 41 5.09 41.6 63.5 46.0 79.1 C 4 H - F 5 H 37 7.95 45.9 82.2 47.1 91.0 C 4 H - F 5 H 65 8.22 42.2 77.1 44.6 87.1 C 4 H - F 5 H 26 8.26 46.2 73.2 46.9 87.1 C 4 H - F 5 H 64 14.17 44.0 81.1 45.5 89.2 107 and transgenic C 4 H - F 5 H poplar and analyzed as described. This analysis was performed in order to elucidate any molecular size differences between samples. A s Figure 34 illustrates, no difference in molecular sized distribution is observed after size exclusion chromatography of kraft lignin solubilized from pulp produced at an H-factor o f 1096. The same results were determined for the other pulp conditions used (Appendix D). Tamura et al. (2001) also report that transgenic peroxidase suppressed lines compare to control lines, in that there is no observable difference in kraft liquor lignin degree of polymerization. The results of the current study suggest that although transgenic lines contain greater proportions o f syringyl monomers in the lignin, relative to wild-type, the solubilized lignin is of a similar size. 108 0 10 20 30 40 50 Time (min.) Figure 34. Elution times (relative molecular weight) of black liquor lignin liberated into the pulping liquor at an H-factor of 1096 for wild-type and transgenic trees. 109 Chapter 5 Conclusions Previous work performed with transgenic poplar over-expressing C 4 H - F 5 H strongly suggested that this construct was capable of changing the lignin composition such that delignification efficacy would be improved (Meyer et al, 1998; Franke et al, 2000). The results of the present investigation have confirmed this supposition. This has been possible by showing that the C 4 H - F 5 H phenotype is maintained over a 2-year growing period and that F 5 H is a major regulatory enzyme for the formation of syringyl monomers, as evidenced by the S :G ratios that have been determined using thioacidolysis and D F R C analyses. Chemical and morphological analysis of the transgenic wood indicates that plieotrophic effects do not result from over-expression of C 4 H - F 5 H . The total amounts of lignin, carbohydrate, and extractives do not appear to be altered in the transgenic lines when compared to wild-type. Furthermore, image analysis suggests that transgenic lines do not exhibit larger or more vessel elements than wild-type, which could potentially aid in deligninification. Nor is there an observable trend regarding the relationship between cell wall thickness, number of fibres, or fibre diameter and mole % syringyl units. Thus, microscopic observations indicate that the transgenic trees share structural similarity with the wild-type. However, a significant trend emerges with regard to lignin composition, the lignin monomer analysis shows that the syringyl:guaiacyl ratio may be increased from 1.9 in the wild-type to as high as 14.2 in one transgenic line. Klason lignin analysis indicates that the percentage of acid soluble lignin increases while acid insoluble lignin decreases as the S:G ratio increases such that total lignin remains the same. The benefit o f this substantial increase in mole % syringyl monomers in the transgenic lines is evidenced by the pulping analysis. Pulp kappa values for the transgenic lines were reduced by 23 units relative to the wild-type. The pulp yields for transgenic lines were comparable when pulped to the same residual lignin content. These results were achieved for 110 both the 1-year and 2-year old wood samples, indicating that the C 4 H - F 5 H construct is stable over a 2-year growing period. Transgenic lines produced pulp with higher viscosity values, suggesting a greater degree of cellulose polymerization, and they also exhibited lower coarseness fibres, but demonstrated similar lengths when pulped to the same residual kappa. D E D bleaching experiments have shown that the pulps derived from C 4 H - F 5 H transgenic lines are more amenable to the bleaching process as ISO brightness values were increased by 30 units compared to wild-type. Collectively, these results illustrate that significantly milder conditions may be used to achieve fibre for paper that is of an equal or greater quality than the wild-type trees. Furthermore, these results definitively confirm that trees may be transgenically engineered for specific end uses, such as the pulp and paper industry. A s such, this work has illustrated the potential to produce transgenic trees having qualities that significantly reduce the time and energy required to produce paper. Finally, not only does this investigation suggest a practical means of reducing industrial processing costs, but the environmental impact of processing the wood derived from C 4 H - F 5 H transgenic trees may be significantly reduced. Ill Chapter 6 Future Work The success that has been realized thus far for C 4 H - F 5 H transgenic poplar w i l l provide many other opportunities for study of this construct. The current study has shown delignification efficiency can be substantially improved in these transgenic lines. However, an in depth analysis of the resulting paper strength characteristics has yet to be evaluated to definitively show its value for papermaking. Additionally, an evaluation of lignin-carbohydrate interactions in both the wood and the resulting pulp would aid the understanding of the deligninification process for these transgenic trees. It is quite l ikely that the lignin polymeric structure has been changed as a result of increased syringyl monomer content in the lignin of C 4 H - F 5 H over-expressing poplar. Therefore, preparations for 2 D - N M R analysis of ball-milled lignin derived from C 4 H - F 5 H transgenic poplar have been initiated and these results w i l l provide information regarding the structure of lignin polymer, thereby identifying the effect of this over-expression at a molecular level. Innovative research in the manufacture of carbon fibres has shown the capability o f using kraft lignin as a precursor material (Kadla et al, 2002). In particular, hardwood kraft lignin could be formed into fibres while softwood lignin could not, and Kadla et al. (2002) have suggested that highly condensed lignin resists softening and cannot be extruded. This research represents another potential end use for C 4 H - F 5 H over-expressing transgenic poplar. A very important investigation that w i l l determine the tree's suitability for any industrial use w i l l be an evaluation of their performance in long-term field trials. It w i l l be essential to determine i f the construct is stable once the tree reaches maturity, and i f the tree is capable of growing well under the variable conditions experienced outside of the greenhouse. Furthermore, the societal repercussions of transgenic research demands that ecological impact of the transgenic trees after release in the field is thoroughly evaluated (Mullen and Bertrant, 1998; Merkle and Dean, 2000). 112 The preliminary results of the D F R C analysis suggest that syringyl monomer content could approach 100% i f a poplar double transformant over-expressing F 5 H - C O M T could be produced, with F 5 H expression levels similar or higher than those observed for the C 4 H - F 5 H 64 line. A n extension of this theory would be to apply this construct to a gymnosperm, as conifer lignin rarely contains syringyl monomers a multi-gene construct containing F 5 H , C O M T , and S A D may be used to produce a coniferous tree that can produce substantial amounts of syringyl monomers. The repercussions of this type of transformation could significantly impact pulp and paper production of the future by improving the delignification efficacy of more industrially utilized tree species. 113 Bibliography Adams, T., B . Cowan, et al. (1983). Pulp and Paper Manufacture V o l . 5 Alkaline Pulping. Ste-Anne-de-Bellevue, M c G r a w - H i l l . Anon (2001). World population prospects the 2000 Revision, Population Division Department of Economic and Social Affairs United Nations. Sept 2002. U R L : http://www.un.org/esa/population/publications/wpp2000/highlights.pdf Atanassova, R., N . Favet, et al. (1995). "Altered lignin composition in transgenic tobacco expressing O-methyltransferase sequences in sense and antisense orientation." The Plant Journal 8: 465-477. Balatinecz, J. and D . Kretschmann (2001). Properties and utilization of poplar wood. Poplar Culture in North America. D . Dickmann, J. Isebrands, J. Eckenwalder and J. Richardson. Ottawa, National Research Council o f Canada: 277-291. Bate, N . , J. Orr, et al. (1994). "Quantitative relationship between phenylalanine ammonia-lyase levels and phenylpropanoid accumulation in transgenic tobacco identifies a rate-determining setp in natural product synthesis." Proceedings o f the National Academy of Sciences, U S A 91: 7608-7612. Baucher, M . , M . Bernard-Vailhe, et al. (1999). "Down-regulation of cinnamyl alcohol dehydrogenase in transgenic alfalfa (Medicago sativa L. ) and the effect on lignin composition and digestibility." Plant Molecular Biology 39: 437-447. Baucher, M . , B . Chabbert, et al. (1996). "Red xylem and higher lignin extractability by down-regulating a cinnamyl alcohol dehydrogenase in poplar." Plant Physiology 112: 1479-1490. Baucher, M . , B . Monties, et al. (1998). "Biosynthesis and Genetic Engineering of Lignin ." Critical Reviews in Plant Sciences 17: 125-197. Boudet, A . (1998). " A new view of lignification." Trends in Plant Science 3: 67-71. Boudet, A . (2000). "Lignins and lignification: selected issues." Plant Physiology and Biochemistry 38: 81-96. Brown, C. (2000). The global outlook for future wood supply from forest plantations. Rome, Food and Agriculture Organization of the United Nations: 145. Bugos, R., V . Chiang, et al. (1991). " c D N A cloning, sequence analysis and seasonal expression of lignin-bispecific caffeic acid /5-hydroxyferulic acid O-methylotransferase of aspen." Plant Molecular Biology 17: 1203-1215. Campbell, M . and R. Sederoff (1996). "Variation in lignin content and composition." Plant Physiology 110: 3-13. 114 Campinhos, E . (1999). "Sustainable plantations of high-yield Eucalyptus trees for production of fibre: Aracruz case." New Forests 17: 129-143. Chabannes, M . , A . Barakate, et al. (2001a). "Strong decrease in lignin content without significant alteration of plant development is induced by simultaneous down-regulation of cinnamoyl C o A reductase (CCR) and cinnamyl alcohol dehydrogenase ( C A D ) in tobacco plants." The Plant Journal 28: 257-270. Chabannes, M . , K . Ruel, et al. (2001b). "In situ analysis of lignins in transgenic tobacco reveals a differential impact o f individual transformations on the spatial patterns of lignin deposition at the cellular and subcellular levels." The Plant Journal 28: 271-282. Chappie, C. (1998). "Molecular-genetic analysis of plant cytochrome P450-dependent monoxygenases." Annual Review of Plant Physiology and Plant Molecular Biology 49: 311-343. Chen, C. (1992). Nitrobenzene and cupric oxide oxidations. Methods in Lignin Chemistry. S. L i n and C. Dence. Berl in, Spriner-Verlag: 578. Chen, C , H . Meyermans, et al. (2000). "Cell-specific and conditional expression of caffeoyl-Coenzyme A-3-O-methyltransferase in poplar." Plant Physiology 123: 853-867. Chiang, V . and M . Funaoka (1990). "The difference between guaiacyl and guaiacyl-syringyl lignins in their responses to kraft delignification." Holzforschung 44: 309-313. Chiang, V . , R. Puumala, et al. (1988). "Comparison of softwood and hardwood kraft pulping." T A P P I Journal 71: 173-176. Clark, J. (1962). "Effects of fiber coarseness and length I. bulk, burst, tear, fold, and tensile tests." T A P P I Journal 45: 628-634. C O F I (2000). The world's forests and forest industry C O F I 2 0 0 0 fact book, COFI . 2002. Dence, C. (1992). The determination of lignin. Methods in Lignin Chemistry. S. L i n and C. Dence. Berl in, Springer-Verlag: 578. Dickmann, D . (2001). A n overview of the genus Populus. Poplar Culture in North America. D . Dickmann, J. Isebrands, J. Eckenwalder and J. Richardson. Ottawa, N R C Research Press: 1-42. Dixon, R., F. Chen, et al. (2001). "The biosynthesis of monolignols: a "metabolic grid", or independent pathways to guaiacyl and syringyl units?" Phytochemistry 57: 1069-1084. Dwivedi , U . , W . Campbell, et al. (1994). "Modification o f lignin biosynthesis in transgenic Nicotiana trhrought expression of an antisense O-methyltransferase gene from Populus.'" Plant Molecular Biology 26: 61-71. 115 Ehlting, J., D . Buttner, et al. (1999). "Three 4-coumarate:coenzyme A ligases in Arabidopsis thaliana represent two evolutionarily divergent classes in angiosperms." The Plant Journal 19: 9-20. Elkind, Y . , R. Edwards, et al. (1990). "Abnormal plant development and down-regulation of phenylpropanoid biosynthesis in transgenic tobacco containing a heterologous phenylalanine ammonia-lyase gene." Proceedings of the National Academy of Sciences, U S A 87: 9057-9061. F A O (2001). State of the World's Forests 2001, Food and Agriculture Organization of the United Nations. September 2002. U R L : www.fao.org/docrep/003/y0900e/y0900e00.htm Fengel, D . and G . Wegener (1989). Wood: chemistry, ultrastructure, reactions. Berlin, Walter de Gruyter. Feuillet, C. , V . Lauvergeat, et al. (1995). "Tissue- and cell-specific expression of a cinnamyl alcohol dehydrogenase promoter in transgenic poplar plants." Plant Molecular Biology 27: 651-667. Franke, R., C. McMichael , et al. (2000). "Modified lignin in tobacco and poplar plants over-expressing the Arabidopsis gene encoding ferulate 5-hydroxylase." The Plant Journal 22: 223-234. Fujita, M . and H . Harada (2001). Ultrastructure and formation of wood cell wall . Wood and Cellulosic Chemistry. D . Hon and N . Shiraishi. New York, Marcel Dekker: 914. Galliano, H . , W . Heller, et al. (1993). "Ozone induction and purification of spruce cinnamyl alcohol dehydrogenase." Phytochemistry 32: 557-563. Gellerstedt, G . and E . Lindfors (1984). "Structural changes in lignin during kraft pulping." Holzforschung 38: 151-158. Goffner, D . , M . Campbell, et al. (1994). "Purification and characterization of cinnamoyl-coenzyme A : N A D P oxidoreductase in Eucalyptus gunnii." Plant Physiology 106: 625-632. Goffner, D . , I. Joffroy, et al. (1992). "Purification and characterization of isoforms of cinnamyl alcohol dehydrogenase from Eucalyptus xylem." Planta 188: 48-53. Grand, C. (1984). "Ferulic acid 5-hydroxylase: a new cytochrome P-450-dependent enzyme from higher plant microsomes involved in lignin synthesis." Federation of European Biochemical Societies Letters 169: 7-11. Grima-Pettenati, J. and D . Goffner (1999). "Lignin genetic engineering revisited." Plant Science 145: 51-65. Guo, D. , F. Chen, et al. (2001). "Downregulation of caffeic acid 3-O-methyltransferase and caffeoyl C o A 3-O-methyltransferase in transgenic alfalfa: impacts on lignin structure and implications for the biosynthesis of G and S lingnin." The Plant Cel l 13: 73-88. 116 Gurnagul, N . , D . Page, et al. (1992). "The effect of cellulose degradation on the strength of wood pulp fibres." Nordic Pulp and Paper Research Journal 7: 152-154. Halpin, C , M . Knight, et al. (1994). "Manipulation o f lignin quality by downregulation o f cinnamyl alcohol dehydrogenase." The Plant Journal 6: 338-350. Harding, S., J. Leshkevich, et al. (2002). "Differential substrate inhibition couples kinetically distinct 4-coumarate:coenzyme A ligases with spatially distinct metabolic roles in quaking aspen." Plant Physiology 128: 428-438. Higuchi, T., T. Ito, et al. (1994). "Red-brown color of lignified issues of transgenic plants with antisense C A D gene: wine-red lignin from coniferyl aldehyde." Journal of Biotechnology 37: 151-158. Hu, W. , S. Harding, et al. (1999). "Repression of lignin biosynthesis promotes cellulose accumulation and growth in transgenic trees." Nature Biotechnology 17: 808-812. Hu, W. , A . Kawaoka, et al. (1998). "Compartmentalized expression of two structurally and functionally distinct 4-coumarate:CoA ligase genes in aspen (Populus tremuloides)." Proceedings of the National Academy of Sciences, U S A 95: 5407-5412. Humphreys, J., M . Hemm, et al. (1999). "New routes for lignin biosynthesis defined by biochemical characterization of recombinant ferulate 5-hydroxylase, a multifunctional cytochrome P450-dependent monooxygenase." Proceedings of the National Academy of Sciences, U S A 96: 10045-10050. Jouanin, L . , T. Goujon, et al. (2000). "Lignification in transgenic poplars with extremely reduced caffeic acid O-methyltransferase activity." Plant Physiology 123: 1363-1373. Kadla, J., S. Kubo, et al. (2002). "Lignin-based carbon fibers for composite fiber applications." Carbon 40: 2913-2920. Kajita, S., S. Hishiyama, et al. (1997). "Structural characterization of modified lignin in transgenic tobacco plants in which the activity of 4-coumarate:Coenzyme A ligase is depressed." Plant Physiology 114: 871-879. Kneusel, R., U . Matern, et al. (1989). "Formation of trans-caffeoyl C o A from trans-4-coumaroyl C o A by Zn 2 +-dependent enzymes in cultured plant cells and its activation by an elicitor-induced p H shift." Archives of Biochemistry and Biophysics 269: 455-462. Kuroda, H . , M . Shimada, et al. (1975). "Purification and properties of Omethyltransferase involved in the biosynthesis of gymnosperm lignin." Phytochemistry 14: 1759-1763. Kutsuki, H . , M . Shimada, et al. (1982). "Distribution and roles of p-hydroxycinnamate:CoA ligase in lignin biosynthesis." Phytochemistry 21: 267-271. Lapierre, C. and B . Monties (1986). "Thioacidolysis of poplar lignins: identification of monomelic syringyl products and characterization of guaiacyl-syringyl lignin fractions." Holzforschung 40: 113-118. 117 Lapierre, C , B . Poller, et al. (2000). Genetic engineering of poplar lignins: impact of lignin alteration on kraft pulping performances. Lignin: historical biological, and materials perspectives. W . Glasser, R. Northey and T. Schultz, 2000 American Chemical Society. 742: 1999:145-160. Lapierre, C , B . Pollet, et al. (1999). "Structural alterations o f lignins in transgenic poplars with depressed cimmamyl alcohol dehydrogenase or caffeic acid o-methyltransferase activity have an opposite impact on the efficency of industrial kraft pulping." Plant Physiology 119:153-163. Lapierre, C , B . Pollet, et al. (1995). "New Insights into the molecular architecture of hardwood lignins by chemical degradative methods." Research on Chemical Intermediates 21: 398-412. Lee, D . and C. Douglas (1996). "Two divergent members of a tobacco 4-coumarate:coenzyme A ligase (4CL) gene family." Plant Physiology 112: 193-205. Lee, D. , K . Meyer, et al. (1997). "Antisense suppression of 4-coumarate:Coenzyme A ligase activity in Arabidopsis leads to altered lignin subunit composition." The Plant Cel l 9: 1985-1998. L i , L . , X . Cheng, et al. (2001). "The last step of syringyl monolignol biosynthesis in angiosperms is regulated by a novel gene encoding sinapyl alcohol dehydrogenase." The Plant Ce l l 13: 1567-1585. L i , L . , K . Osakabe, et al. (1999). "Secondary xylem-specific expression of caffeoyl-coenzyme A 3-O-methyltransferase plays an important role in the methylation pathway associated with lignin biosynthesis in loblolly pine." Plant Molecular Biology 40: 555-565. L i , L . , J. Popko, et al. (2000). "5-hydroxyconiferyl aldehyde modulates enzymatic methylation for syringyl monolignol formation, a new view o f monolignol biosnthesis in angiosperms." The Journal of Biological Chemistry 275: 6537-6545. L i , L . , J. Popko, et al. (1997). " A novel multifunctional O-methyltransferase implicated in a dual methylation pathway associated with lignin biosynthesis in loblolly pine." Proceedings of the National Academy of Sciences, U S A 94: 5461-5466. L u , F. and J. Ralph (1997a). "Derivatization followed by reductive cleavage ( D F R C method), a new method for lignin analysis: protocol for analysis of D F R C monomers." The Journal of Agriculture and Food Chemistry 45: 2590-2592. L u , F. and J. Ralph (1997b). " D F R C Method for lignin analysis. 1. New method for p-aryl ether cleavage: lignin model studies." The Journal of Agriculture and Food Chemistry 45: 4655-4660. L u , F. and J. Ralph (1998). "The D F R C method for lignin analysis. 2. monomers from isolated lignins." The Journal of Agriculture and Food Chemistry 46: 547-552. 118 MacKay, J., D . O'Malley, et al. (1997). "Ingeritance, gene expression, and lignin characterization in a mutant pine deficient in cinnamyl alcohl dehydrogenase." Proceedings of the National Academy of Sciences, U S A 94: 8255-8260. Marita, J., J. Ralph, et al. (1999). " N M R characterization of lignins in Arabidopsis altered in the activity of ferulate 5-hydroxylase." Proceedings of the National Academy of Sciences, U S A 96: 12328-12332. Maury, S., P. Geoffroy, et al. (1999). "Tobacco O-methyltransferases involved in phenylpropanoid metabolism. The diferent caffeoyl-coenzyme A/5-hydroxyferuloyl-coenzyme A 3/5-O-methyltransferase and caffeic acid/5-hydroxyferulic acid 3/5-0-methyltransferase classes have distinct substrate specificities and expression patterns." Plant Physiology 121: 215-223. Merkle, S. and J. Dean (2000). "Forest tree biotechnology." Current Opinion in Biotechnology 11: 298-302. Meyer, K . , J. Cusumano, et al. (1996). "Ferulate-5-hydroxylase from Arabidopsis thaliana defines, a new family of cytochrome P450-dependent monooxygenases." Proceedings of the National Academy o f Sciences. U S A 93: 6869-6874. Meyer, K . , A . Shirley, et al. (1998). "Lignin monomer composition is determined by the expression of a cytochrome P450-dependent monooxygenase in Arabidopsis." Proceedings of the National Academy of Sciences, U S A 95: 6619-6623. Meyer, V . (1998). Practical High-Performance Liquid Chromatography. West Sussex, John Wiley and Sons. Meyermans, H . , K . Morreel, et al. (2000). "Modifications in lignin and accumulation of phenolic glucosides in poplar xylem upon down-regulation of caffeoyl-coenzyme A O-methyltransferase, and enzyme involved in lignin biosynthesis." The Journal of Biological Chemistry 275(47): 36899-36909. Mimms, A . , M . Kocurek, et al. (1989). Kraft pulping a compilation of notes. Atlanta, T A P P I . Mullen, T. and S. Bertrant (1998). "Environmental release of transgenic trees in Canada -potential benefits and assessment of biosafety." The Forestry Chronicle 74: 203-219. N i , W. , N . Paiva, et al. (1994). "Reduced lignin in transgenic plants containing a caffeic acid O-methyltransferase antisense gene." Transgenic Research 3: 120-126. O'Connell, A . , K . Holt, et al. (2002). "Improved paper pulp from plants with suppressed cinnamoyl-CoA reductase or cinnamyl alcohol dehydrogenase." Transgenic Research 11: 495-503. O'Malley, D . , S. Porter, et al. (1992). "Purification, characterization, and cloning of cinnamyl alcohol dehydrogenase in loblolly pine (Pinus taeda L . ) . " Plant Physiology 98: 1364-1371. 119 Osakabe, K . , C . Tsao, et al. (1999). "Coniferyl aldehyde 5-hydroxylation and methylation direct syringyl lignin biosynthesis in angiosperms." Proceedings of the National Academy of Sciences, U S A 96: 8955-8960. Panshin, A . J. and C. de Zeeuw (1980). Textbook of Wood Technology. New York, McGraw-H i l l . Pena, L . and A . Seguin (2001). "Recent advances in the genetic transformation of trees." Trends in Biotechnology 19: 500-506. Pierrel, M . , Y . Batard, et al. (1994). "Catalytic properties of the plant cytochrome P450 C Y P 7 3 expressed in yeast." European Journal of Biochemistry 224: 835-844. Pilate, G. , E . Guiney, et al. (2002). "Field and pulping performances of transgenic trees with altered lignification." Nature Biotechnology 20: 607-612. Pincon, G . , M . Chabannes, et al. (2001). "Simultaneous down-regulation of caffeic/5-hydroxy ferulic acid-O-methyltransferase I and cinnamoyl-coenzyme A reductase in the progeny from a cross between tobacco lines homozygous for eash transgene. Consquences for plant development and lignin synthesis." Plant Physiology 126: 145-155. Piquemal, J., C . Lapierre, et al. (1998). "Down-regulation of cinnamoyl-CoA reductase induces significant changes o f lignin profiles in transgenic tobacco plants." The Plant Journal 13: 71-83. Ralph, J. and J. Grabber (1996). "Dimeric (3-ether thioacidolysis products resulting from incomplete ether cleavage." Holzforschung 50: 425-428. Ralph, J., C. Lapierre, et al. (2001). "Elucidation of new structures in lignins of C A D - and COMT-deficient plants by N M R . " Phytochemistry 57: 993-1003. Ralph, J. and J. M a c K a y (1997). "Abnormal lignin in loblolly pine mutant." Science 277:235-239. Rolando, C , B . Monties, et al. (1992). Thioacidolysis. Methods in Lign in Chemistry. S. L i n and C. Dence. Berlin, Springer-Verlag: 578. Roth, R., A . Boudet, et al. (1997). "Lignification and cinnamyl alcohol dehydrogenase activity in developing stems of tomato and poplar: a spatial and kinetic study through tissue printing." Journal of Experimental Botany 48: 247-254. Sakakibara, A . and Y . Sano (2001). Chemistry of lignin. Wood and cellulosic chemistry. D . Hon and N . Shiraishi. New York, Marcel Dekker: 914. Sarni, F., C. Grand, et al. (1984). "Purification and properties of cinnamoyl-CoA reductase and cinnamyl alcohol dehydrogenase from poplar stems (Populus x euramericana)." European Journal of Biochemistry 139: 259-265. Seth, R. (1995). "The effect of fiber length and coarseness on the tensile strength of wet webs: a statistical geometry explanation." T A P P I Journal 78: 99-102. 120 Seth, R. and B . Chan (1997). "Measurement of fiber coarseness with optical fiber length analyzers." T A P P I Journal 80: 217-221. Sewalt, V . , W . N i , et al. (1997a). "Reduced lignin content and altered lignin composition in transgenic tobacco down-regulated in expression o f :-phenylalanine ammonia-lyase or cinnamate 4-hydroxylase." Plant Physiology 115: 41-50. Sewalt, V . , W. N i , et al. (1997b). "Lignin impact on fiber degradation: increased enzymatic digestibility of genetically engineered tobacco (Nicotiana tabacum) stems reduced in lignin content." Journal o f Agricuture and Food Chemistry 45: 1977-1983. Shimada, M . , H . Fushiki, et al. (1972). "O-methyltransferase activity from Japanese black pine." Phytochemistry 11: 2657-2662. Sitbon, F. , S. Hennion, et al. (1999). "Enhanced ethylene production and peroxidase activity in IAA-overproducing transgenic tobacco plants is associated with increased lignin content and altered lignin composition." Plant Science 141: 165-173. Sjostrom, E . (1993). Wood chemistry: fundamentals and applications. San Diego, Academic Press. Stanturf, J., C. van Oosten, et al. (2001). Ecology and silviculture of poplar plantations. Poplar Culture in North America. D . Dickmann, J. Isebrands, J. Eckenwalder and J. Richardson. Ottawa, N R C Research Press: 153-206. Stenius, P. (2000). Forest Products Chemistry. Helsinki, Fapet Oy. Stewart, D . , N . Yahiaoui, et al. (1997). "Fourier-transform infarared and Raman spectroscopic evidence for the incorporation of cinnamaldehydes into the lignin of transgenic tobacco (Nicotiana tabacum L. ) plants with reduced expression of cinnamyl alcohol dehydrogenase." Planta 201: 311-318. Tamura, T., N . Morohoshi, et al. (2001). "Suitability of peroxidase-suppressed transgenic hybrid aspen (Populus sieboldii x Populus gradidentata) for pulping." Holzforschung 55: 335-339. Teutsch, H . , M . Hasenfratz, et al. (1993). "Isolation and sequence o f a c D N A encoding the Jerusalem artichoke cinnamate 4-hydroxylase, a major plant cytochrome P450 involved in the general phenylpropanoid pathway." Proceedings of the National Academy of Sciences, U S A 90: 4102-4106. Tsai, C , J. Popko, et al. (1998). "Suppression of O-methyltransferase gene by homologous sense transgene in quaking aspen causes red-brown wood phenotypes." Plant Physiology 117:101-112. Van Doorsselaere, J., M . Baucher, et al. (1995). " A novel lignin in poplar trees with a reduced caffeic acid/5-hydroxyferulic acid O-methyltransferase activity." The Plant Journal 8: 855-864. 121 Wang, T. and S. Aitken (2001). "Variation in xylem anatomy of selected populations of lodgepole pine." Canadian Journal o f Forest Research 31: 2049-2057. Whetten, R., J. Mackay, et al. (1998). "Recent advances in understanding lignin biosynthesis." Annual Review of Plant Physiology and Plant Molecular Biology 49: 585-609. Ye , Z. (1997). "Association of caffeoyl coenzyme A 3-O-methyltransferase expression with lignifying tissues in several dicot plants." Plant Physiology 115: 1341-1250. Y e , Z . , R. Kneusel, et al. (1994). " A n alternative methylation pathway in lignin biosynthesis in Zinnia." The Plant Ce l l 6: 1427-1439. Ye , Z . and J. Varner (1995). "Differential expression of two O-methyltransferases in lignin biosynthesis in Zinnia elegans." Plant Physiology 108: 459-467. Zhong, R., W . Morrison, et al. (1998). "Dual methylation pathways in lignin biosynthesis." The Plant Cel l 10: 2033-2045. 122 Appendix A ^ 3 1 2 1 -O l 3 Total Lignin I Y///X Insoluble Lignin ESSS Soluble Lignin I I I 6 5 . 6 7 0 . 7 8 2 . 5 8 3 . 6 8 8 . 8 8 9 . 1 8 9 . 2 9 3 . 4 Mol % S Figure 35. L ignin content of pulp generated from the wild-type and C 4 H - F 5 H transgenic poplar cooked at an H-factor of 856. Values represent the average o f two replicates while error bars represent a range o f 0.1-1.0 (insoluble), 0.08-0.3 (soluble), and 0.4-1.2 (total). Q 5 . 0 4 . 5 4 . 0 3 . 5 3 . 0 2 . 5 2 . 0 1.5 1.0 0 . 5 0 . 0 3 Total Lignin 1 V///X Insoluble Lignin ISS3 Soluble Lignin I 6 i i 6 5 . 6 7 0 . 7 7 9 . 5 8 2 . 5 8 3 . 6 8 8 . 8 8 9 . 1 8 9 . 2 9 3 . 4 Mol % S Figure 36. Lignin content of pulp generated from the wild-type and C 4 H - F 5 H transgenic poplar cooked at an H-factor of 1547. Values represent the average of two replicates while error bars represent a range of 0.1-1.0 (insoluble), 0.05-0.2 (soluble), and 0.2-1.2 (total). 123 Appendix B Figure 37. Carbohydrate contents of pulp produced from wild-type and C 4 H - F 5 H transgenic poplar pulped at an H-factor of 856. Values represent the average of two replicates while error bars represent a range o f 0.3-4.1 (Glu) and 0.05-1.2 (Xyl) . Figure 38. Carbohydrate contents of pulp produced from wild-type and C 4 H - F 5 H transgenic poplar pulped at an H-factor of 1547. Values represent the average of two replicates while error bars represent a range of 04-7.4 (Glu) and 0.01-0.8 (Xyl) . 124 Appendix C Table 6. Pulp coarseness (mg/m) of wild-type and C 4 H - F 5 H transformed poplar processed at different H-factors. Tree Line S:G Ratio H-Factor 2110 1784 1547 1096 856 Wild-type 1.90 0.087 0.109 0.190 0.255 0.389 C 4 H - F 5 H 85 2.41 0.107 0.191 0.239 C 4 H - F 5 H 2 1 3.87 0.079 0.117 N D C 4 H - F 5 H 82 4.72 0.107 0.154 0.132 C 4 H - F 5 H 4 1 5.09 0.120 0.161 0.213 C 4 H - F 5 H 37 7.95 0.106 0.121 0.126 C 4 H - F 5 H 65 8.21 0.086 0.089 0.103 C 4 H - F 5 H 26 8.26 0.073 0.078 0.081 C 4 H - F 5 H 64 14.17 0.098 0.108 0.105 Coarseness values are the average three replicates where between 15,000-20,000 individual fibres were measured per pulp sample, averages have a range of 0.0004-0.07 mg/m. 125 Appendix D E o < 6 5 . 6 'A' 7 0 . 7 8 2 . 5 8 3 . 6 8 8 . 8 •!h l Vi K'. * ii 8 9 . 1 - 8 9 . 2 / ' / >« ! . V \ - — • 9 3 . 4 — i — i — i — i — . i , i I 1 1 1 1 I I • I 0 1 0 2 0 3 0 4 0 Time (min.) Figure 39. Elution times (relative molecular weight) o f black liquor lignin liberated into the pulping liquor at an H-factor of 856 for wild-type and transgenic trees. Figure 40. Elution times (relative molecular weight) of black liquor lignin liberated into the pulping liquor at an H-factor of 1547 for wild-type and transgenic trees. 126 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items