Bioactive and Protective Polyphenolics from Roots and Stumps of Conifer Trees (Norway Spruce and Scots Pine)

Chemistry Faculty of Science University of Helsinki Finland

BIOACTIVE AND PROTECTIVE POLYPHENOLICS FROM ROOTS AND STUMPS OF CONIFER TREES (NORWAY SPRUCE AND SCOTS PINE)

Harri Latva-Mäenpää

ACADEMIC DISSERTATION

To be presented, with the permission of the Faculty of Science of the University of Helsinki, for public examination in Lecture hall 10, University main building, on 9th June 2017, at 12 noon.

Helsinki 2017 Supervisors Research Director, Professor Kristiina Wähälä Organic Chemistry Faculty of Medicine University of Helsinki Finland

Docent Pekka Saranpää Principal Research Scientist New business opportunities Natural Resources Institute Finland (Luke) Finland

Reviewers Professor Kumi Yoshida Organic Chemistry Graduate School of Information Science University of Nagoya Japan

Professor Corrado Tringali Organic Chemistry Dipartimento di Scienze Chimiche University of Catania Italy

Opponent Dr. Emerson Ferreira Queiroz Senior Lecturer School of Pharmaceutical Sciences Phytochemistry and Bioactive Natural Products University of Geneva Switzerland

ISBN 978-951-51-3465-3 (paperback) ISBN 978-951-51-3466-0 (PDF)

Unigrafia Oy Helsinki 2017

ABSTRACT

The current era of bioeconomy strives to develop new, innovative products from natural raw materials in a sustainable way. In this PhD thesis, we have searched potential bioactive compounds from different parts of the neglected roots and stumps of conifer trees for high value products. Our aim was to establish deeper knowledge concerning chemical compositions of specified and defined biomass sources (roots and stumps of conifer trees) as well as the chemistry of certain natural polyphenolic compounds in different conditions. Using methods of natural products chemistry, such as solvent extraction and chromatographic separation and analysis techniques (GC-MS, HPLC- DAD/MS), we found that the root bark of Norway spruce is a rich source of stilbenoid glucosides (astringin, isorhapontin and piceid). These have structural similarities with the most studied stilbenoid, resveratrol. We also found that the stumps of Scots pine are a source of stilbenoids, pinosylvin and its derivatives. Stilbenoids are bioactive compounds and they have shown a variety of beneficial activities in studies related for example to human health. They have also shown antimicrobial and protective properties in trees and also in tests against micro-organisms. The highest concentrations of stilbenoid glucosides were found from the inner bark of the root zones closest to the stem of Norway spruce. This study also revealed that the root neck of Norway spruce is a new source of the bioactive lignan hydroxymatairesinol (HMR). Chromatographic and spectroscopic methods were developed to isolate pure stilbenoids from the bark of Norway spruce roots for further experiments. The complete structural elucidation of stilbenoids from Norway spruce was performed by HPLC-DAD-NMR. The HPLC-DAD method was further optimized to isolate stilbenoids in a semipreparative scale. The invividual stilbenoid molecules were isolated from the extracts and their photochemical stabilities (under fluorescent and UV light) were studied by HPLC-DAD, MS and NMR. Naturally-occurring trans stilbenoids undergo photochemical transformations to cis forms. However, the extended UV irradiation caused stilbenoids to form phenanthrene structures by intramolecular cyclisation. This is the first time that these new phenanthrene structures are reported to be derived from stilbenoids of spruce bark. Our results show that that the roots and stumps of conifer trees contain bioactive polyphenolic compounds, such as stilbenoids and lignans. These molecules and the compounds formed thereof may have commercial value and the information gained would offer a base platform for the possible new commercial high-value products developed from forest biomass. Our study has also provided new scientific knowledge of natural compounds and their properties that may lead to new findings and deeper understanding of the

3 physiology and internal protection processes of plants by their bioactive secondary metabolites. Roots and stumps of conifer trees (Norway spruce and Scots pine) are a vast source of biomass containing potentially bioactive polyphenolic extractives as shown in this work, but this biomass is currently used for low- value energy production only. With already existing harvesting techniques, roots and stumps of conifer trees could be used as a source of commercially valuable biochemicals.

ACKNOWLEDGEMENTS

This study was carried out in a cooperation between the Laboratory of Organic Chemistry, Department of Chemistry, University of Helsinki and the Finnish Forest Research Institute, Metla (nowadays Natural Resources Institute Finland, Luke). I wish to thank both organizations and several people for contributing their resources, time and expertise to this thesis by helping and supporting me to complete this work. I am grateful to my supervisors Professor Kristiina Wähälä and Docent Pekka Saranpää to give me this opportunity to dive into the interesting world of biomass and polyphenolic compounds. A major appreciation also to Docent Tytti Sarjala for her support. This has been a long journey but all of you have kindly provided me the tools to grow up as a researcher and also as a person. Your endless support, coaching and mentoring has taken me this far. I wish to thank Professors Kumi Yoshida and Corrado Tringali for reviewing the manuscript of the thesis and Docent Emerson Ferreira Queiroz for agreeing to act as my opponent at the public defence of this dissertation. Many thanks to my co-authors, all personnel and students (especially Daniel for his great work) in the laboratories of Metla/Luke in Vantaa and Parkano and in the laboratory of Organic Chemistry, University of Helsinki. Special thanks also to Anneli, Eeva, Irmeli and three Tapios for their valuable work with our samples. I wish to thank all my colleagues for their time and support to help and discuss how to proceed in the project. During the project, I was able to start working in the natural products industry to develop and commercialize products derived from biomass. The combination of academic research and business has given me valuable perspectives for the future. I am grateful to an entrepreneur Mr. Stein Ulve for this opportunity to learn so much about the business and its reality. We Finns may still need you fearless Norwegians to carry on the risks and attack the developing markets. Thanks also to my current working colleagues. Besides the work, I wanted to highlight the brotherhood in our North Stars society. You guys have given me something that is more valuable than work, friendship. I warmly thank my parents Merja and Hannu and my sister Henna for showing the examples how hard work and continuous education of yourself will lead to great things. Finally, the most important, the love of my life, Laura. There are no words that would adequately describe the value of your support. You and your love are always available. You have always believed in me and encouraged me to success in everyday life. Without you I wouldn’t be at this point. You, our boy Noel and the future newcomer have shown to me the most presious things in life. I am so grateful to have all of you with me.

5 CONTENTS

Abstract ...... 3 Acknowledgements ...... 5 Contents ...... 6 List of original publications ...... 8 Abbreviations ...... 9 1 Introduction ...... 10 1.1 Background ...... 10 1.1.1 Bioeconomy, business, politics, sustainability ...... 10 1.1.2 Roots and stumps as a raw material of the future ...... 12 1.1.3 Modern analysis methods build knowledge of natural products chemistry ...... 14 1.2 Aims of the study ...... 15 2 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine ...... 16 2.1 Stilbenoids ...... 16 2.1.1 Biosynthesis of stilbenoids ...... 16 2.1.2 Stilbenoids in Norway spruce and Scots pine ...... 18 2.1.3 Extraction and analysis of stilbenoids from coniferous biomass ...... 19 2.1.4 Antioxidant activity of stilbenoids ...... 21 2.1.5 Medicinal properties of stilbenoids ...... 22 2.1.6 Antimicrobial and protective properties of stilbenoids ..... 23 2.2 Stability of stilbenoids ...... 26 2.3 Commercial utilisation of potential bioactive compounds .... 27 3 Experimental ...... 29 3.1 Sample material ...... 29 3.2 Extraction ...... 31 3.3 Analysis methods ...... 31 3.4 Isolation of stilbenoids after extraction ...... 33 3.5 Stability of stilbenoids ...... 34 4 Results and discussion ...... 36 4.1 General extractive profiles of the root and stump biomass of conifer trees ...... 36

4.2 Bottom of the root neck of norway spruce as a source of lignans ...... 44 4.3 Extraction and purification of stilbenoids from the bark of Norway spruce ...... 45 4.4 Structural elucidation of stilbenoids by UV, MS and NMR ... 47 4.5 Stability of stilbenoids ...... 49 4.5.1 Storage...... 49 4.5.2 Photochemical reactions and stability (under fluorescent light and UV irradiation) ...... 50 4.5.3 Fractionation and structural elucidation of new compounds derived from stilbenoids ...... 56 5 Concluding remarks ...... 60 References ...... 62

7 Introduction

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following publications:

I Latva-Mäenpää, H., Laakso, T., Sarjala, T., Wähälä, K. and Saranpää, P. (2013) Variation of stilbene glucosides in bark extracts obtained from roots and stumps of Norway spruce (Picea abies [L.] Karst.). Trees - Structure and Function, 27 (1), 131–139.

II Latva-Mäenpää, H., Laakso, T., Sarjala, T., Wähälä, K. and Saranpää, P. (2014) Root neck of Norway spruce as a source of bioactive lignans and stilbenes. Holzforschung, 68 (1), 1–7.

III Mulat, D., Latva-Mäenpää, H., Koskela, H., Saranpää, P. and Wähälä, K. (2014) Rapid chemical characterisation of stilbenes in the root bark of Norway Spruce by off-line HPLC/DAD-NMR. Phytochemical Analysis, 25 (6), 529–536.

IV Latva-Mäenpää, H., Mulat, D., Wufu, R., Saranpää, P. and Wähälä, K. (2017) Photochemical stability and transformation of stilbenes from the root bark of Norway spruce. Manuscript in preparation

The publications are referred to in the text by their roman numerals (I-IV). The published articles (I-III) are reproduced in the printed version of this thesis with the kind permission of the publishers.

Other publications:

Jyske, T., Laakso, T., Latva-Mäenpää, H., Tapanila, T. and Saranpää, P. (2014). Yield of stilbene glucosides from the bark of young and old Norway spruce stems. Biomass & Bioenergy, 71, 216–227.

Muilu-Mäkelä, R., Vuosku, J., Hamberg, L., Latva-Mäenpää, H., Häggman, H. and Sarjala, T. (2015) Osmotic stress affects polyamine homeostasis and phenolic content in proembryogenic liquid cell cultures of Scots pine. Plant, Cell, Tissue and Organ Culture, 122 (3), 709–726.

ABBREVIATIONS

AD Alzheimer’s disease AMD Age-related macular degeneration AMR Antimicrobial resistance ASE Accelerated Solvent Extraction C17 Heptadecanoic acid C21 Heneicosanoic acid D1.3, mm Stem diameter DAD Diode Array Detection DW Dry weight EFSA European Food Safety Authority EI-MS Electron Impact Ionization-Mass spectrometry ESI Electron-Spray Ionization FDA United States Food and Drug Administration FID Flame Ionization Detection FW Fresh Weight GC Gas Chromatography GC-MS Gas Chromatography-Mass Spectrometry HPLC High-Performance Liquid Chromatography HMR Hydroxymatairesinol h Hours hW Heartwood ICP-AES Inductively Coupled Plasma Atomic Emission Spectrometry iB inner bark Luke Natural Resources Institute Finland MDR Multidrug resistance Metla Finnish Forest Research Institute MS Mass Spectrometry NMR Nuclear Magnetic Resonance spectroscopy OTMS Trimethylsilyloxy groups oB Outer bark PHWE Pressurised Hot Water Extraction PS Pinosylvin PSM Pinosylvin monomethyl ether RP-HPLC Reversed Phase Liquid Chromatography RN Root Neck sW Sapwood (outer) part of wood material TOF Time Of Flight tR Retention time in HPLC chromatogram UV Ultraviolet WHO World Health Organization wt Weight

9 Introduction

1 INTRODUCTION

1.1 BACKGROUND

1.1.1 BIOECONOMY, BUSINESS, POLITICS, SUSTAINABILITY

With a growing interest in plant polyphenolics, there is a need to find alternative sources of these compounds in reasonable amounts. The initial idea of the present study was that polyphenolic compounds and other extractives could also be gained from the root and stump biomass of conifer trees abundant in Finland, Norway spruce (Picea abies [L.] Karst.) and Scots pine (Pinus sylvestris), which are mainly utilized for bioenergy production or left in the forest after harvesting. This study will support the use the use of renewable natural resources in an efficient way to produce new high-value products thereof and is in line with the new era of bioeconomy and its new biorefinery concepts. According to Finnish governmental bioeconomy strategy (updated in 2016), the bioeconomy is based on renewable natural resources and new operating models. It relies on renewable natural resources to produce food, energy, products and services. Bioeconomy strives to reduce our dependence on fossil natural resources, to prevent biodiversity loss and to create new economic growth and jobs in line with the principles of sustainable development. The most important renewable resources in Finland are biomass or organic matter in the forests, soil, fields, water bodies and the sea, and fresh water. Bioeconomy has a significant role in Finnish national economy by its 16% share and its 60 billion euros output (Finnish Bioeconomy strategy 2016). The aim in Finland is to increase the output to 100 billion euros and to create 100000 new jobs (300 000 jobs being the current figure) by 2025. The forest sector has already a major role in bioeconomy since over one half of bioeconomy output today relies on Finnish forests.

The Finnish bioeconomy strategy will be implemented by the following strategic goals (Finnish Bioeconomy strategy 2016): x A competitive operating environment will be created for bioeconomy growth, x New businesses will be generated in the bioeconomy by means of risk financing, bold experiments and the crossing of sectoral boundaries, x The bioeconomy competence base will be upgraded by developing education, training and research, x Availability of biomasses, well-functioning raw material markets and sustainability of the use of biomass will be secured.

The Finnish National Support Group to the Forest-based Sector Technology Platform (2016) has given its recommendations for the Finnish forest-based bioeconomy R&D actions by cross-sectoral development work: strengthening the competence and knowledge and identifying of new value-added products to create new business and also affect positively on export opportunities. One of the exact research themes identified was the development of bark extractives (and their small molecules) to new bioactive products.

The complex and developing environment of biomass-refining The design and relationships paths in the area of biomass-refining must not be seen as straightforward due to multiple factors affecting the whole picture (Figure 1.1). Value creation and development of new high-value products based on non-utilised natural resources is influenced by political and environmental decisions. Also the developing concept of bioeconomy and corporate/governmental strategies will play a role in defining the wider picture of biorefining. The biomass may be refined to higher value products (natural extracts or isolated compounds with functional properties) that will finally lead to new bioproducts. This will create new sustainable business. Research efforts and building of knowledge and understanding are needed to successfully execute the process of new value creation. Enhanced understanding of the chemistry of natural bioactive molecules and their behaviour will build new knowledge related to natural compounds, which may also help to explain the phenomena in plant physiology. Studies of natural products chemistry will serve our society in two ways, which should be seen by a synergestic view of economic growth and deeper understanding of nature.

Fig. 1.1. Design and relationship paths of the field of research and utilization of biomass.

11 Introduction

1.1.2 ROOTS AND STUMPS AS A RAW MATERIAL OF THE FUTURE Wood-derived biomass (wood, bark, needles, leaves, roots etc.) contains bioactive compounds, which could be extracted and utilised as new products. A vast source of raw material is available from industrial side streams (e.g. debarking of logs) and residues from harvesting processes (e.g. roots and stumps) (Figure 1.2).

Fig. 1.2. Sampling of the roots of Norway spruce (II).

Norway spruce and Scots pine are among the most widespread conifers in Europe and they have vast economic significance in forestry for pulp and sawn timber products. According to statistical information provided by Natural Resources Institute Finland – Luke (formerly Finnish Forest Research Institute – Metla), the total roundwood removals of Norway spruce and Scots pine were annually between 21.3 – 24.0 million m3 and 25.9 – 29.6 million m3, respectively. 1.3 million m3 of stump biomass was annually used in energy production in Finland between 2002 and 2011 (Bergström and Matisons 2014), but the harvestable stump volume would have been 8.0 million m3 (3.5, 3.0 and 1.5 million m3 of spruce, pine and broadleaf stumps, respectively). Norway spruce and Scots pine typically grow on mineral soil, which is characteristic of Finnish forests, but spruce growing on peatland was also studied. Besides efficiently used forest areas with the relatively high growth mentioned, there are over 0.5 million ha of drained nutrient poor peatlands in Finland on which continuous forestry production is not feasible. In those areas there is between 15–50 t ha-1 of exploitable wood biomass. From that biomass 20–30% is stumps and thick roots. The physical properties of peatland are characterized by high organic matter content and much lower bulk density (Päivänen and Paavilainen 1996). Roots and stumps are usually left in the forest after felling the timber or are used for combustion and energy production, which may cause emissions that

are comparable to the emissions from fossil fuels (Repo et al. 2011). However, when desirable by environmental, economic or political decisions to remove roots and stumps from the forests, these materials should be also considered to be used for the production of high-value products. With already existing removal techniques, roots and stumps of conifer trees could be used as a source of commercially valuable biochemicals. In this PhD thesis, we have searched potential biochemical and future bioactive compounds from different parts of the roots and stumps of conifer trees for high value products (Figure 1.3).

Fig. 1.3. Structure and sampling scheme of the roots and stumps of Norway spruce.

The general chemical composition (containing cellulose, hemicelluloses, lignin and extractives) of various parts of wood-biomass of Norway spruce and Scots pine are shown in Table 1.1. Tree bark is the first barrier of the tree against various harmful external threaths. Due to attacks, the concentrations of secondary metabolite compounds (extractives) such as terpenoid resins and polyphenols increase to protect the tree (Evensen et al. 2000, Zeneli et al. 2006, Wadke et al. 2016). The bark of Norway spruce stems contains phenolic compounds such as stilbenoids, flavonoids and tannins (Pietarinen et al. 2006, Li et. al 2012). However, the function of the root bark is not so clearly defined compared to stem bark. Especially stilbenoids, protective compounds of the conifer trees, are seen as one the most interesting polyphenolic compounds groups in plants, and they have shown a variety of bioactive properties. Due to this potential, these compounds are reviewed in more detail in the following chapters of this thesis.

13 Introduction

Table 1.1 Proportions of the main chemical compound groups (%) in the indicated biomass components of Norway spruce and scots pine (Bergström and Matisons 2014)

Cellulose Hemicelluloses Lignin Extractives NORWAY SPRUCE Stem wood 42.0 27.3 27.4 2.0 Bark 26.6 9.2 11.8 32.1 Branches 29.0 30.0 22.8 16.4 Needles 28.2 25.4 8.4 43.3 Stump 42.9 27.9 29.4 3.8 Roots 29.5 19.2 25.5 15.7 SCOTS PINE Stem wood 40.7 26.9 27.0 5.0 Bark 22.2 8.1 13.1 25.2 Branches 32.0 32.0 21.5 16.6 Needles 29.1 24.9 6.9 39.6 Stump 36.4 28.2 19.5 18.7 Roots 28.6 18.9 29.8 13.3

1.1.3 MODERN ANALYSIS METHODS BUILD KNOWLEDGE OF NATURAL PRODUCTS CHEMISTRY Modern analytical methods are needed to understand more of the chemistry of natural compounds before their larger-scale commercial utilization. These methods provide the tools for the identification, analysis and also isolation of chemical compounds derived from nature. Deeper knowledge of the chemical properties (such as the stability of stilbenoids) of natural compounds is needed to fulfill the compliance requirements for the safe use and regulatory issues of the new products developed. Gas chromatography-mass spectrometry (GC-MS) (Pietarinen et al. 2006, Balas and Popa, 2007) and high-performance liquid chromatography-diode array detection (HPLC-DAD) (Mannila and Talvitie, 1992, Wolfender et al., 1998) and high-performance liquid chromatography-mass spectrometry (HPLC-MSn) are presently used techniques for the identification and quantification of stilbenoids and other small-sized polyphenolics in plant extracts (Buiarelli et al., 2007). However, to our knowledge, for example liquid chromatography combined with nuclear magnetic resonance (NMR) spectroscopy (HPLC-NMR) affording unambiguous identification of novel structures has not been utilized in stilbenoid research at all. In our studies, we have used the following analytical methods: GC-MS, HPLC-DAD, HPLC-MS, MS, NMR and HPLC-NMR.

1.2 AIMS OF THE STUDY

There is very little information about the extractive compositions of wood and bark material of the roots and stumps of Norway spruce and Scots pine. Only a few studies about stilbenoids and other polyphenolic compounds from root bark (Lindberg et al. 1992, Pan and Lundgren, 1995) or fine roots, mycorrhizas and non-mycorrhizal roots of Norway spruce (Beyer et al. 1993, Münzenberger et al. 1990) have been published. In earlier studies concerning roots, samples represented only a certain fraction of the root and the site type of the sample trees has not been specified in previous studies. Due to a lack of specific information of the sample material and unspecified sampling setting, there was a need for more data of different roots fractions and their extractive composition. Our aim was to describe an extraction and analysis procedure for roots and stumps of conifer trees (which are largely an uncovered source of potential bioactive compounds) and analyse the compositions of the extracts from the stump and also from different specified cross sections of the roots of the trees, which were of varying ages and sizes. Another objective of this study was to describe the chemical composition of the extracts from the heartwood and sapwood of the root neck (RN) area of Norway spruce. Furthermore, the localization of stilbenoid glucosides and the chemical composition of the extracts in the inner and outer part of the root bark were examined. Also, the efficiency of pressurised hot water extraction (PHWE) as a green technique for industrial scale production was evaluated. One of the objectives of this study was to develop and establish a simple and rapid procedure of off-line HPLC-DAD-NMR for the unambiguous structural assignment of closely related stilbenoid structures in root bark of of Norway spruce (III). In addition, the aim was to optimize and scale up isolation methods for stilbenoids by preparative chromatography. The stability studies of stilbenoids have given contradictory results and been concentrated mainly on resveratrol and its glucoside piceid. Therefore, the objective of this work was to study the stability of other biologically interesting stilbenoid glucosides such as astringin and isorhapontin and their aglucones piceatannol and isorhapontigenin, which have not yet been studied (IV). To resolve this issue, our aim was to study the effects of light and storage on the stability of stilbenoids and put more effort on the identification and characterization of new compounds formed during our experiments. By all these studies, our aim has been to establish deeper knowledge concerning chemical compositions of specified and defined biomass sources (roots and stumps of conifer trees) as well as the chemistry of certain natural polyphenolic compounds in different conditions. This information would offer a base platform for the possible new commercial high-value products developed from forest biomass.

15 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

2 BIOACTIVE POLYPHENOLICS OF ROOTS AND STUMPS OF NORWAY SPRUCE AND SCOTS PINE

The main emphasis of our study is on stilbenoids. Stilbenoids are one of the most interesting groups of polyphenolic compounds and they are already used in a variety of applications in the fields of medicine, nutraceuticals, cosmetics and materials (Likhtenstein 2010). The most well-known and studied stilbenoid compound is trans-resveratrol (Figure 2.2).

2.1 STILBENOIDS

The basic structure of stilbenes is 1,2-diphenylethene, which has two isomeric forms, trans (E) and cis (Z) (Figure 2.1). The trans form is more stable because the cis form is sterically hindered (Likhtenstein 2010). Stilbene derivatives, called stilbenoids, are secondary metabolites of plants and act as phytoalexins (i.e. defense substances with antimicrobial properties). Stilbenoid synthesis in plants is induced as a response to biotic and abiotic stresses (Chong et al. 2009). Stilbenoids may be used as starting materials for various chemical syntheses because of their reactive hydroxyl (OH) groups.

Fig. 2.1 Chemical structure of the trans stilbene molecule.

2.1.1 BIOSYNTHESIS OF STILBENOIDS Plant stilbenoids are formed from phenylalanine via the phenylpropanoid pathway (Hammerbacher et al. 2011, Li et al. 2012, Kiselev et al. 2016a, Kiselev et al. 2016b). The pathway mediated by stilbene synthase may go through various reactions such as isomerization, glycosylation, methoxylation, oligomerization and isoprenylation to produce stilbene derivatives (Chong et al. 2009). According to Hammerbacher et al. (2011), the first step of stilbene biosynthesis in spruce is the formation of resveratrol and other stilbenoids. For example resveratrol is then modified by 3’-hydroxylation and 3-O- glycosylation to yield one of the major stilbene glucosides, astringin (Figure

2.2). The formation of isorhapontin (the other major stilbenoid glucoside) involves 3’-O-methylation.

Fig. 2.2. A suggested biosynthetic pathway for astringin formation in Norway spruce (OGlu = -O- β-D-glucoside) (Hammerbacher et al. 2011).

Stilbenoids in Norway spruce bark are mainly in a glucosidic form (Weissmann 1981, Mannila and Talvitie 1992). They are isorhapontin (3-O-β- D-glucosyl-4',5-dihydroxy-3'-methoxystilbene), astringin (3-O-β-D-glucosyl- 3',4',5-trihydroxystilbene) and piceid (3-O-β-D-glucosyl-4',5- dihydroxystilbene), which is also known by other trivial names, such as polydatin and resveratrol-glucoside. Spruce bark may also contain smaller amounts of the stilbene aglucones isorhapontigenin (3,4',5-trihydroxy-3'- methoxystilbene), piceatannol (3,3',4',5-tetrahydroxystilbene) and resveratrol (3,4',5-tetrahydroxystilbene). The aglucones may be produced from their glucosides by hydrolysis (Wang et al. 2007). Scots pine heartwood contains pinosylvin (3,5-dihydroxystilbene) and its methylated derivatives (Harju et al. 2009). The structures of stilbenoids studied and reviewed in this work are presented in Tables 2.1 and 2.2.

17 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

Table 2.1. Stilbenoids from Norway spruce. Table 2.2. Stilbenoids from Scots pine.

Stilbenes R1 R2 Stilbenes R1 R2 trans-resveratrol OH H trans-pinosylvin OH OH

trans-piceatannol OH OH trans-pinosylvin OH OCH3

trans-isorhapontigenin OH OCH3 monomethyl ether

trans-isorhapontin OGlu* OCH3 trans-pinosylvin OCH3 OCH3 trans-astringin OGlu* OH dimethyl ether trans-piceid OGlu* H *OGlu = -O-β-D-glucoside

2.1.2 STILBENOIDS IN NORWAY SPRUCE AND SCOTS PINE The concentration of stilbenoids may vary significantly between tree species and also in other plants, such as grapevines (Guerrero et al. 2016). Stilbenoids are secondary metabolities with protective functions and they occur as free compounds or as derivatives such as glucosides in woody biomass (Zeneli et al. 2006, Evensen et al. 2000, Wadke et al. 2016). Trees have been shown to produce stilbenoid aglucones from their glucosides after fungal attack (Viiri et al. 2001). Stilbenoids also react with tannins to form complex structures including flavonoid, stilbene and glucose units (Figure 2.3) (Zhang and Gellerstedt 2008). In addition, stilbenoids may also form dimers with each other (Figure 2.4) (Li et al. 2008). Stilbenoid glucosides have been observed in the bark of stemwood and in unspecified root sections of Norway spruce (Solhaug 1990, Lindberg et al. 1992, Beyer et al. 1993). Stilbenoids have been localized to axial parenchyma cells of the phloem of Norway spruce (Jyske et al. 2014, Jyske et al. 2015, Jyske et al. 2016). Stilbenoid concentrations have shown variations between growth seasons and parts of the tree or bark. Stilbenoids of Scots pine (pinosylvin) are mainly localized to the heartwood region of the wood and they were reported for the first time by Erdtman as early as 1939 (Venäläinen et al. 2004, Hoveltstad et al. 2006). The highest concentrations of pinosylvin compounds have been found from pine knots (Willför et al. 2003, Hovelstad et al. (2006). Formation of pinosylvin was induced in young seedlings of pine after drilling holes through the stem (Harju et al. 2009)

The reported amounts of stilbenoids in different plant parts of Norway spruce and Scots pine are presented in Chapter 4.1.

Fig. 2.3. Condensed tannin structure including stilbene, flavonoid and glucose units (R1 = glucose 2 of H, R = OH or OCH3, T = tannin) (Zhang and Gellersted 2008). Fig. 2.4. Stilbene glucoside dimers (R = H or CH3, OGlc = -O-β-D- glucoside) (Li et al. 2008).

2.1.3 EXTRACTION AND ANALYSIS OF STILBENOIDS FROM CONIFEROUS BIOMASS Extraction Stilbenoids may be obtained from biomass by solvent extraction. The raw material used is collected from forest and then usually dried and ground before extraction processes. The basic drying methods for biomass are air-drying and freeze-drying. Stilbenoids can be extracted from biomass by polar organic solvents such as acetone, ethanol and methanol and with water-organic solvent mixtures (Lindberg et al. 1992, Pietarinen et al. 2006, Co et al. 2012, Fang et al. 2013, Pineiro et al. 2016, I, II). Hydrophobic compounds may be removed with nonpolar solvents (pentane, hexane) from the biomass before the actual extraction of stilbenoids and other phenolic compounds of interest (Evensen et al. 2000, Pietarinen et al. 2006, III). One of the traditional methods for the biomass extraction to obtain stilbenoids is Soxhlet extraction (Pietarinen et al. 2006, Välimaa et al. 2007, I). Accelerated Solvent Extraction (ASE) is nowadays one of the most used and less labor-intensive techniques developed for sequential extraction. ASE has been applied for the extraction of stilbenoids from the bark and knot material of Norway spruce and Scots pine (Pietarinen et al. 2006, Välimaa et al. 2007, Co et al. 2012, Fang et al. 2013, III).

19 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

Environmentally benign extraction methods are needed for the isolation of natural products on an industrial scale. Chemat et al. (2012) have introduced six principles of green extraction. One of those is the use of water as a solvent. Mantegna et al. (2012) have applied ultrasound-assisted water extraction with cyclodextrin encapsulation for the isolation of resveratrol and piceid from Polygonum cuspidatum. It is important to critically evaluate what kind of enhancement methods for the extraction processes (such as ultrasound) are already applicable in an industrial scale. Also solvent recycling and energy efficiency are critical parameters to take into account, when defining the larger scale commercial methods.

Analysis methods Proper separation of stilbenoids from other compounds by chromatography is needed in order to analyse stilbenoids from the extracts obtained. Solhaug (1990) used paper chromatography to separate stilbenoid glucosides from the bark and needle extracts of Norway spruce. The detection of compounds was performed by ultraviolet (UV) and fluorescence spectrometry. Reversed phase liquid chromatography (RP-HPLC) is one of the most commonly used methods for the separation and analysis of stilbenoids. Stilbenoids are easily separated on a C18 column and methanol or acetonitrile with water are used as a mobile phase. Small amounts of buffers may be added to the mobile phase to enhance the separation or detection. Lindberg et al. (1992) and Evensen et al (2000) used HPLC with UV detection to analyse stilbenoids from the bark of Norway spruce. HPLC coupled to diode array detection (DAD) and a mass spectrometer (MS) has also been applied to the analysis of bark stilbenoids (Co et al. 2012) and products derived from trans- resveratrol (Montsko et al. 2008). Mass spectrometry was also utilised in a tandem MS mode to obtain mass fragmentations of stilbenoids. Gas chromatography (GC) is also a traditionally used method for the analysis of stilbenoids. Stilbenoids are analysed by GC as silylated derivatives, formed by the chemical conversion of hydroxyl groups (OH) of stilbenoids to trimethyl silyloxy groups (OTMS). The detection and identification of stilbenoids is usually done by GC-MS (Viiri et al. 2001, Pietarinen et al. 2006, Hovelstad et al. 2006, Fang et al. 2013). When retention times of analytes are known, also flame ionization detection (FID) may be applied for the analysis of stilbenoids (Pietarinen et al. 2006). A quantification of known analytes is usually performed by GC-FID but also GC-MS has been applied for this purpose. Nuclear magnetic resonance (NMR) spectroscopy gives information about the chemical structure of the compounds studied. The compounds need to be separated and isolated from extracts before NMR analyses to receive clear, compound-specific signals. NMR has been applied to the structural elucidation of stilbenoids and their glucosides as well as of stilbenoid dimers from the bark extracts of Norway spruce and pine species, and from hairy root cultures of Vitis vinifera (Pan and Lundgren 1995, Zhang and Gellerstedt

2008, Li et al. 2008, Ioannidis et al. 2016, Tisserant et al. 2016). In recent years, HPLC–NMR has become an attractive technique in the field of natural product research for the expedient identification of organic compounds (Jaroszewski 2005a, 2005b). The use of HPLC-NMR in conjunction with HPLC–DAD, HPLC-MS and HPLC–MS/MS analysis proved to be valuable for the identification of certain classes of natural products in plant extracts (Wolfender et al. 1998). However, as of today, there have been no reports about the use of HPLC-NMR for the rapid and unambiguous structural identification of stilbenoids in plant extracts.

Isolation methods Methods for isolating stilbenoids from the extracts, for example for the characterization, biological testing or other applications have been developed. Adsorption-desorption methods with polymeric adsorbents may be used for purification (Soto et al. 2011). Relatively pure stilbenoids can be obtained from purified extracts by flash chromatography or preparative HPLC (Mannila and Talvitie 1992, Shibutani et al. 2004).

2.1.4 ANTIOXIDANT ACTIVITY OF STILBENOIDS Numerous research findings have shown that resveratrol, the most studied stilbenoid, is a potential anti-oxidant across a variety of assays. Due to the structural similarities between resveratrol and other stilbenoids found in plants, it has been hypothesized that other stilbenoids and also extracts rich in stilbenoids may possess potential anti-oxidant activities. According to Merillon (1997), different isomers (cis and trans) of stilbenoids may have different antioxidative effects. The isomeric form of stilbenoids has not always been specified in the literature, but stilbenoids are naturally-occurring mainly in the trans form. The potency of antioxidant activity may differ between different stilbenoids due to the positions of –OH groups in the structures (Castellano et al. 2014). Stilbenoids have shown promising medicinal properties connected to to their antioxidant activity against harmful oxidative stress and inflammation pathways, which are related to diseases, such as obesity, type 2 diabetes, age- related macular degeneration (AMD), and Alzheimer’s disease (AD) (Reinisalo et al. 2015). It is suggested that other stilbenoid compounds, such as pinosylvin, piceatannol, or pterostilbene, may have higher biological activity than resveratrol and deserve to receive more scientific attention. The antioxidant activity of stilbenoids of this study are reviewed in Table 2.3.

21 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

Table 2.3. Antioxidant activity of stilbenoids and extracts rich in stilbenoids studied

Antioxidative indication Compound/Extract Reference

Inhibition of chemiluminescence

Pinosylvin Perecko et al. 2008

Inhibition of lipid peroxidation

Astringin Harlamow 2007

Isorhapontigenin Wang et al. 2001

Isorhapontin Harlamow 2007

Norway spruce bark extract Pietarinen et al. 2006

Piceatannol Waffo-Teguo et al. 1998

Piceatannol Harlamow 2007

Scots pine knot extract Pietarinen et al. 2006

Inhibition of oxidative DNA damage

Isorhapontigenin Wang et al. 2001

Scavenging of peroxyl radicals

Astringin Harlamow 2007

Isorhapontigenin Wang et al. 2001

Isorhapontigenin Harlamow 2007

Isorhapontin Harlamow 2007

Norway spruce bark extract Pietarinen et al. 2006

Norway spruce bark extract Harlamow 2007

Piceatannol Waffo-Teguo et al. 1998

Piceatannol Hung et al. 2001

Piceatannol Zheng and Ramirez 2000

Piceatannol Harlamow 2007

Scots pine knot extract Pietarinen et al. 2006

2.1.5 MEDICINAL PROPERTIES OF STILBENOIDS Stilbenoids, especially resveratrol, have shown functional properties that could be utilized in medicinal products. These compounds have potential to be used as pure compounds or as lead molecules for further chemical synthesis of even more active and bioavailable stilbenoid derivatives. Resveratrol is the most studied naturally-occurring stilbenoid. Resveratrol and its synthetic derivatives possess a variety of biological activities, such as anticancer, antioxidant and cardiovascular protection, anti-inflammatory, antiviral, chemopreventive and neuroprotective effect, and immunomodulation that can be applied for medicinal products (Jang et al. 1997, Roupe et al. 2006, Rimando and Suh 2008, Liu et al. 2015, Bhullar and Hubbard 2015, Reinisalo et al. 2015, Shahpiri et al. 2016). The antioxidant activity of stilbenoids in combination with their photoprotective properties against UV radiation may offer new pharma-photoprotective formulations against the harmful

development of UV-induced injuries (Regev-Shoshani et al. 2003, Stevanato et al. 2014). Resveratrol is the principal stilbenoid found in wine and for that reason resveratrol is implied in health benefits of wine such as decreased platelet aggregation, promoting vasorelaxation, suppressing atherosclerosis and reducing lipid peroxidation (Renaud and de Lorgeril 1992, Seigneur et al. 1990, Fitzpatrick et al. 1993, Wang et al. 2005, Fuhrman et al. 1995). Resveratrol has been considered as a potential phytoestrogen compound but its bioavailability is one the main concerns related to its efficacy in humans (Giuliani et al. 2016, Hu et al. 2016, Landete et al. 2016). However, the efficacy of intake of stilbenoids (especially resveratrol) has shown to reduce the risk of certain health diseases, such as diabetes (Tresserra-Rimbau et al. 2016). Based on encouraging therapeutic evidence, trans-resveratrol research has fueled a great deal of interest in characterizing structurally similar stilbenoid compounds. Stilbenoids structurally related to trans-resveratrol, found in a variety of foods, medicinal plants and trees, have shown a multitude of biological activities comparable to trans-resveratrol (Rimando and Suh 2008, Waffo-Teguo et al. 1998). The interesting medicinal activities of other stilbenoids studied in this thesis are reviewed in Table 2.4. As a conclusion, the efficacy of stilbenoids is based on their ability to fight against oxidative stress inducing different signaling pathways in the cells.

2.1.6 ANTIMICROBIAL AND PROTECTIVE PROPERTIES OF STILBENOIDS According to World Health Organization (WHO), antimicrobial resistance (AMR) threatens the effective prevention and treatment of an ever-increasing range of infections caused by bacteria, parasites, viruses and fungi. It happens when such microorganisms change on exposure to antimicrobial drugs (such as antibiotics, antifungals, antivirals, antimalarials, and anthelmintics). Due to this serious health problem, there is a need to find new ways to treat various diseases and also replace the current antibiotics. Stilbenoids derived from conifer trees have shown protective properties that have functions against variety of microorganism. The molecular structure of stilbenoids, such as certain hydroxyl group positions, is suggested to be responsible for these activities (Sekine et al. 2009). Stilbenoids have shown potential as natural antibacterial, antifungal and antitermite agents. These properties are reviewed in Table 2.5.

23 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

Table 2.4.Medicinal activity of stilbenoids and extracts rich in stilbenoids studied

Medicinal indication Compound/Extract Study model Reference Age-related macular degeneration Pinosylvin In vitro Koskela et al. 2014 Anticancer Piceatannol In vivo Kimura et al. 2000 Piceatannol In vivo Waffo-Téguo et al. 2001 Astringin In vivo Waffo-Téguo et al. 2001 Piceatannol In vitro Kim et al. 2009 Piceatannol In vitro Wolter et al. 2002 Piceatannol In vitro Wesolowska et al. 2010 Pinosylvin In vitro Park et al. 2012 Pinosylvin In vivo Park et al. 2012 Pinosylvin In vitro Simard et al. 2008 monomethyl ether Pine knot extract In vivo Yatkin et al. 2014 Anti-inflammatory (Arthritis) Pinosylvin In vivo Macickova et al. 2010 Pinosylvin In vivo Jancinova et al. 2012 Pinosylvin In vivo Laavola et al. 2015 Pinosylvin In vivo Laavola et al. 2015 monomethyl ether Cardioprotection Piceatannol In vitro Hung et al. 2001 Piceatannol In vitro Shi et al. 2008 Isorhapontigenin In vitro Li et al. 2005 Immune system Isorhapontigenin In vivo Nan Fang and Tao Liu 2002

Table 2.5. Antimicrobial activity of stilbenoids and extracts rich in the stilbenoids studied

Compound/ Study organisms Reference Extract

ANTIBACTERIAL Pine knot extract Burkholderia multivorans, Bacillus coagulans Lindberg et al. 2004 and Alcaligenes xylosoxydans Pine knot extract, Bacillus cereus, Staphylococcus aureus, Välimaa et al. 2007 pinosylvin, pinosylvin Listeria monocytogenes, Lactobacillus monomethyl ether plantarum, Escherichia coli, Salmonella infantis, Pseudomonas fluorescens, Candida albicans, Saccharomyces cerevisiae, Aspergillus fumigatus and Penicillium brevicompactum Astringin, Bacillus cereus, Listeria monocytogenes, Harlamow 2007 isorhapontin, Staphylococcus aureus, Escherichia coli, piceatannol, Pseudomonas fluorescens, Salmonella infantis isorhapontigenin, and Saccharomyces cerevisiae resveratrol, pinosylvin, pinosylvin monomethyl ether ANTIFUNGAL Pinosylvin, Pinosylvin Polyporaceae, Telephoraceae and Agaricaceae Hart, 1981 monomethyl ether Pinosylvin, Pinosylvin Ascomycetes and Fungi Imperfecti Hart, 1981 monomethyl ether Pine heartwood extract Coniophora puteana, Gloeophyllum trabeum, Lu et al. 2016 and Rhodonia (Poria) placenta Pinosylvin 16 fungal species Seppänen et al. 2004 Pinosylvin, pinosylvin Trametes versicolor, Phanerochaete Celimene et al. monomethyl ether, chrysosporium, Neolentinus lepideus, 1999 pinosylvin dimethyl Gloeophyllum trabeum and Postia placenta ether Astringin, isorhapontin, Phaelous schweinitzii, Sparassis crispa, Woodward and isorhapontigenin Pearce 1988 Isorhapontin Formitopsis pinicola, Phaelous schweinitzii Beyer et al. 1993 Spruce bark extract Ceratocystis polonica, Formitopsis pinicola. Alfredsen et al. 2008 Spruce bark extract Ceratocystis polonica, Ophiostoma penicillatum Evensen et al. 2000 ANTITERMITE Isorhapontin, astringin, Reticulitermes speratus (Kolbe) Shibutani et al. piceid, isorhapontigenin 2004

25 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

2.2 STABILITY OF STILBENOIDS

Stilbenoids exist in the trans and cis stereoisomeric forms (Figure 2.5), but naturally they occur overwhelmingly in the trans isomer due to that being thermodynamically more stable (and energetically more favorable) than the cis isomer (Likhtenshtein 2010). However, trans stilbenes may be isomerized to the cis form under various conditions such as UV radiation (light energy is absorbed by the stilbene molecule and the isomerization takes place). There are skepticism and also conflicting findings about the stability and therefore future use of stilbenoids (Shi et al. 2008). According to Piñeiro et al. (2006), the utilization of the beneficial effects of resveratrol is limited because it is an easily oxidizable and extremely photosensitive compound. In contrast, Prokop et al. (2006) showed that piceid and its aglycone, resveratrol, are stable polyphenols. Trans-resveratrol converted to cis-resveratrol in solution exposed to daylight (Kolouchová-Hanzlíková et al. 2004). The same photoisomerization reaction takes place when the solution is exposed to UV light (Trela and Waterhouse 1996). The authors did not find any other product than cis-resveratrol when trans-resveratrol was exposed to UV light at 366 nm. Montsko et al. (2008) used HPLC-MS and -MS/MS together with UV spectroscopy and semiempirical quantum chemical calculations to determine the UV isomerization reaction of trans-resveratrol solutions under 365 nm UV radiation.

Fig. 2.5. Trans (left) and cis (right) isomers of stilbenes.

Besides the formation of cis-resveratrol, they claimed the formation of an oxidized derivative of trans-resveratrol with a triple bond at the centre of the molecule or the formation of a phenanthrene structure. The formation of trihydroxyphenanthrene was suggested by Tříska et al. (2012). Yang et al. (2012) showed that UV radiation of trans-resveratrol leads to the formation of a highly fluorescent compound, resveratrone, which was properly characterized by UV, MS and NMR. Excluding the last reference, the characterizations of possible new compounds derived from trans stilbenoids have been mainly tentative based only on UV or MS spectra without utilizing more detailed structural information, such as NMR spectra. In summary, the findings about the chemical stability of trans-resveratrol and trans-piceid in a solution are not in agreement with each other. Although the cis isomer was formed in every case when trans stilbenoid was exposed to UV or fluorescent light, the formation of additional oxidized product(s) was

not observed by all investigators. A clear conclusion about the stability of trans-piceid and trans-resveratrol in solution cannot be made.

2.3 COMMERCIAL UTILISATION OF POTENTIAL BIOACTIVE COMPOUNDS

Wood-derived polyphenolics, such as stilbenoids and lignans, have various functions in plants: defence against UV light, oxidative stress, and a variety of pathogens. The potential functional properties of stilbenoids presented in earlier chapters describe the possibilities of using these compounds in commercial products. Commercial applications based on the antioxidative properties include orally administered dietary supplements, health-promoting (functional) foods and pharmaceuticals. Resveratrol is the most sold stilbenoid compound in the nutraceutical market. Also the lignan hydroxymatairesinol is offered as a commercial food supplement product. Antimicrobial properties of stilbenoids in combination with their antioxidant properties could be utilized in cosmetic products. Protection against harmful microorganisms and/or UV radiation may also be a useful feature in the field of protecting different products, materials and surfaces. Stilbenoid derivatives have already been used in electronics industries (Likhtenstein 2010). Nowadays, resveratrol is the major stilbenoid compound sold commercially. According to Nutraingredients (Frost and Sullivan report), the global resveratrol supply market was valued at 40 million euros in 2012 with 90 % of volume accounted for by the US supplements market. It is expected by Front Research that the global resveratrol markets will grow by 4.5 % annually by 2021 the Asia-Pacific region being the fastest growing area. However, it should be noted that these resveratrol products are derived from other sources than woody biomass, for example from grapewine (skin) biomass. The lignan hydroxymatairesinol (HMR) derived from woody biomass (Norway spruce knots) is one of the few examples sold commercially. This lignan product is sold in supplement markets as a natural antioxidant against several diseases. The main indication areas of HMR are women’s menopause and hot flashes. The ingredient product was first studied and developed in Finland. To develop commercial products from the extracts based on natural raw materials, such as forest biomass and industrials side streams, it is important to review the potential products from at least two main perspectives:

1. Regulatory perspective: Does the raw material used (biomass) have already existing regulatory status? Is the product already in use or is it novel? Novel products may need to go through a deeper inspection and evaluation process to get registered and approved. It should be noted that different markets have their regional legislation: for example, the

27 Bioactive polyphenolics of roots and stumps of Norway spruce and Scots pine

European food markets are controlled by EFSA, US markets by FDA and Asian markets by country-specific authorities. Food markets are also divided usually to food products and supplement product registration (pharmaceutical products have also their own registration path). The cosmetics products may be regulated by different legislation. The regulation for natural products refined from plants (biomass) are “plant part”-specific. For example, the use of Norway spruce young shoots is not considered as novel but the use of Norway spruce bark would be novel by EFSA. Issues related to legislation and registration of new compounds and products thereof need to be thought carefully (Singh and Singh 2012).

2. Efficacy and safety perspective: The products developed need to have defined efficacy and safety documentation and also technical documentation. This kind of documentation may not be necessary (if the regulatory status is easy and allows proceeding immediately) but in the very competitive and international market of natural products, these documents are often required to attain commercial interest and actual sales.

3 EXPERIMENTAL

3.1 SAMPLE MATERIAL

All the root and stump samples were collected one day after the trees were felled and stored in -20°C prior to any further handling (I, II, III, IV). The samples from different parts of the collected biomass (such as bark and wood material) were visually inspected and separated by knife. All the samples prepared were then freeze-dried and powdered by a mill prior to extraction. The dried and powdered samples were stored in -20°C.

Root and stump samples The samples of Norway spruce and Scots pine were collected from drained peatland forests in Parkano, Western Finland (I). Another samples of Norway spruce were collected from a mineral soil forest in Parkano, Western Finland. Matured trees of different ages and sizes were included in sampling from both sites. For the analysis of nutrient content of Norway spruce, samples of the current year needles were collected for the nutrient analyses from the uppermost branch of the sample trees. The roots and stumps from each tree were divided into four different zones S, A, B, and C (Figure 3.1). The S zone was "the stump sample" above the root neck. The A zone was the thickest cross-section below-ground root part adjacent to the stem. The B zone was the next part of the root, which were a bit further out with a smaller cross-sectional diameter than in zone A. The C zone represented the outermost part of the roots with a diameter less than 2 cm.

Fig. 3.1. Location of sample zones in stump (S) and roots (A, B and C).

29 Experimental

Root neck samples Location of the root neck area is between the thick roots (“zone A”) and the stump (“zone S”) (Figure 3.2a). Due to its role by connecting the root system and the tree stem, the root neck (RN) may have an interesting extractive composition. The RN samples of Norway spruce were collected from a paludified mineral forest in Parkano, Western Finland (II). The root neck samples were divided into different zones (Figure 3.2b-f & Table 3.2). The wood samples were taken from both the heartwood (hW) (3 cm from wood pith) and the sapwood (sW) region (3 x 3 cm pieces). Samples of the inner and outer parts of the bark were studied separately. Macroscopic bark samples were visually separated into: 1) the inner bark (iB) sample: comprising conducting phloem and inner bark layers with stone cells and 2) the outer bark (oB) sample: comprising the layers of periderm and old, dead phloem (Figure 3.2f).

Fig. 3.2. Scheme of a sampling process of the root neck (RN) of Norway spruce. a) The sample zones of our earlier analyses of roots and stumps of Norway spruce (Latva-Mäenpää et al. 2013). b) A stump of Norway spruce (Tree 3). c) RN samples were taken from the stump of Norway spruce. d) The RN samples were further divided into five different zones (e.g. zone 1 is from the bottom of the RN. In bark analyses an extra sample was taken underneath the RN denoted 0). e) One zone from the root neck was divided into heartwood and sapwood. Bark was also divided to inner (iB) and outer (oB) parts. f) A microscope image of the bark of Norway spruce (Outer bark: the layers of periderm and old, dead phloem, Inner bark: conducting phloem + inner bark with stone cells)

3.2 EXTRACTION

Soxhlet extraction with acetone The total amounts of acetone soluble extracts of Norway spruce and Scots pine roots and stump samples from peatland forest site were measured by weighing the dried residues after the Soxhlet extraction (I). A sample of 2 g of a freeze- dried wood and bark from each sampling zone was weighed and extracted for 6 h in a Soxhlet apparatus with 60 mL of acetone. The extracts were evaporated to dryness by a rotary evaporator. Dried extracts were then weighed to calculate the extract yield.

Acetone extraction by sonication Freeze dried and finely ground wood and bark samples or Norway spruce and Scots pine from different sampling zones were prepared and extracted by sonication according to Papers I (for root and stumps samples) and II (for root neck samples). 20 mg of samples were extracted by 2 mL of pure acetone (I) or acetone-water mixture (95:5, V/V) (II) in an ultrasonic water bath for 30 minutes and 1 mL of the supernatant was dried under N2 stream.

Accelerated solvent extraction (ASE) Ethanol-water extract of the bark of Norway spruce roots was obtained by sequential extraction of the powdered bark (2 g) by using accelerated solvent extraction (ASE) (Dionex) (III and IV). The bark sample was extracted first with n-hexane to remove the hydrophobic constituents, followed by ethanol- water mixture (95:5, V/V) to collect the hydrophilic stilbenoid constituents. The extraction was followed by further concentration steps to yield the stilbenoid enriched fraction described in Chapter 3.4.

Pressurized hot water extraction (PHWE) A root neck sample of Norway spruce was subjected to the pressurized hot water extraction (PHWE) method (II). A 1.65 g sample of the heartwood of Norway spruce root neck was placed into the extraction chamber inside an oven. The oven was heated to 140°C and the water flow-rate of 4 mL min-1 was used for the 30 min extraction. A description of a similar method has been given by Leppänen et al. (2011) for the extraction of wood hemicelluloses. We evaluated its applicability for the extraction of polyphenolic compounds.

3.3 ANALYSIS METHODS

Gas chromatography (GC-MS) The dried extract samples were silylated with N-trimethylsilyl imidazole (TMSI in pyridine) and analysed by GC-MS (gas chromatography Agilent Hewlett Packard 6890 and mass spectrometer Hewlett Packard 5890 MSD)

31 Experimental

with a DB-5MS capillary column (30 m × 0.25 mm i.d. × 0.25 μm) (I and II). Compounds were identified by their mass spectra with mass spectral libraries. Quantitative analyses of identified compounds were performed by comparison with the internal standards (heptadecanoic acid, heneicosanoic acid and betulinol) corrected with external standards, such as stilbenoid glucoside piceid (The correction factor of 0.79 was used for the quantification of stilbenoid glucosides with heptadecanoic acid). All results in Chapter 4, given in mg g–1 or % (w w–1), were calculated on a freeze-dried wood basis.

Liquid chromatography (HPLC-UV and HPLC-MS) Separation of stilbenoids from the extract of the root bark of Norway spruce was carried out by HPLC system (III and IV). HPLC separation was performed by Hewlett Packard model 1100 series and Schimadzu Prominencence HPLC intruments on a C18 column (4.6 mm × 150 mm, 5 μm) using a mobile phase of H2O and MeOH. The UV profiles of eluates were monitored with DAD in the range of 200-400 nm. Detection was carried out at 286 nm and 306 nm for cis and trans stilbenoids, respectively (260 nm was also used for detection in UV experiments). The mobile phase (H2O and MeOH) gradient program was further modified for the structural characterisation of the compounds formed after UV radiation. For the HPLC-DAD-ESI-MS and HPLC-DAD-ESI-MS/MS analyses, an Agilent 1100 Series liquid chromatography system coupled to a Bruker Esquire 3000plus ion trap mass spectrometer were used. The detailed methods are described in paper III. The HPLC conditions were the same as those given for the analytical HPLC-DAD analysis. DAD was monitored at 280, 286, and 306 nm and the negative ion mode with scan range of m/z = 100-1000 was used to obtain MS spectra.

Liquid chromatography combined with nuclear magnetic resonance spectroscopy (HPLC-DAD-NMR) The HPLC-DAD-NMR experiments were carried by the methods described in Paper III. The stilbenoids were separated from the extracts on C18 (4.6 mm × 150 mm, 5 μm) column by Agilent 1200 LC equipment and peaks detected were trapped in separate storage loops. The multiple injections of the extracts were performed to obtain enough material for further studies. The compounds fractionated were then analysed separately by NMR (at 11.75 T using a Bruker DRX 500 NMR Spectrometer) after evaporation of the solvent. Each of the dried eluate was dissolved in 40 μL of DMSO-d6 and were transferred into 1.7 mm NMR tubes for 1D and 2D NMR measurements. All NMR spectra were acquired at 27 °C. 1D 1H NMR spectra with 50122 data points were recorded with water solvent peak suppression. 2D NMR spectra were recorded using standard pulse programs (COSY-45, NOESY, TOCSY, HSQC and HMBC). 1H and 13C chemical shifts were referenced to solvent signals of DMSO-d6, δ (1H) = 2.5 ppm and δ (13C) = 39.52 ppm.

Mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (NMR) The mass spectrometry analyses were conducted for stilbenoid glucosides by high resolution ESI-TOF-MS (Bruker Daltonics micro-TOF-Q I) with negative ion mode and for stilbenoid aglucones by EI-MS (MS Jeol JMS SX-2 high resolution sector instrument) (IV). Acetonitrile-water was used as the solvent. The NMR analyses were performed by 500 MHz and 300 MHz spectrometers. The 1H and 13C spectra were recorded in CD3OD (300 MHz, 500MHz). 1H and 13C chemical shifts were referenced to solvent signals of CD3OD, δ (1H) = 3.31 ppm and δ (13C) = 49.86 ppm. The following methods were used to analyse the products formed in UV irradiation studies presented in the Chapter 3.5 concerning the experiments of the UV stability.

3.4 ISOLATION OF STILBENOIDS AFTER EXTRACTION

Prepurification and enrichment of stilbenoids The prepurification of the bark extract was performed in order to remove soluble sugars from the extracts (III and IV). The non-hydrolysed bark extract containing stilbenoid glucosides was first evaporated to dryness by rotavapor in order to remove the ethanol from the extract. The dried extract was then dissolved in water and then enriched by polymer resin (XAD-7HP) filled column, which was previously conditioned with water. Polyphenolic compounds, such as stilbenoids, were absorbed to a resin material and water soluble compounds such as sugars (not absorbed to the resin) were washed away by rinsing the column with water. Stilbenoid glucosides were collected from ethanol-water eluates.

Hydrolysis of stilbenoid glucosides The bark extract rich in stilbenoid glucosides (astringin, isorhapontin and piceid) was hydrolysed to obtain stilbenoid aglucones (piceatannol, isorhapontigenin and resveratrol) by a specific enzyme (β-glucosidase), which breaks the glucosidic bond of the molecules (III and IV). The dried bark extract was dissolved in ammonium acetate buffer solution (0.1 mol/L, pH=5), β- glucosidase enzyme solution (3mg/mL) was suspended in the above solution and the mixture was incubated in a laboratory oven at 39°C for 3 days. The incubated solution was then directly transferred onto a column filled with polymer resin XAD-7HP and stilbenoids were collected after washing with water from the ethyl acetate eluates.

Semi-preparative isolation of stilbenoids by HPLC Preparative isolation of bark stilbenoids was performed by a Waters 600 HPLC instrument on a Waters Xbridge-C18 column (19 × 150 mm, 5 μm) using a mobile phase of water and acetonitrile with a flow rate of 7 mL/min (IV). The

33 Experimental

mobile phase gradient programs for non-hydrolysed and hydrolysed samples were 90:10 (t) 0 min, 85:15 (t) 5 min, 70:30 (t) 25 min, 50:50 (t) 30 min, 10:90 (t) 35 min, 90:10 (t) 37-40 min and 90:10 (t) 0 min, 85:15 (t) 10 min, 70:30 (t) 35 min, 50:50 (t) 38 min, 10:90 (t) 40 min, 90:10 (t) 43-45 min, respectively. The method developed was later transferred also to the Shimadzu Prominence HPLC system. The fractions of the hydrolyzed extract and the non-hydrolysed extract were collected at 7.5, 11.2, 12.2 min and 13.3, 18.3, 19.5 min. All the collected fractions (7 mL of each) were evaporated to dryness by Biotage evaporator and dissolved in 10 mL of MeOH. The dissolved stilbenoids were filtered through a 0.45 μm microporous membrane and stored in freezer at -20 °C until analysis. The chemical structures of the stilbenoids isolated were determined by NMR and MS. The mobile phase (H2O and MeOH) gradient program was further modified for the isolation of stilbenoids for the structural characterization of the compounds formed under UV irradiation and this fractionation was performed by Shimadzu Prominence HPLC instrument; the modified mobile phase gradient program was 72:28 (t) 0 min, 60:40 (t) 9 min, 50:50 (t) 13.5 min, 30:70 (t) 15 min, 10:90 (t) 17.5 min and 72:28 (t) 18-21.5 min with a flow rate of 14 mL/min. The fractions were dried with vacuum centrifuge and identified by UV, MS and NMR.

3.5 STABILITY OF STILBENOIDS

Storage conditions Stock solutions of reference trans-resveratrol (290 μg/mL), trans-piceatannol (380 μg/mL) and trans-piceid (270 μg/mL) and the isolated stilbenoids produced from our own experiments (trans-astringin (120 μg/mL), trans- piceid (40 μg/mL), trans-isorhapontin (320 μg/mL), trans-piceatannol (90 μg/mL), trans-resveratrol (14 μg/mL) and trans-isorhapontigenin (190 μg/mL).) were prepared in MeOH. Also a crude extract (produced from the bark of Norway spruce roots) prior to concentration on a column packed with XAD-7HP was directly taken for stability study. The stability of the samples was tested under conditions described in Table 3.3. Aliquots of each sample were taken after 0 hr (control sample), 24 hrs, 1 week and 2 weeks for HPLC-DAD, HPLC-DAD/ESI-MS and HPLC-DAD/ESI- MS/MS.

Table 3.3. Conditions in the stability studies of stilbenoids

Stability conditions in solution Stability conditions in solid form form

(1) Light protected: solutions of stilbenoids (1) Light protected: solid extracts were transferred were transferred into a glass vial and capped into a glass vial and capped under ambient air under ambient air conditions. The samples conditions. The samples were stored in freezer (- were stored in freezer (-20°C) with protection 20°C) with protection from light. from light.

(2) Light unprotected: solutions of stilbenoids (2) Light unprotected capped glass vial: solid were transferred into a glass vial and capped extracts were transferred into a glass vial and capped under ambient air conditions. The samples under ambient air conditions. The vial was placed placed on top of a laboratory bench were on top of a laboratory bench and exposed to exposed to continuous fluorescent light by a continuous fluorescent light by a general laboratory general laboratory lamp. lamp.

(3) Light unprotected uncapped glass vial: solid extracts were transferred into a glass vial and left uncapped to increase the contact of air with the sample. The vial was placed on top of a laboratory bench and exposed to continuous fluorescent light by a general laboratory lamp.

Photostability of stilbenoids under UV radiation The stability of the isolated stilbenoids (in methanol solution) under ultraviolet (UV) radiation was tested by using UV lamp at 366 nm wavelength (UV-A radiation). The stilbenoids were placed in vials on the work surface of the UV lamp system and samples were taken at intervals of 0 h (control sample), 10 min, 30 min, 1h, 2h, 5h and 24h for HPLC-DAD analysis. Also GC- MS (presented in Chapter 3.3) was applied to analyse samples at 0 h, 2h and 24h. Spruce bark-derived stilbenoids were isolated by semiprep-HPLC for the structural elucidation of new compounds formed during UV radiation of the samples. The stilbenoids isolated were UV irradiated for 6 days and the compounds formed were isolated by prep-HPLC, dried with vacuum centrifuge and identified by MS and NMR.

35 Results and discussion

4 RESULTS AND DISCUSSION

The extractive profiles of Norway spruce and Scots pine roots and stumps are presented in Chapter 4.1. Also the differences in stilbenoid composition of the roots and stumps of Norway spruce bark between the peatland and mineral soil forests and localization between inner and outer bark are presented. Finally, the amounts of stilbenoids (stilbenoid glucosides of Norway spruce bark and pinosylvin compounds in Scots pine) in different tree parts are compared. Chapter 4.2. describes the analysis of lignans from the root neck of Norway spruce and also presents a green water extraction method suitable for polyphenolic compounds from forest biomass. Extraction, isolation and purification methods for stilbenoids derived from the bark of Norway spruce roots are described in Chapter 4.3. More detailed structural elucidation of these stilbenoid compounds is presented in Chapter 4.4. The stability and photochemical reactions including fractionation and analysis of new compounds formed is the topic of Chapter 4.5.

4.1 GENERAL EXTRACTIVE PROFILES OF THE ROOT AND STUMP BIOMASS OF CONIFER TREES

Roots and and stumps of Norway spruce and Scots pine The total amount of acetone soluble extracts of Norway spruce and Scots pine roots and stumps of the peatland forest site were analysed (I). The largest amounts of extracts in Norway spruce bark were found in the zone A closest to the stem and in wood samples from the zone C closest to the root tip (Figure 4.1 and Table 4.1). For Scots pine, the largest amount of extracts in bark was found from the zones B and C and in wood samples from the stump (zone S).

Fig. 4.1. Cross-sections of root zones from right to left A, B and C in Norway spruce (left) and Scots pine (right) show the eccentric growth of roots (The scale bars in the figures shows illustrates the length of 5 cm).

Table 4.1. The total amount of acetone soluble extracts per unit of dry weight (dry wt) of the wood and bark zones of Norway spruce and Scots pine roots and stumps

%-.extractives per dry weight %-.extractives per dry weigth Sampling zone (Norway spruce) (Scots pine)

Wood Bark Wood Bark

Stump (zone S) 1,93 ± 0,42 13,38 ± 2,94 6,56 ± 3,52 10,90 ± 4,22

Root zone A 2,09 ± 0,95 18,18 ± 2,72 2,32 ± 0,62 14,65 ± 4,97

Root zone B 1,67 ± 0,31 13,44 ± 3,92 3,17 ± 0,67 16,83 ± 4,22

Root zone C 3,46 ± 1,17 16,40 ± 4,21 7,13 ± 2,00 16,19 ± 3,17

Compounds identified from Norway spruce The wood of Norway spruce roots and stumps contained mainly mono- and oligosaccharides such as fructose, glucose and sucrose (Figure 4.2) (I). Fatty acids and resin acids (pimaric acid, abietic acid and dehydroabietic acid) were identified. Interestingly, the lignan hydroxymatairesinol (HMR) was found in zones A and B of some wood samples.

Fig. 4.2. GC-MS chromatograms of extracts obtained from Norway spruce roots (I).

37 Results and discussion

The bark of Norway spruce roots and stumps were found to contain stilbenoid glucosides and also free stilbenoids. Astringin (3-O-β-D-glucosyl- 3',4',5-trihydroxystilbenoid) and isorhapontin (3-O-β-D-glucosyl-4',5- dihydroxy-3'-methoxystilbene 3-O-β-D-glucoside) were the major compounds and piceid, piceatannol (aglycone of astringin) and isorhapontigenin (aglycone of isorhapontin) were the minor stilbenoids (Figure 4.3). Also catechin and β- sitosterol were identified as minor compounds.

Fig. 4.3. Chemical structures of stilbenoid compounds. Compounds were identified and analysed by GC-MS after silylation with TMSI. (R1=R2=OH: trans-piceatannol, R1=OH, R2=H: trans- resveratrol, R1=OH, R2=OCH3: trans-isorhapontigenin, R1=OGlu, R2=OH: trans-astringin, R1=OGlu, R2=H: trans-piceid, R1=OGlu, R2=OCH3: trans-isorhapontin, *OGlu=3-O-β-D-glucoside).

Compounds identified from Scots pine The wood samples of Scots pine roots and stumps contained sugars and fatty and resin acids. The stump extract of pine wood contained also stilbenoids, pinosylvin (3,5-dihydroxystilbene) and pinosylvin monomethyl ether (3- hydroxy-5-methoxystilbene) but stilbenoids were not found from root zones (only minor amounts in root zone A). Catechin was detected from the extracts of pine bark.

Stilbenoid variation of Norway spruce bark obtained from peatland and mineral soil forest sites Stilbenoid glucoside concentrations in Norway spruce bark varied between 0.5 and 8.3 % in different zones of the roots and stumps (I). In both soil-type sites the highest concentrations of stilbenoid glucosides were found in zone A nearest to the stem. Isorhapontin was the major stilbenoid compound in root zones, whereas astringin was in stump samples. The concentration of astringin and isorhapontin decreased when the diameter of roots decreased in the bark samples from the mineral soil forest site. The same phenomenon was not observed in the peatland forest site. The concentrations of the stilbenoid aglycones isorhapontigenin and piceatannol and also of catechin were the largest in zone A. The highest concentration of isorhapontin (5.8%) was found in the roots of the bark of trees grown on mineral soil. The highest concentration of astringin (1.6%) was found in the thick roots of the bark of trees grown on peatland. However, no such a clear difference in astringin concentrations was observed than with isorhapontin between the forest sites. Large variations in the

concentration of resin acids were observed in the bark of stumps. Fatty acid and resin acid contents were relatively low in wood samples taken from both sites. Total contents of resin acids in our study (I) were comparable with those reported in the study of Hovelstadt et al. (2006).

Table 4.2. The mean concentrations (mg g-1. dry wt) of resin acids and stilbenoid glucosides in the bark of stumps (s) and in different root zones (A, B and C) of Norway spruce obtained from the peatland and mineral soil forest sites (I).

Extractives in bark of Norway spruce roots and stumps, mg g-1. dry wt (mean ± standard error of the mean)

PEATLAND S A B C (April 2009. n=5) Resin acids 7.71 ± 1.41 3.07 ± 1.42 2.40 ± 0.74 5.45 ± 1.33 Piceid 0.32 ± 0.07 1.04 ± 0.29 0.41 ± 0.20 0.58 ± 0.29 Astringin 7.01 ± 0.78 12.25 ± 4.16 4.36 ± 1.92 4.65 ± 2.86 Isorhapontin 4.32 ± 2.86 17.85 ± 4.08 10.77 ± 2.79 19.83 ± 6.29

MINERAL SOIL S A B C (June 2010. n=4) Resin acids 19.96 ± 6.47 4.41 ± 0.78 5.04 ± 1.69 4.57 ± 1.05 Piceid 0.46 ± 0.07 1.33 ± 0.38 1.19 ± 0.27 0.74 ± 0.24 Astringin 6.72 ± 1.88 11.41 ± 3.75 9.49 ± 3.10 2.57 ± 1.27 Isorhapontin 4.56 ± 1.33 45.55 ± 12.14 42.39 ± 8.94 22.20 ± 4.31

Isorhapontin concentrations between different zones (A, B, C and S) and between trees from both sites (peatland and mineral) showed larger variation between different root zones than the tree-to-tree variation (I). The mean isorhapontin concentration of tree 3 (9.99 mg g-1) of the mineral soil forest site was lower than that found for the other trees (32.16–36.67 mg g-1) from the same site. Tree 3 had shorter roots compared to the other trees. For astringin, the same phenomenon was not observed so clearly. Large variation in concentrations of stilbenoid glucosides may be due to the differences between the physiological stages (root age) of sample zones. Seasonal variation of stilbenoid glucosides may partially explain the varying concentrations in the samples from peatland and mineral soil forest sites in this study. This has also been suggested by Solhaug et al. 1990 and Underwood and Pearce 1991. Correlations between different nutrient contents and stilbenoid glucoside concentrations showed some correlation between zinc (Zn) and isorhapontin (r=0.72, p<0.05) (Table 4.3).

39 Results and discussion

Table 4.3. Mean nutrients content of the needles of Norway spruce.

Nutrients, dry wt (mean ± standard error of the mean)

PEATLAND, April 2009, n=5) MINERAL SOIL, June 2010, n=4 (optimum (optimum Analysed value Analysed value value*) value*) N mg g-1 14.0 ± 1.00 (12.5-19.1) N mg g-1 12.2 ± 0.20 (13.0-15.0) K mg g-1 3.43 ± 0.37 (>6.20) K mg g-1 3.53 ± 0.69 (4.5-6.0) P mg g-1 1.97 ± 0.37 (2.3-3.5) P mg g-1 1.49 ± 0.17 (1.2-1.5) Ca mg g-1 3.78 ± 0.20 n.a Ca mg g-1 5.64 ± 0.36 (2.0-3.0) B mg kg-1 11.8 ± 1.00 (10-30) B mg kg-1 13.3 ± 1.58 (8.0-25) S mg kg-1 882 ± 33.0 (900-1200) S mg kg-1 839 ± 59.5 (>900) Mg mg kg-1 1.35 ± 0.06 (1.2-2.0) Mg mg kg-1 1.55 ± 0.09 (0.8-1.0) Cu mg kg-1 1.88 ± 0.10 (2.0-5.0) Cu mg kg-1 1.72 ± 0.07 (>3.0) Zn mg kg-1 22.7 ± 3.69 (16-80) Zn mg kg-1 36.5 ± 2.19 (>25) Mn mg kg-1 350 ± 26,2 (200-2000) Mn mg kg-1 877 ± 126 (80-500)

Isorhapontin concentrations (up to 5.8%) from the samples of the mineral soil forest site in our study (I) were larger than those obtained by Lindberg et al. (1992) and Beyer et al. (1993). However, the results from their studies were presented per fresh weight (FW) of the bark, where as our results are per dry weight (DW). Balas and Popa (2007) used an air drying method in preparation prior to their analysis of wood extract and reported lower stilbenoid glucoside concentrations, which may be caused by their pre-handling method.

Distribution of stilbenoids between inner and outer bark of Norway spruce The major stilbenoid glucosides (isorhapontin and astringin) were found to be mainly localized in the inner bark (iB), which contains the conducting parenchyma and stone cells, whereas the outer bark contains the layer of periderm and old, dead phloem (II). The average concentrations of isorhapontin and astringin in the inner bark were 9.8% and 5.4%, respectively. The concentrations were times 30-200 times higher in the inner bark compared to the outer bark (Figure 4.4). Also other compounds, such as sucrose, piceid, catechin, piceatannol, isorhapontigenin and dehydroabietic acid were quantified from inner bark.

Fig 4.4. GC-MS chromatograms of the extracts of the inner and outer parts of the bark of Norway spruce (above=inner/below=outer part of the bark)

Stilbenoid concentrations in Norway spruce The concentrations of stilbenoid glucosides (astringin, isorhapontin and piceid) in various parts of Norway spruce bark are presented in Tables 4.4a (for roots) and 4.4b (for stump, stem and needles). The results of these studies conclude that the stilbenoid glucoside concentrations are higher in roots than in stem and stump. Isorhapontin was the most abundant stilbenoid in root bark (especially in mineral soilds samples), whereas in stem bark samples astringin and isorhapontin concentration were closer to each other. Only small amounts of stilbenoids were found from the needles (0.06%). It can be concluded that the root bark of Norway spruce is a promising source of bioactive stilbenoids.

41 Results and discussion

Table 4.4a. The concentrationof stilbenoids (isorhapontin, astringin, piceid) found in roots of Norway spruce (DW; dry weight, FW; fresh weight).

Sampling Stilbenoid Amount Sampling area Notes [Ref.] zone (%)

outermost 2.2 (DW) Finland (mineral soil) 51-82 year roots “C” isorhapontin old trees, 1.9 (DW) Finland (peatland) (bark) root

0.26 (DW) Finland (mineral soil) diameter less Latva-Mäenpää et al. astringin than 2 cm, 0.47 (DW) Finland (peatland) (2013) distance 0.07 (DW) Finland (mineral soil) from the root piceid neck 54-170 Finland (peatland) 0.06 (DW) cm medium roots 4.2 (DW) Finland (mineral soil) “B” (bark) isorhapontin 1.1 (DW) Finland (peatland) 51-82 year old trees, Finland (mineral soil) 0.95 (DW) distance Latva-Mäenpää et al. astringin 0.44 (DW) Finland (peatland) from the root (2013) neck 20-132 0.12 (DW) Finland (mineral soil) cm piceid 0.04 (DW) Finland (peatland) thick roots 4.6 (DW) Finland (mineral soil) “A” (bark) isorhapontin 1.8 (DW) Finland (peatland) 51-82 year old trees, Finland (mineral soil) Latva-Mänpää et al. 1.1 (DW) distance astringin (2013) 1.2 (DW) Finland (peatland) from the root neck 10-84 0.13 (DW) Finland (mineral soil) cm piceid 0.10 (DW) Finland (peatland) root collar isorhapontin 2.0 (FW) Sweden (peatland) (bark) 30 year old astringin 2.7 (FW) Sweden (peatland) Lindberg et al. (1992) trees piceid 0.28 (FW) Sweden (peatland) roots (bark) isorhapontin 1.8 (FW) Sweden (peatland) 30 year old trees, 0.3-1.0 astringin 1.8 (FW) Sweden (peatland) Lindberg et al. (1992) m from the piceid 0.23 (FW) Sweden (peatland) stem roots (bark) 15 year old isorhapontin 1.3 (FW) Germany Beyer et al. (1993) trees fine roots 15 year old isorhapontin 0.3 (FW) Germany Beyer et al. (1993) (bark) trees

Table 4.4b. The concentrations of stilbenoids (isorhapontin, astringin, piceid) found in Norway spruce (DW; dry weight, FW; fresh weight).

Sampling Stilbenoid Amount Sampling area Notes [Ref.] zone (%)

stump “S” 0.46 (DW) Finland (mineral soil) (bark) isorhapontin Finland (peatland) 0.43 (DW) 51-82 year Finland (mineral soil) old trees, 0.67 (DW) Latva-Mäenpää et al. astringin stump (2013) 0.70 (DW) Finland (peatland) diameter 0.05 (DW) Finland (mineral soil) 116-397 mm piceid 0.03 (DW) Finland (peatland) Total stilbenoid 0.5-2.7 18 years old stem (bark) Finland Jyske et al. (2014) glucosides (DW) tree clones Total stilbenoid 1.2-4.8 37 years old stem (bark) Finland Jyske et al. (2014) glucosides (DW) tree clones

stem (bark) isorhapontin 1.4 (FW) Sweden (peatland) 30 year old trees, 1 m astringin 2.5 (FW) Sweden (peatland) Lindberg et al. (1992) above soil piceid 0.27 (FW) Sweden (peatland) level stem (bark) isorhapontin 0.33 (FW) Norway 15-20 year old trees, 2 Solhaug (1990) astringin 0.14 (FW) Norway m above ground stem (bark) 15 year old isorhapontin 0.4 (FW) Germany Beyer et al. (1993) trees needles isorhapontin 0.03 (FW) Norway 15-20 year old trees, 2 Solhaug (1990) astringin 0.03 (FW) Norway m above ground

Stilbenoid concentrations in Scots pine The concentrations of stilbenoids (pinosylvin, pinosylvin monomethyl ether, pinosylvin dimethyl ether) in various parts of Scots pine are presented in Table 4.5. Our results for the concentrations of pinosylvins found from the stumps (0.74%) are rather similar to those found from the stem heartwood. However, our stump samples contained also the sapwood part and our preliminary results suggest that the heartwood of pine stumps contains 5% of pinosylvin (data not published). These preliminary results suggest that heartwood of the stumps of Scots pine may be promising source of stilbenoids and thus this requires further research.

43 Results and discussion

Table 4.5. The concentrations of stilbenoids (pinosylvin, pinosylvin monomethyl ether, pinosylvin dimethyl ether) found in Scots pine (DW; dry weight, FW; fresh weight).

Sampling zone Stilbenoid Sampling Notes [Ref.] amount (%) area Finland 51-67 year old trees, stump Latva-Mäenpää et al. stump “S” 0.74 (DW) (peatland) diameter 131-297 mm (2010, 2012)

stem heartwood 0.50-0.75 34 year old trees Venäläinen et al. (2004) (aver. sampling height 151 cm) Finland (DW) stem heartwood 0.47-0.64 34 year old trees Venäläinen et al. (2004) (aver. sampling height 91 cm) Finland (DW) stem heartwood (0 m) 2.0 (DW) Norway 87 year old trees Hovelstad et al. (2006) (3.5 m) 1.7 (DW) Norway (5 m) 1.3 (DW) Norway stem heartwood (0 m) 0.28 (DW) Norway 34 year old trees Hovelstad et al. (2006) (3.5 m) 0.23 (DW) Norway knots (3.5 m) 2.1 (DW) Norway 87 year old trees Hovelstad et al. (2006) (5 m) 8.4 (DW) Norway knots 34 year old trees Hovelstad et al. (2006) (3.5 m) 2.6 (DW) Norway stem heartwood 72-84 year old trees Willför et al. (2003) (1.5 m) 0.9-1.2 (DW) Finland knots 16-18 and 72-84 year old Willför et al. (2003) with dead branch (3-15 m) 1.8-5.8 (DW) Finland trees

knots 16-18 and 72-84 year old Willför et al. (2003) with living branch (3-20 m) 1.2-7.0 (DW) Finland trees

4.2 BOTTOM OF THE ROOT NECK OF NORWAY SPRUCE AS A SOURCE OF LIGNANS

Lignans in the heartwood the root neck of Norway spruce This was the first time that lignans were found and analysed in heartwood (hW) of the root neck (RN) of Norway spruce (I and II). The extracts from parts close to the lowermost part of the RN consisted only of lignans without other accompanying compounds in detectable amounts. Two hydroxymatairesinol (HMR) epimers (7(S)-hydroxymatairesinol, and 7(R)-allo- hydroxymatairesinol) were the main lignans detected (Figure 4.5). Also other lignans lariciresinol, secolariciresinol, 7-hydroxylariciresinol (lignan A), 7- isoliovil, and 7-todolactol were tentatively identified.

Fig. 4.5. Molecular structures of major lignans found from the root neck of Norway spruce, hydroxymatairesinol (HMR) epimers

The total concentrations of HMR in the hW of different parts of the RN increased towards the lowermost part of the RN. Furthermore, some sW samples close to the bottom also contain relatively high concentrations of HMR. Bottom of the RN also showed a higher concentration of HMR (5-6%) than the hW in the stems or thick roots (Willför et al. 2003b, 2005b, Ekman 1979). The highest concentration of HMR found from the hW sample of RN was over 10%. This new source of lignans was found to be comparable to the other sources, such as knots (Willför et al. 2003b, 2005b, Piispanen et al. 2008). The reliability of the HMR quantification was tested by application of three internal standards. Small differences between the responses of these standards were found. However, the quantification profiles were similar. It can be safely concluded that the lowermost part of the RN is as valuable a lignan source as the knots in general.

Pressurized hot water extraction (PHWE) of lignans The PHWE method originally designed for the extraction of wood hemicelluloses was found to be promising method to yield HMR from Norway spruce (II). Our study revealed that the extract produced composed mainly of lignans with the two HMR epimers predominating. According to our studies, PHWE is more efficient than acetone/water extraction and it can be considered as a potential green method for the large-scale extraction of lignans.

4.3 EXTRACTION AND PURIFICATION OF STILBENOIDS FROM THE BARK OF NORWAY SPRUCE

The process to extract and isolate stilbenoids from the bark of the roots of Norway spruce was developed (III). Stilbenoid glucosides and stilbenoid aglycones (by additional enzymatic hydrolysis) were sequentially extracted by

45 Results and discussion

accelerated solvent extraction (ASE) and then enriched by polymeric adsorbent followed by an evaporation step. A general protocol, which was developed for the isolation, identification and quantification of stilbenoids in the bark of the roots of Norway spruce is presented in Figure 4.6.

Fig. 4.6. General protocol developed for the isolation, identification and quantification of stilbenoids in the bark of the roots of Norway spruce (III).

The method development for the purification of the bark stilbenoids by HPLC consisted of the optimization of the separation process in the analytical scale and then for scaling it up to the larger sample amounts by semipreparative HPLC (III and IV). The scale up was carried out by increasing the flow rate, sample amount and modification of analytical gradient system to suit the bigger size semi-preparative column. The purity of the fractionated stilbenoids was relatively high. Only in the case of trans-piceid, a small amount of another stilbenoid (trans-isorhapontin) was co-eluted. This was due to the close

retention times of the compounds and also to the presence of large amount of trans-isorhapontin in the extract.

4.4 STRUCTURAL ELUCIDATION OF STILBENOIDS BY UV, MS AND NMR

Identification of stilbenoids in the extract of the bark of roots of Norway spruce The HPLC-DAD chromatograms of the bark extracts obtained from the Norway spruce roots are shown in Figures 4.7 (non-hydrolysed extract) and 4.8 (hydrolysed extract) (III). Major peaks observed showed two broad maximum absorption bands at λ 210-220 and 310-320 nm (see Table 4.7), which are characteristic UV absorptions of trans stilbenoids (Fuendjiep et al. 2002). The molecular masses of compound 1 (tR=11.7), 2 (tR=16.3) and 3 (tR=17.5) in Figure 4.7 were determined as 406, 390 and 420 Da by HPLC-DAD-ESI- MS analysis. The MS/MS spectra of all three compounds exhibited a loss of hexosyl unit (162 u) from the molecular ion, which is characteristic for stilbenoid glycosides.

Fig. 4.7. HPLC-DAD chromatogram non-hydrolysed extract rich in stilbenoid glucosides from Norway spruce root bark

It is not possible to identify the exact structures of the three peaks in Figure 4.7 by comparing their UV absorption and mass spectra with published data (Buiarelli et al., 2007; Hu et al., 2008; Fan et al., 2009; JinLong et al., 2009), because stilbenoids are isomers, showing the same UV absorption and mass spectra. Therefore, additional information should be obtained from off-line HPLC-DAD-NMR to unambiguously identify the constituents of the extract. The gradient system used for HPLC-DAD was modified for optimum “loop collection” of the major chromatographic peaks during off-line HPLC-DAD- NMR analysis. Three major peaks were collected with the same elution order

47 Results and discussion

as in the HPLC-DAD analysis. A single injection or several injections (when required) of the extract into the HPLC column provided enough material to get good quality 1D NMR spectra for compound identification. Also 2D NMR spectra were obtained when enough sample material was available by using standard pulse programmes (correlation spectroscopy (COSY)-45, nuclear Overhauser effect spectroscopy (NOESY), total correlation spectroscopy (TOCSY), heteronuclear single-quantum coherence (HSQC) and heteronuclear multiple bond correlation (HMBC)). The information obtained from off-line HPLC-DAD-NMR analysis together with HPLC-DAD-ESI-MS and HPLC-DAD-ESI-MS/MS confirmed the structural elucidation of the compounds 1, 2 and 3. The structure of compound 3 (as the most abundant compound of the extract) was first established as trans-isorhapontin by all the NMR methods described below. By the same methods, the compound 1 was identified as trans-astringin. The identification of compound 2 was based only on its proton signals obtained from 1H NMR, COSY-45 and TOCSY spectra, and compound 2 was identified as trans-piceid. Our studies utilised a wide range of NMR techniques to provide fully definitive structural elucidation of these natural compounds. Some simple 1H NMR chemical shifts have been reported previously (Ribeiro et al. 1999, Chu et al. 2002 and Ngoc et al. 2008) and they are in agreement with our data. Stilbenoid aglucones were identified from the enzymatically hydrolyzed extract (III). The molecular masses of compounds 4, 5 and 6 were determined as 244, 228 and 258 Da by HPLC-DAD-ESI-MS analysis. The [M-H]- ions of the compounds 4, 5 and 6 are identical with the MS2 [M-H]- ions of compounds 1, 2 and 3, respectively. This relationship revealed that compounds 4, 5 and 6 correspond to aglucone forms of the stilbenoid glucosides. 1H- and COSY-45 NMR spectra fully established the structures 4, 5 and 6.

Fig. 4.8. HPLC-DAD chromatogram of hydrolysed extract rich in stilbenoids from Norway spruce root bark

Compound 4 was identified as trans-piceatannol,. compound 5 was identified as trans-resveratrol and compound 6 was identified as trans-

isorhapontigenin. 1H NMR data of 4 and 6 were in agreement with those in literature (Lojanapiwatna et al. 1982 and Kim et al., 2009). All the stilbenoids extracted and identified in this study were successfully isolated by semipreparative HPLC (Figure 4.9) for further experiments.

Fig. 4.9. HPLC-DAD chromatograms of different fractions of concentrated bark extract obtained from the roots of Norway spruce by semi-preparative-HPLC.

4.5 STABILITY OF STILBENOIDS

4.5.1 STORAGE Trans isomers of the isolated stilbenoids proved stable for 2 weeks (366 h) at -20 °C if protected from fluorescent light (Figure 4.10). The recovery of stilbenoids was in the range of 97-101% for the trans isomers of the five isolated stilbenoids (astringin, isorhapontin, piceid, resveratrol and isorhapontingenin) and 73-106% for trans-piceatannol. The recovery of trans- piceatannol showed a decrease at 1 week but the same was not observed at 2 weeks. Trans-piceatannol showed a broad peak and unresolved chromatogram, which in turn affects the integration of its peak. In such broad trans-piceatannol peak, inconsistent recovery results are expected. Similar to isolated stilbenoids, the constituents of crude extract in solution were also stable when protected from fluorescent light at -20 ºC.

49 Results and discussion

120

100

80 trans-astringin 60 trans-piceid trans-isorhapontin 40 trans-piceatannol 20 trans-resveratrol Recovery (%) Recovery trans-isorhapontigenin 0 0 h 24 h 1 week 2 weeks Storage time

Fig. 4.10. Effect of storage time on stability of isolated stilbenoids under light protected and low temperature (-20 ºC) conditions

4.5.2 PHOTOCHEMICAL REACTIONS AND STABILITY (UNDER FLUORESCENT LIGHT AND UV IRRADIATION)

Stability of stilbenoids under fluorescent light Isolated stilbenoids were not stable when exposed to light. When trans stilbenoids were exposed to fluorescent light for 2 weeks, the area of main peak reduced in time and a new major peak was observed at longer retention time with increased peak area (Figure 4.11).

Fig. 4.11. HPLC–UV chromatogram at 306 nm of isolated trans-resveratrol, exposed to fluorescent light for 1 and 2 weeks in methanol solution

The structures of the new peaks were first tentatively identified from information obtained by HPLC-DAD, HPLC-DAD/MS and HPLC- DAD/MS/MS analyses (Table 4.6). The new products obtained in this study (except the degradation products of trans-astringin and trans-piceatannol) showed UV absorption maxima at ~220 and ~280 nm, which is diagnostic for cis stilbenoids (Hu et al. 2008). The molecular weights of the new products were determined from characteristic m/z of [M-H]- ion and the new products had the same molecular weight as their starting materials. The new products of trans stilbenoids (isorhapontin, isorhapontingenin, piceid and resveratrol) were identified as cis isomers and these results were confirmed by NMR. Our findings are in agreement with those of Trela and Waterhouse (1996) and Jensen et al. (2010), whose study showed that trans-piceid and trans- resveratrol solutions exposed to light were isomerized to cis-piceid and cis- resveratrol respectively. However, our studies are providing new information about the stability of other naturally-occurring stilbenoids than resveratrol.

Table 4.6. Characteristics of original stilbenoids and new products obtained from stilbenoid exposed to fluorescent light in methanol solution (analysed by HPLC-DAD-MS)

- Sample name Products tR (min) UV profile [M-H] , m/z trans-astringin 11.8 trans stilbenoid 405 trans-astringin new peak 16 λ max~260 403 trans-piceid 16.3 trans stilbenoid 389 trans-piceid new peak 27.5 cis stilbenoid 389 trans-isorhapontin 17.5 trans stilbenoid 419 trans-isorhapontin new peak 28.3 cis stilbenoid 419 trans-piceatannol 18.6 trans stilbenoid 243 trans-piceatannol no new peak detected* trans-resveratrol 25 trans stilbenoid 227 trans-resveratrol new peak 32.3 cis stilbenoid 227 trans-isorhapontigenin 26.4 trans stilbenoid 257 trans-isorhapontigenin new peak 33.1 cis stilbenoid 257 *due to a low concentration of the sample, no new peak was detected.

Exposure of trans-astringin to fluorescent light yielded a new product at tR=16.0 by HPLC-DAD with maximum absorption at ~260 nm, which is different from the UV max absorptions (~220 and ~280 nm) of cis stilbenoids. The molecular weight was determined to be 404 Da as can be deduced from [M-H]- = 403 m/z. The new product exhibited a loss of two protons compared to the starting material. Apart from the identification of new products from trans stilbenoids exposed to light, we also calculated the recovery of original material and the yield of new products. Light unprotected conditions caused a strong instability of trans-piceid, trans-resveratrol and trans-astringin whereas trans- isorhapontingenin and trans-isorhapontin where more stable after 2 weeks fluorescent light exposure at room temperature (Figure 4.12). Isomerization

51 Results and discussion

from trans to cis may be deterred for isorhapontin and its aglucone due to steric hindrance caused by the bulky methoxy substituent in their stucture.

100 90 80 70 60 50 trans-astringin 40 trans-piceid Recovery (%) Recovery 30 trans-isorhapontin 20 trans-resveratrol 10 trans-isorhapontigenin 0 0 50 100 150 200 250 300 350 400 Exposure time (h)

Fig. 4.12. Effect of fluorescent light exposure on the stability of isolated trans stilbenoids in MeOH solution

New compounds were formed during the exposure of stilbenes to fluorescent light (Figure 4.13). Maximum conversion was achieved from the isomerization of trans forms of resveratrol and piceid to their corresponding cis isomers after 2 weeks fluorescent light exposure at room temperature. As shown in Figure 4.13, the conversion rate is fast for 36 hours but becomes slower after that. Although a quantitative solution kinetics study was not undertaken, the decrease in the rate of conversion after 36 h is probably due to the degradation of the new products (cis and oxidized stilbenoids) to other compounds. The possibility of degradation of cis stilbenoids can be inferred from the work of Trela and Waterhouse (1996). In their study, 95% pure 79 μmol/L cis- resveratrol in ethanol was placed on the laboratory counter in borosilicate glass and exposed to the overhead fluorescent lighting. They found that the starting material was isomerized to 91% cis-resveratrol over 30 days. This supports our finding in which the formation of a cis stilbenoid became slower after 36 h exposure of a trans isomer to fluorescent light.

100

90 peak derived from trans-astringin cis-piceid 80 cis-isorhapontin 70 cis-resveratrol 60 cis-isorhapontigenin

Yield (%) Yield 40

0 0 100 200 300 400 Exposure time (h)

Fig. 4.13. The yield of cis stilbenoid and another compound (derived from trans-astringin) due to the exposure of trans isomers to fluorescent light in MeOH solution

Stability of crude extracts Crude extract solutions were not significantly destablized by exposure to fluorescent light. Crude extract solutions showed an increase in the amounts of its constituents for 1 week and decrease slowly after 1 week under light exposure conditions and at room temperature. A crude extract contains a variety natural compunds which may decrease the amount of light energy reaching the stilbenoids so that their degradation in the crude extract due to light exposure is lessened. The results of stability assessment revealed that the evaluated stilbenoid glucosides (trans-astringin, trans-isorhapontin and trans-piceid) in a solid crude extract are stable compounds. Our results are supported by resveratrol studies of Prokop et al. (2006) and Jensen et al. (2010). Within the method’s precision, trans stilbenoids showed 100% recovery under all the three conditions evaluated. No trends indicating degradation are seen for the three stilbenoids either under light protected or light unprotected conditions. Although uncapped vials increase the contact of sample with atmospheric air, oxidation and isomerization of the evaluated stilbenoids were not observed. This revealed that solid stilbenoids (trans-astringin, trans-isorhapontin and trans-piceid) are stable compounds for at least 2 weeks with negligible degradation for the three conditions evaluated. Not only under light protected conditions, but also when exposed to fluorescent light at ambient lab conditions solid stilbenoids remain as the thermodynamically more stable trans isomers.

53 Results and discussion

UV stability of stilbenoids The isomerization of trans stilbenoids to cis forms started immediately after exposure to UV light (366 nm) and most of the trans structures were lost after 10 – 30 min (Figure 4.14). Trans-astringin and trans-isorhapontin showed the lowest rate of the disappearance of the trans form. The yield of cis stilbenoids was highest at the beginning of the UV irradiation (10 – 60 min) and then it started to decrease (Figure 4.15). This decrease was due the transformation of cis stilbenoids to new products.

100 trans-astringin 90 trans-piceid 80 70 trans-isorhapontin 60 trans-resveratrol

50 trans-isorhapontigenin 40 Recovery (%) Recovery 30 20 10 0 01030601203001440 Exposure time (min)

Fig. 4.14. Effect of UV light exposure on the stability of isolated trans stilbenoids in MeOH

100 90 80 70 60 cis-astringin 50 cis-piceid

Yield (%) Yield 40 cis-isorhapontin 30 cis-resveratrol 20 cis-isorhapontigenin 10 0 01030601203001440 Exposure time (min)

Fig. 4.15. The yield of cis stilbenoids due to the exposure of trans isomers to UV light in MeOH solution

The GC-MS information of trans stilbenoids isolated and the compounds derived from those during UV irradiation (366nm) are presented in Table 4.7. In general, trans stilbenoids isomerized rapidly to cis forms and after that all the compounds lost two hydrogens from their structures by oxidation. In the case of piceatannol, the cis form was not detected by GC-MS. The UV profiles of new compounds formed showed an absorbance maxima at 260 nm (the same phenomenon was observed earlier with astringin under fluorescent light described in Table 4.7). In the hydrolysed crude extract, all three trans stilbenoid aglucones isomerized to cis forms, which were major compounds after 2 and 24 h. The new peaks with loss of two hydrogens derived from stilbenoid aglucones were found from the hydrolysed crude extract after 24 h UV radiation by GC-MS as in the case of isolated stilbenoid aglucones. The same phenomenon was observed in the non-hydrolysed extracts containing the trans stilbenoid glucosides.

Table 4.7. Characteristics of original stilbenoids and new products obtained from stilbenoids exposed to UV light (366 nm) in solution by GC-MS and HPLC-DAD

Major GC-MS GC-MS compound Retention HPLC-DAD Sample Products (TMSi), at UV time (UV profile) m/z (0h,2h,24h) (min) trans-astringin 0h 41.72 532 trans stilbenoid trans-astringin cis-astringin 2h 32.04 532 cis stilbenoid new peak 24h 34.48 530 λ max~260 trans-piceid 0h 38.42 444 trans stilbenoid trans-piceid cis-piceid 2h 30.30 444 cis stilbenoid new peak 24h 31.77 442 λ max~260 trans-isorhapontin 0h 42.45 474 trans stilbenoid cis-isorhapontin 2h/24h 31.62 474 cis stilbenoid trans- new peak 1 24h 30.43 472 λ max~260 isorhapontin new peak 2 24h 34.48 472 λ max~260 new peak 3 24h 43.54 472 λ max~260 trans-piceatannol 0h 22.76 532 trans stilbenoid trans- new peak 1 2h 21.57 530 λ max~260 piceatannol new peak 1 24h 21.57 530 λ max~260 trans-resveratrol 0h 20.00 444 trans stilbenoid trans-resveratrol cis-resveratrol 2h 13.89 444 cis stilbenoid new peak 24h 18.64 442 λ max~260 trans-isorhapontigenin 0h 22.55 474 trans stilbenoid trans- cis-isorhapontigenin 2h 15.62 474 cis stilbenoid isorhapontigenin new peak 24h 16.36 472 λ max~260

55 Results and discussion

4.5.3 FRACTIONATION AND STRUCTURAL ELUCIDATION OF NEW COMPOUNDS DERIVED FROM STILBENOIDS

Isolation and identification new compounds formed during UV radiation Semipreparative HPLC was utilized for the isolation of new compounds derived from the trans stilbenoids studied. New compounds were identified by MS and NMR. All the stilbenoids showed similar kind of transformations during UV radiation; rapid isomerization from trans to cis form and after longer time of radiation, a formation of new compounds with the loss of two protons (-2 H). The mass spectra of the fractions derived from trans stilbenoids studied are presented in Tables 4.8 (stilbenoid glucosides, astringin and isorhapontin by ESI-TOF-MS) and 4.9 (hydrolyzed stilbenoids, piceatannol and isorhapontigenin by EI-MS).

Table 4.8. Identification of compounds derived from trans-astringin (F1) and trans-isorhapontin (F3) by HR ESI-TOF-MS

Compound Stilbenoid [M-](m/z) Calcd. Formula

F1.1 trans-astringin 405.1180 405.1186 C20H21O9

F1.2 Unknown 403.1036 403.1029 C20H19O9

F1.3 cis-astringin 405.1243 405.1186 C20H21O9

F3.1 trans-isorhapontin 419.1479 419.1342 C21H23O9

F3.2 Unknown 417.1321 417.1186 C21H21O9

F3.3 Unknown 417.1327 417.1186 C21H21O9

F3.4 cis-isorhapontin 419.1487 419.1342 C21H23O9

F3.5 Unknown 417.1321 417.1186 C21H21O9

Table 4.9. Identification of compounds derived from trans-piceatannol (F4) and trans- isorhapontigenin (F6) by EI-MS

M Compound Stilbenoid Formula (m/z)

F4.1 Unknown 242 C14H10O4 F4.2 cis-piceatannol 244 C14H12O4 F6.1 trans-isorhapontigenin 258 C15H14O4 F6.2 Unknown 256 C15H14O4 F6.3 Unknown 256 C15H12O4 F6.4 cis-isorhapontigenin 258 C15H14O4 F6.5 Unknown 256 C15H14O4

The structural elucidation of unknown fractions formed from astringin (and its aglycone piceatannol) and isorhapontin (and its aglycone isorhapontigenin) For the new compound derived from trans-astringin (F1.2), the ESI-MS spectrum showed 403.1036 m/z as the parent peak. Its molecular weight is two units less than that of trans- or cis-astringin. In the 1H-NMR spectrum, there are six protons signals between 6.00-10ppm,. Alkene H bond signals which had larger coupling constant (around 12-16Hz) were absent from the aromatic range. The signal δ 9.008(1H, s) could be an aldehyde proton, but the aldehyde carbon signal (around 200ppm) however does not appear in the carbon spectrum. This proton connected with carbon (about 104ppm) is therefore an aromatic proton. The signals 7.435(1H, d, J=9Hz) and 7.319(1H, d, J=9Hz) belong to a pair of ortho-coupled aromatic protons. They do not couple with other aromatic protons. The signals 6.968(1H, d, J=2.4Hz) and 6.843(1H, d, J=2.7Hz) had almost the same coupling constants, so they may be from the same aromatic ring. 2D NMR (COSY, HSQC, and HMBC) also supported these findings. Thus, he new structure identified (F1.2) is 7-O-β-D- glucosyl-2,3,5-trihydroxyphenanthrene (Figure 4.16).

Fig. 4.16. The proposed structure of F1.2, a phenanthrene molecule derived from the stilbenoid astringin by to a photochemical cyclization and dehydrogenation induced by UV irradiation.

For the new compounds derived from trans-isorhapontin, the negative ESI- MS spectrum showed the strongest peak at 417.1321m/z (for F3.2, F3.3, F3.5) two units less than trans- and cis-isorhapontin. 1H-NMR and 2D NMR signals of F3.2 were almost same as those in fraction F1.2, except for the OCH3 signal at 3.992. F3.5 showed the shift of -OCH3 signal to 3.432 ppm, which may be due to the formation of different isomers of phenanthrenes since the positions of –OH and –OCH3 groups next to each other have two alternative arrangements in the cis isomer (which is the starting point of phenanthrene formation). F3.3 was the mixture of the fractions F3.2 and F3.5. The new structures identified (F3.2 and F3.5) are 7-O-β-D-glucosyl-3,5-dihydroxy-2- methoxyphenanthrene and 7-O-β-D-glucosyl-2,5-dihydroxy-3-methoxyphenanthrene (Figure 4.17).

57 Results and discussion

Fig. 4.17. The proposed structures of F3.2 and F3.5, phenanthrene molecules derived from the stilbenoid isorhapontin due to a photochemical cyclization and dehydrogenation induced by UV irradiation.

The same kind of phenanthrene NMR signals were also observed for the new compounds derived from trans-piceatannol and trans-isorhapontigenin presumably by similar transformation pathways. This data acquired from NMR in combination with our earlier UV and MS data suggest that stilbenoids go through photochemical transformation to hydroxy-phenanthrene structures by intramolecular cyclisation a formation of a new C-C bond. These suggested phenanthrene sructures formed are supported by the earlier data based on the studies of trans-resveratrol and its glucoside, piceid (Leong et al. 1999, Montsko et al. 2008, Tříska et al. 2012, Yang et al. 2012) and for studies of pure stilbene molecule itself (Moore et al. 1963 and Sigman et al. 2001). However, the data available from the literature are not based on the same molecules studied (and also the source of compounds is different from those stduied earlier), which increases the value of our findings. A suggested mechanism of photochemical transformation of cis stilbenoids to phenanthrenes is shown in Figure 4.18.

Fig. 4.18. Proposed mechanism of photochemical transformation of cis stilbenoids to phenanthrenes.

This is the first time that photochemical transformations of trans-astringin and trans-isorhapontin (and of their aglycones, trans-piceatannol and trans- isorhapontigenin) derived from Norway spruce root bark have been reported and described (Figure 4.19). These molecules formed by photochemical transformation of naturally oocurring stilbenoids have not been reported in the literature previously.

Fig. 4.19. Proposed photochemical transformation of selected trans stilbenoids (astringin and isorhapontin and their aglycones piceatannol and isorhapontigenin) derived from Norway spruce root bark. R = -H or -O-β-D-glucoside (Glu).

59 Concluding remarks

5 CONCLUDING REMARKS

In this study, we have found out that the bark of roots and stumps of Norway spruce are rich in stilbenoid glucosides (astringin, isorhapontin and piceid) with structural similarities with the most studied stilbenoid, resveratrol. We also found that the stumps of Scots pine are a source of stilbenoids, pinosylvin and its derivatives. We described the simple and relatively fast methods to analyse the extractive profiles from conifer biomass by chromatographic and spectrometric methods (GC and MS). This is the first time that the biomass from the roots and stumps of these Finnish conifer trees with a vast economic and environmental significance have been studied with this level of fractionation (certain zones of biomass). Samples from two different forest areas (peatland and mineral soil) were studied and great variation between the forest sites but also between single samples was observed. The highest contents of stilbenoid glucosides were found from the root zone closest to the stem. The stilbenoid glucosides were localized to the inner part of the bark of Norways spruce roots, where they may act physiologically as a reserve for protective compounds against external threats. This study revealed a new source of the bioactive lignan hydroxymatairesinol (HMR). The root neck of Norway spruce was found to be a rich and new source of HMR, which has been previously extracted from the knots. The lowest part of the root neck was the best source of HMR. This novel finding may explain the role of HMR in the protective function of the tree and its roots (in comparison to branches and knots). Pressurised hot water extraction (PHWE) showed its potential as a green extraction method for wood lignans. The HPLC-DAD-NMR method was developed and described for the first time for the structural elucidations of stilbenoids from the bark of Norway spruce roots. We also present the whole process chain for the sampling, extraction, concentration/pre-purification and further isolation of spruce bark stilbenoids. Stilbenoids were efficiently extracted and refined from biomass with industrially applicable methods and also identified by HPLC-MSn and HPLC-DAD-NMR methods. The HPLC method was further extended to a semipreparative scale. We identified new compounds derived from naturally-occurring stilbenoids from the bark of Norway spruce roots. The invividual stilbenoid molecules were isolated from the extracts and their photochemical stabilities (under fluorescent and UV light) were studied. Stilbenoids occur naturally in trans form and may isomerise to the cis form by light energy. Isomerisation was much faster in the presence of UV light. The extended irradiation with UV light revealed that the stilbenoids undergo photochemical transformations via the trans to cis isomerization. The molecular structure of stilbenoids changes

by intramolecular cyclisation (new C-C bond and loss of two protons) to phenanthrene structures. This is the first time that these new phenanthrene structures are reported to be derived from stilbenoids of spruce bark. The compounds identified appear to be completely novel molecules not found earlier. Stilbenoids are bioactive compounds and they have shown a variety of beneficial activities in studies related for example to human health. They have also shown antimicrobial and protective properties in trees and also in tests against micro-organisms. According to literature, stilbenoids derived from conifer trees are promising compounds for utilisation as drugs, antioxidative food supplements or protective agents. Also natural phenanthrenes have been reported to possess a variety of biological activities (Kovacs et al. 2008). Nevertheless, more studies are needed to establish how stilbenoids may be effectively isolated from biomass in a larger scale. The chemical properties of stilbenoids such as stability, reactivity and safety need to be studied in detail before they can be utilised. Also open-minded testing of stilbenoids for potential commercial applications is essential. However, it can be concluded that stilbenoids have real potential as commercial biochemicals and Norway spruce and Scots pine are rich sources of these compounds. As a conclusion, our studies have shown that the roots and stumps of conifer trees contain bioactive polyphenolic compounds of relatively small molecule sizes. These molecules and the compounds formed thereof may have commercial value and be used as single compounds or as lead compounds in specific indication areas (and generally natural antioxidants) in the food supplements, pharmaceuticals, cosmetics and other industries (Figure 5.1). Our study has also provided new scientific knowledge of natural compounds and their properties that may lead to new findings and deeper understanding of the physiology and internal protection processes of plants by their bioactive secondary metabolites.

Fig. 5.1. Value chain thinking about research and utilisation of stilbenoids.

61 References

REFERENCES

Agrawal, P.K. (1992). NMR spectroscopy in the structural elucidation of oligosaccharides and glycosides. Phytochemistry, 31, 3307–3330.

Alfredsen, G., Solheim, H. and Slimestadt, R. (2008) Antifungal effect of bark extracts from some European tree species. European Journal of Forest Research, 127, 387–393

Balas, A. and Popa, V.I. (2007). On characterization of some bioactive compounds extracted from Picea abies bark. Biotechnology Letters, 123209– 3216.

Bergström, D. and Matisons, M. (2014) Forest Refine, 2012-2014 – Efficient forest biomass supply chain management for biorefineries, Synthesis report.

Beyer, U., Tesche, M., Heller, W. and Sandermann, H. (1993) Fungistatic effectiveness of phenolic compounds in Norway spruce (Picea abies) and the effect of SO2. Forstwissenschaftliches Centralblatt, 112, 251–256.

Bhullar, K. and Hubbard, B. (2015) Lifespan and healthspan extension by resveratrol. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease, 1852 (6), 1209–1218.

Buiarelli, F., Coccioli, F., Jasionowska, R., Merolle, M. and Terracciano, A. (2007). Analysis of some stilbenes in Italian wines by liquid chromatography/tandem mass spectrometry. Rapid Communications of Mass Spectrometry, 21, 2955–2964.

Castellano, G., Lara, A. and Torrens, F. (2014) Classification of stilbenoid compounds by entropy of artificial intelligence. Phytochemistry, 97, 62–69.

Celimene, C., Micales, J., Ferge, L. and Young, R. (1999) Efficacy of pinosylvins against white-rot and brown-rot fungi. Holzforschung, 53 (5), 491–497.

Chemat, F., Abert Vian, M. and Cravotto, G. (2012) Green extraction of natural products: Concepts and principles. International Journal of Molecular Sciences, 13, 8615–8627.

Chong, J., Poutaraud, A. and Hugueney, P. (2009) Metabolism and roles of stilbenes in plants. Plant Science, 177, 143–155.

Co, M., Fagerlund, A., Engman, L., Sunnerheim, K., Sjöberg, P. and Turner, C. (2012) Extraction of antioxidants from spruce (Picea abies) bark using eco- friendly solvents. Phytochemical Analysis, 23, 1–11.

Erdtman, V.H. (1939) Die phenolischen Inhaltsstoffe des Kiefernkernholzes, ihre physiologische Bedeutung und hemmende Einwirkung auf die normale Aufschließbarkeit des Kiefernkernholzes nach dem Sulfitverfahren. Liebigs Annalen der Chemie, 539, 116–127.

Evensen. P., Solheim, H., Hoiland, K. and Stenersen, J. (2000) Induced resistance of Norway spruce, variation of phenolic compounds and their effects on fungal pathogens. Forest Pathology, 30, 97–108.

Fan, P., Marston, A., Hay, A.E and Hostettmann, K. (2009) Rapid separation of three glucosylated resveratrol analogues from the invasive plant Polygonum cuspidatum by high-speed countercurrent chromatography. Journal of Separation Science, 32, 2979–2984.

Fang, W., Hemming, J., Reunanen, M., Eklund, P., Pineiro, E.C., Poljansek, I., Oven, P. and Willför, S. (2013) Evaluation of selective extraction methods for recovery of polyphenols from pine. Holzforschung, 67 (8), 843–851.

Fitzpatrick, D.F., Hirschfield, S.L. and Coffey, R.G. (1993) Endothelium- dependent vasorelaxing activity of wine and other grape products. The American Journal of Physiology – Heart and Circulatory Physiology, 265, 774–778.

Forest Bioeconomy NRA (2016) Recommendations for the Finnish forest- based bioeconomy R&D by the Finnish National Support Group to the Forest- based Sector Technology Platform (FTP Finnish NSG task force group). http://www.biotalous.fi/wp- content/uploads/2016/02/Recommendations_Finnish_Forest_based_bioec onomy_RD.pdf

Front Research: http://www.frontresearch.com/news/global-resveratrol- market-to-grow-by-4-5-annually-by-2021/

Fuendjiep, V., Wandji, J., Tillequin, F., Mulholland, D.A., Budzikiewicz, H., Fomum, Z.T., Nyemba, A.M. and Koch, M. (2002) Chalconoid and stilbenoid glycosides from Guibourtia tessmanii. Phytochemistry, 60, 803–806.

Fuhrman, B., Lavy, A. and Aviram, M. (1995) Consumption of red wine with meals reduces the susceptibility of human plasma and low-density lipoprotein to lipid peroxidation. The American Journal of Clinical Nutrition, 61 (3), 549– 554.

63 References

Giuliani, C., Marzorati, M., Innocenti, M., Vital M., Pieper D.H., Van de Wiele T. and Mulinacci N. (2016) Dietary supplement based on stilbenes: a focus on gut microbial metabolism by the in vitro simulator M-SHIME®. Food & Function, 7, 4564–4575.

Guerrero, R., Biais, B., Richard, T., Puertas, B., Waffo-Teguo, P., Merillon, J- M. and Cantos-Villar, E. (2016) Grapevine cane’s waste is a source of bioactive stilbenes. Industrial Crops and Products, 94, 884–892.

Hammerbacher, A., Ralph, S., Boehlmann, J., Fenning, T., Gershenzon, J. and Schmidt, A. (2011) Biosynthesis of the major tetrahydroxystilbenes in spruce, astringin and isorhapontin, proceeds via resveratrol and is enhanced by fungal infection. Plant Physiology, 157, 876–890.

Harju, A., Venäläinen, M., Laakso, T. and Saranpää, P. (2009) Wounding response in xylem of Scots pine seedlings shows wide genetic variation and connection with the constitutive defense of heartwood. Tree Physiology, 29 (1), 19–25.

Harlamow, R.,Willför, S., Hemming, J., Honkalampi-Hämäläinen, U., von Wright, A. and Holmbom, B. (2007) Extraction and biotests of stilbenes in spruce inner bark. Proceedings of the Meeting 50 years of the Phytochemical Society of Europe, Cambridge, UK, 11–14 April 2007.

Hart, J.H. (1981) Role of phytostilbenes in decay and disease resistance. Annual Review of Phytopathology, 19 (1), 437–458.

Hovelstad, H., Leirset, I., Oyaas, K. and Fiksdahl, A. (2006) Screening analyses of stilbenes, resin acids and lignans in Norwegian conifers. Molecules, 11, 103–114.

Hu, B., Liu, X., Zhang, C. and Zeng, X. (2016) Food macromolecule based nanodelivery systems for enhancing the bioavailability of polyphenols. Journal of Food and Drug analysis, in press.

Hu, Y., Ma, S., Li, J., Yu, S., Qu, J., Liu, J. and Du, D. (2008) Targeted isolation and structure elucidation of stilbene glycosides from the bark of Lysidice brevicalyx Wei guided by biological and chemical screening. Journal of Natural Products, 71, 1800–1805.

Hung, L.M., Chen, J.K. Lee, R.S. Liang, H.C. and Su, M.J. (2001) Beneficial effects of astringinin, a resveratrol analogue, on the ischemia and reperfusion damage in rat heart. Free Radical Biology and Medicine, 30 (8), 877–883.

Ioannidis, K., Melliou, E., Alizoti, P. and Magiatis, P. (2016) Identification of black pine (Pinus nigra Arn.) heartwood as a rich source of bioactive stilbenes by qNMR. Journal of the Science of Food and Agriculture, DOI: 10.1002/jsfa.8090

Jancinova, V., Perecko, T., Nosal, R., Harmatha, J, Smidrkal, J. and Drabikova, K. (2012) The natural stilbenoid pinosylvin and activated neutrophils: effects on oxidative burst, protein kinase C, apoptosis and efficiency in adjuvant arthritis. Acta Pharmacologica Sinica, 30(10), 1285– 1892.

Jang, M., Cai, L., Udeani, G.O. Thomas, C., Beecher, C., Fong, H., … Pezzuto, J. (1997) Cancer chemopreventive activity of resveratrol, a natural product derived from grapes. Science, 275, 218–220.

Jensen, J.S., Wertz, C.F and O’Neill, V.A. (2010) Preformulation stability of trans-resveratrol and trans-resveratrol clucoside (piceid). Journal of Agricultural and Food Chemistry, 58, 1685–1690.

Jin Long, C., Meng Meng, M., Yan Wei, L., Li Sheng, W., Chu Tse, W. and Hai Feng, D. (2009) Rhaponticin from rhubarb rhizomes alleviates liver steatosis and improves blood glucose and lipid profiles in KK/Ay diabetic mice. Planta Med., 75, 472–477.

Jyske, T. Laakso, T., Latva-Mäenpää, H. Tapaniola, T. and Saranpää, P. (2014) Yield of stilbene glucosides from the bark of young and old Norway spruce stems. Biomass and Bioenergy, 71, 216–227.

Jyske, T., Kuroda, K., Suuronen, J-P. Pranovich, A., Roig-Juan, S., Aoki, D. and Fukushima, K. (2016) In planta localization of stilbenes within Picea abies phloem. Plant Physiology, 172, 913–928.

Jyske, T., Suuronen, J-P. and Pranovich, A. (2015) Seasonal variation in formation, structure, and chemical properties of phloem in Picea abies as studied by novel microtechniques. Planta, 242(3), 613–629.

Kilpeläinen, P., Leppänen, K., Spetz, P., Kitunen, V., Ilvesniemi, H., Pranovich, A and Willför, S. (2012) Pressurised hot water extraction of acetylated xylan from birch sawdust. Nordic Pulp & Paper Research Journal, 27, 680–688.

Kim, E.J., Park, H., Park, S.Y., Jun, J.G. and Park, J.H. (2009) The grape component piceatannol induces apoptosis in DU145 human prostate cancer cells via the activation of extrinsic and intrinsic pathways. Journal of Medicinal Food, 12, 943–951.

65 References

Kimura, Y., Baba, K. and Okuda, H. (2000) Inhibitory effects of active substances isolated from Cassia garrettiana heartwood on tumor growth and lung metastasis in Lewis lung carcinoma-bearing mice (part 2). Anticancer Research, 20, 2923–2930.

Kiselev, K., Grigorchuk, V., Ogneva, Z., Suprun, A. and Dubrovina, A. (2016a) Stilbene biosynthesis in the needles of spruce Picea jezoensis. Phytochemistry, 131, 57–67.

Kiselev, K., Ogneva, Z., Suprun, A. and Zhuravlev, Y. (2016b) Expression of the stilbene synthase genes in the needles of spruce Picea jezoensis. Russian Journal of Genetics, 52(11), 1157–1163.

Kolouchová-Hanzlíková, I., Melzoch, K., Filip, V and Smidrkal, J. (2004) Rapid method for resveratrol determination by HPLC with electrochemical and UV detections in wines. Food Chemistry, 87, 151–158.

Koskela, A., Reinisalo, M., Hyttinen, J., Kaarniranta, K. and Karjalainen, R. (2014) Pinosylvin-mediated protection against oxidative stress in human retinal pigment epithelial cells. Molecular Vision, 20, 760–769.

Kovacs, A., Vasas, A and Hohmann, J. (2008) Natural phenanthrenes and their biological activity. Phytochemistry, 69, 1084–1110.

Laavola, M., Nieminen, R., Leppänen, T., Eckerman, C., Holmbom, B. and Moilanen, E. (2015) Pinosylvin and monomethylpinosylvin, constituents of an extract from the knot of Pinus sylvestris, reduce inflammatory gene expression and inflammatory responses in vivo. Journal of Agricultural and Food Chemistry, 63 (13), 3445–3453.

Landete, J., Arqués, J., Medina, M., Gaya, P., de Las Rivas, B. and Muñoz, R. (2016) Bioactivation of phytoestrogens: intestinal bacteria and health. Critical Reviews in Food Science and Nutrition, 56(11), 1826–1843.

Latva-Mäenpää, H., Laakso, T., Phan, D-T., Pihlajamaa, T., Sarjala, T., Wähälä, K. and Saranpää, P. (2010) Bioactive and protective compounds from roots and stumps. In: Polyphenols Communications 2010. p. 94-95. Montpellier, France, 24–27 August, 2010.

Latva-Mäenpää, H., Laakso, T., Pihlajamaa, T., Phan, D-T., Sarjala, T., Wähälä, K. and Saranpää, P. (2012) Structure and chemistry of roots and stumps of Norway spruce and Scots pine. In: Final program, proceedings and abstracts book of the 2012 IUFRO Conference, Division 5 Forest Products, p. 187, Estoril, Portugal, 8–13 July, 2012.

Latva-Mäenpää, H., Laakso, T., Sarjala, T., Wähälä, K. and Saranpää, P. (2013) Variation of stilbene glucosides in bark extracts obtained from roots and stumps of Norway spruce (Picea abies [L.] Karst.). Trees – Structure and Function, 27, 131–139.

Latva-Mäenpää, H., Laakso, T., Sarjala, T., Wähälä, K. and Saranpää, P. (2014) Root neck of Norway spruce as a source of bioactive lignans and stilbenes. Holzforschung, 68 (1), 1–7.

Leong, Y-W., Harrison, L. and Powell, A. (1999) Phenanthrene and other aromatic constituents of Bulbophyllum vaginatum. Phytochemistry, 50 (7), 1237–1241.

Leppänen, K., Spetz, P., Pranovich, A., Hartonen, K., Kitunen, V. and Ilvesniemi, H. (2011) Pressurized hot water extraction of Norway spruce hemicelluloses using a flow-through system. Wood Science and Technology, 45, 223–236.

Li, H.L., Wang, A.B., Huang, Y., Liu, D., Wei, C., Williams, G., … Liang, C. (2005) Isorhapontigenin, a new resveratrol analog, attenuates cardiac hypertrophy via blocking signaling transduction pathways. Free Radical Biology & Medicine, 38, 243–257.

Li, S-H., Nagy, N., Hammerbacher, A. Krokene, P., Niu, X.M., Gershenzon, J. and Schneider, B. (2012) Localization of phenolics in phloem parenchyma cells of Norway spruce (Picea abies). ChemBioChem, 13, 2707–2713.

Li, S-H., Niu, X-M., Zahn, S., Gershenzon, J., Weston, J. and Schneider, B. (2008) Diastereomeric stilbene glucoside dimers from the bark of Norway spruce (Picea abies). Phytochemistry, 69, 772–782.

Likhtenstein, G. (2010) Stilbenes - Applications in Chemistry, Life Sciences and Materials Science, Wiley-VCH, Weinheim.

Lindberg, L., Willför, S. and Holmbom, B. (2004) Antibacterial effects of knotwood extractives on paper mill bacteria. Journal of Industrial Microbiology & Technology, 31 (3), 137–147.

Lindberg, M., Lundgren, L., Gref, R. and Johansson, M. (1992) Stilbenes and resin acids in relation to the penetration of Heterobasidion annosum through the bark of Picea abies. The European Journal of Plant Pathology, 22, 95– 106.

67 References

Liu, Yi., Liu, Ya, Chen, H., Yao, X., Xiao, Y., Zheng, X., … Zheng, X. (2015) Synthetic resveratrol derivatives and their biological activities: a review. Open Journal of Medicinal Chemistry, 5, 97–105.

Lojanapiwatna, V., Majaeha, J., Wiriyachitra, P. (1982) Isolation of 3,4',5- trihydroxy-3'-methoxystilbene from Gnetum cuspidatum BL. Journal of the Science Society of Thailand, 8, 59–63.

Lu, J., Venäläinen, M., Julkunen-Tiitto, R. and Harju, A. (2016) Stilbene impregnation retards brown-rot decay of Scots pine sapwood. Holzforschung, 70(3), 261–266.

Macickova, T., Drabikova, K., Nosal, R., Bauerova, K., Mihalova, D., Harmatha, J. and Pecivova, J. (2010) In vivo effect of pinosylvin and pterostilbene in the animal model of adjuvant arthritis. Neuro Endocrinology Letters, 31, 91–95.

Mannila. E. and Talvitie, A. (1992) Stilbenes from Picea abies bark. Phytochemistry, 31, 3288–3289.

Mantegna, S., Binello, A., Boffa, L., Giorgis, M., Cena, C and Cravotto, G. (2012) A one-pot ultrasound-assisted water extraction/cyclodextrin encapsulation of resveratrol from Polygonum cuspidatum. Food Chemistry, 130, 746–750.

Mérillon, J.M., Fauconneau, B., Teguo, P.W., Barrier, L., Vercauteren, J. and Huguet, F. (1997) Antioxidant activity of the stilbene astringin, newly extracted from Vitis vinifera cell cultures. Clinical Chemistry, 43, 1092–1093.

Montsko, G., Pour Nikfardjam, M.S., Szabo, Z., Boddi, K., Lorand, T., Ohmacht, R. and Mark, L. (2008) Determination of products derived from trans-resveratrol UV photoisomerisation by means of HPLC-APCI-MS. Journal of Photochemistry and Photobiology A, 196, 44–50.

Moore, W., Morgan, D. and Stermitz, F. (1963) The photochemical conversion of stilbene to phenanthrene. The nature of the intermediate. Journal of the American Chemical Society, 86 (6), 829–830.

Mulat, D., Latva-Mäenpää, H., Koskela, H., Saranpää, P. and Wähälä, K. (2014) rapid chemical characterisation of stilbenes in the root bark of Norway spruce by off-line HPLC/DAD-NMR. Phytochemical Analysis, 25 (6), 529– 536.

Münzenberger, B., Heilemann, J., Strack, D., Kottke, I. and Oberwinkler, F. (1990) Phenolics of mycorrhizas and non-mycorrhizal roots of Norway spruce. Planta, 182,142–148.

Nan Fang, Y.E. and Tao Liu, G. (2002) Effect of isorhapontigenin on respiratory burst of rat neutrophils. Phytomedicine, 9 (8), 734–738.

Ngoc, T.M., Minh, P.T.H., Hung, T.M., Thuong, P.T., Lee, I.S., Min, B.S. and Bae, K.H. (2008) Lipoxygenase inhibitory constituents from rhubarb. Archives of Pharmacal Research, 31, 598–605.

Nutraingredients (Frost and Sullivan report): http://www.nutraingredients.com/Markets-and-Trends/US-dominates- global-resveratrol-market

Pan, H. and Lundgren, L. (1995) Phenolic extractives from root bark of Picea abies. Phytochemistry, 39, 1423–1428.

Park, E.J., Park, H.J., Chung, H-J., Shin, Y., Min, H.Y., Hong, J.Y., Kang, Y.J., Ahn, Y.H., Pyee, J.H. and Lee, S.K. (2012) Antimetastatic activity of pinosylvin, a natural stilbenoid, is associated with the suppression of matrix metalloproteinases. The Journal of Nutritional Biochemistry, 23 (8), 946– 952.

Perecko, T., Jancinova, V., Drabikova, K., Nosal, R. and Harmatha, J. (2008) Structure-efficiency relationship in derivatives of stilbene. Comparison of resveratrol, pinosylvin and pterostilbene. Neuro Endocrinology Letters, 29 (5), 802–805.

Pietarinen, S., Willför, S., Ahotupa, M., Hemming, J. and Holmbom, B. (2006) Knotwood and bark extracts: strong antioxidants from waste materials. Journal of Wood Science, 52 (5), 436–444.

Pineiro, Z., Marrufo-Curtido, A., Serrano, M. and Palma, M. (2016) Ultrasound-Assisted Extraction of Stilbenes from Grape Canes. Molecules, 21, 784–796.

Piñeiro, Z., Palma, M. and Barroso, C.G. (2006) Determination of trans- resveratrol in grapes by pressurised liquid extraction and fast high- performance liquid chromatography. Journal of Chromatography A, 1110, 61–65.

Prokop, J., Abrman, P., Seligson, A.L. and Sovak, M. (2006) Resveratrol and Its Glycon Piceid Are Stable Polyphenols. Journal of Medicinal Food, 9, 11– 14.

69 References

Päivänen, J. and Paavilainen, E. (1996) Forestry in peatlands. In Vasander H (Ed.), Peatlands in Finland. Finnish Peatland Society. ISBN 952-90-7971-0, pp 72–83.

Reinisalo, M., Kårlund, A., Koskela, A., Kaarniranta, K. and Karjalainen, R. (2015) Polyphenol stilbenes: molecular mechanisms of defence against oxidative stress and aging-related diseases. Oxidative Medicine and Cellular Longevity, 2015;2015:340520. doi: 10.1155/2015/340520.

Repo, A., Tuomi, M. and Liski, J. (2011) Indirect carbon dioxide emissions from producing bioenergy from forest harvest residues. GCB Bioenergy, 3 (2), 107–115.

Regev-Shoshani, G., Shoseyov, O., Bilkis, I. and Kerem, Z. (2003) Glycosylation of resveratrol protects it from enzymic oxidation. Biochemical Journal, 374, 157–163.

Rimando, A.M. and Suh, N. (2008) Biological/Chemopreventive activity of stilbenes and their effect on colon cancer. Planta Medica, 74, 1635–1643.

Roupe, K., Halls, S. and Davies, N.M. (2005) Determination and assay validation of pinosylvin in rat serum: application to drug metabolism and pharmacokinetics. Journal of Pharmaceutical and Biomedical Analysis, 38, 148–154.

Roupe, K.A., Remsberg, C.M., Yáñez, J.A. and Davies, N.M. (2006) Pharmacometrics of stilbenes: seguing towards the clinic. Current Clinical Pharmacology, 1, 81–101.

Saarinen, N., Wärri, A., Mäkelä, S., Eckerman, C., Reunanen, M., Ahotupa, M., Salmi, M., Franke, A., Kangas, L. and Santti, R. (2000) Hydroxymatairesinol, a novel enterolactone precursor with antitumor properties from coniferous tree (Picea abies). Nutrition and Cancer, 36, 207–216.

Seigneur, M., Bonnet, J., Dorian, B., Benchimol, D., Drouillet, F., Gouverneur, G., … Bricaud, H. (1990) Effect of the consumption of alcohol, white wine and red wine on platelet function and serum lipids. Journal of Applied Cardiology, 5 (3), 215–222.

Sekine, N., Ashitani, T., Murayama, T., Shibutani, S., Hattori, S. and Takahashi, K. (2009) Bioactivity of latifolin and its derivatives against termites and fungi. Journal of Agricultural and Food Chemistry, 57 (13), 5707–5712.

Seppänen, S-K., Syrjälä, L., Weissenberg, K., Teeri, T.H., Paajanen, L. and Pappinen, A. (2004) Antifungal activity of stilbenes in in vitro bioassays and in transgenic Populus expressing a gene encoding pinosylvin synthase. Plant Cell Reports, 22 (8), 584–593.

Shahpiri, Z., Bahramsoltani, R., Hosein Farzaei, M, Farzaei, F. and Rahimi, R. (2016) Phytochemicals as future drugs for Parkinson's disease: a comprehensive review. Reviews in the Neurosciences, 27(6), 651–668.

Shi, G., Rao, L., Yu, H., Xiang, H., Yang, H. and Ji, R. (2008) Stabilization and encapsulation of photosensitive resveratrol within yeast cell. International Journal of Pharmaceutics, 349, 83–93.

Shibutani, S., Samejima, M and Doi, S. (2004) Effects of stilbenes from bark of Picea glehnii (Sieb. et Zucc) and their related compounds against feeding behaviour of Reticulitermes speratus (Kolbe). Journal of Wood Science, 50, 439–444.

Sigman, M., Barbas, J., Corbett, S., Chen, Y., Ivanov, I. and Dabestani, R. (2001) Photochemical reactions of trans-stilbene and 1,1-diphenylethylene on silica gel: mechanisms of oxidation and dimerization. Journal of Photochemistry and Photobiology A: Chemistry, 138, 269–274.

Simard, F., Legault, J., Lavoie, S., Mshvildadze, V. and Pichette, A. (2008) Isolation and identification of cytotoxic compounds from the wood of Pinus resinosa. Phytotherapy Research, 22 (7), 919–922.

Singh, T. and Singh, A.P. (2012) A review of natural products as wood protectant. Wood Science and Technology, 46, 851–870.

Solhaug, K. (1990) Stilbene glucosides in bark and needles from Picea species. Scandinavian Journal of Forest Research, 5, 59–67.

Soto, M.L., Moure, A., Dominquez, H. and Parajo, J.C. (2011) Recovery, concentration and purification of phenolic compounds by adsorption: A review. Journal of Food Engineering, 105, 1–27.

Stevanato, R., Bertelle, M. and Fabris, S. (2014) Photoprotective characteristics of natural antioxidant polyphenols. Regulatory Toxicology and Pharmacology, 69, 71–77.

Sustainable Growth from Bioeconomy - The Finnish Bioeconomy Strategy (2016) http://biotalous.fi/wp- content/uploads/2014/08/The_Finnish_Bioeconomy_Strategy_110620141. pdf

71 References

Tisserant, L-P., Hubert, J., Lequart, M., Borie, N., Maurin, N., Pilard, S., … Courot, E. (2016) 13C NMR and LC-MS profiling of stilbenes from elicited grapevine hairy root cultures. Journal of Natural Products, 79, 2846–2855.

Trela, B.C. and Waterhouse, A.L. (1996) Resveratrol: isomeric molar absorptivities and stability. Journal of Agricultural and Food Chemistry, 44, 1253–1257.

Tresserra-Rimbau, A., Guasch-Ferre, M., Salas-Salvado, J., Toledo, E., Corella, D., Castañer, O., … Lamuela-Raventós, R.M. (2016) Intake of total polyphenols and some classes of polyphenols is inversely associated with diabetes in elderly people at high cardiovascular disease risk. The Journal of Nutrition, 146, 767–777.

Tříska, J., Vrchotová, N., Olejníčková, J., Jílek, R. and Sotolář, R. (2012) Separation and identification of highly fluorescent compounds derived from trans-resveratrol in the leaves of Vitis vinifera infected by Plasmopara viticola. Molecules, 17 (3), 2773–2783.

Underwood C.D.T. and Pearce R.B. (1991) Astringin and isorhapontin distribution in Sitka spruce trees. Phytochemistry, 30, 2183–2189.

Wadke, N., Kandasamy, D., Vogel, H., Lah, L., Wingfield, B., Paetz, Z., … Hammerbacher, A. (2016) The bark-beetle-associated fungus, Endoconidiophora polonica, Utilizes the phenolic defense compounds of its host as a carbon source. Plant physiology, 171, 914–931.

Waffo-Teguo, P., Hawthorne, M., Cuendet, M., Mérillon, J.M., Kinghorn, A.D., Pezzuto, J.M. and Mehta, R.G. (1998) Potential cancer-chemopreventive activities of wine stilbenoids and flavans extracted from grape (Vitis vinifera) Cell Cultures. Nutrition and Cancer, 40 (2), 173–179.

Wang, H., Liu, L., Guo, Y-X., Dong, Y-S., Zhang, D-J. and Xiu, Z-L. (2007) Biotransformation of piceid in Polygonum cuspidatum to resveratrol by Aspergillus oryzae. Applied Microbiology and Biotechnology, 75, 763–768.

Wang, Q.L., Lin, M. and Liu, G.T. (2001) Antioxidative activity of natural isorhapontigenin. Japanese Journal of Pharmacology, 87, 61–66.

Wang, Z., Zou, J., Cao, K., Hsieh, T.C., Huang, Y. and Wu, J.M. (2005) Dealcoholized red wine containing known amounts of resveratrol suppresses atherosclerosis in hypercholesterolemic rabbits without affecting plasma lipid levels. International Journal of Molecular Medicine, 16, 533–540.

Weissmann, G. (1981) Untersuchung der rindenextrakte von Picea abies Karst. Holz- als Roh-und Werkstoff, 39, 457–461.

Venäläinen, M., Harju, A.M., Saranpää, P., Kainulainen, P., Tiitta, M. and Velling, P. (2004) The concentration of phenolics in brown-rot decay resistant and susceptible Scots pine heartwood. Wood Science and Technology, 38 (2), 109–118.

Wesolowska, O., Wisniewski, J., Bielawska-Pohl, A., Paprocka, M., Duarte, N., Ferreira, M.J., Dus, D. and Michalak, K. (2010) Stilbenes as multidrug resistance modulators and apoptosis inducers in human adenocarcinoma cells. Anticancer Research, 30, 4587–4593.

Viiri, H., Annila, A. and Kitunen, V. (2001) Induced responses in stilbenes and terpenes in fertilized Norway spruce after inoculation with blue-stain fungus Ceratocystis polonica. Trees – Structure and Function, 15, 112–122.

Willför, S., Hemming, J., Reunanen, M. and Holmbom, B. (2003) Phenolic and lipophilic extractives in scots pine knots and stemwood. Holzforschung, 57 (4), 359–372.

Wolfender, J., Rodriguez, S. and Hostettmann, K. (1998). Liquid chromatography coupled to mass spectrometry and nuclear magnetic resonance spectroscopy for the screening of plant constituents. Journal of Chromatography A, 794, 299–316.

Wolter, F., Clausnitzer, A., Akoglu, B. and Stein, J. (2002) Piceatannol, a natural analog of resveratrol, inhibits progression through the S phase of the cell cycle in colorectal cancer cell lines. The Journal of Nutrition, 132, 298– 302.

Woodward, S. and Pearce, R.B. (1988) The role of stilbenes in resistance of Sitka spruce (Picea sitchensis (Bong.) Carr.) to entry of fungal pathogens. Physiological and Molecular Plant Pathology, 33, 127–149.

Välimaa, A.L., Honkalampi-Hämäläinen, U., Pietarinen, S., Willför, S., Holmbom, B. and von Wright, A. (2007) Antimicrobial and cytotoxic knotwood extracts and related pure compounds and their effects on food- associated microorganisms. International Journal of Food Microbiology, 115, 235–243.

Yang, I., Kim, E., Kang, J., Han, H., Sul, S., Park, S. and Kim, S. (2012) Photochemical generation of a new, highly fluorescent compound from non- fluorescent resveratrol. Chemical communications, 48 (32), 3839–3841.

73 References

Yatkin, E., Polari, L., Laajala, T.D., Smeds, A., Eckerman, C., Holmbom, B., Saarinen, N., Aittokallio, T. and Mäkelä, S. (2014) Novel lignan and stilbenoid mixture shows anticarcinogenic efficacy in preclinical PC-3M-luc2 prostate cancer model. PLoS ONE, 9(4): e93764.

Zeneli, G., Krokene, P., Christiansen, E., Krekling, T. and Gershenzon, J. (2006) Methyl jasmonate treatment of mature Norway spruce (Picea abies) trees increases the accumulation of terpenoid resin components and protects against infection by Ceratocystis polonica, a bark beetle-associated fungus. Tree Physiology, 26:977–988.

Zhang, L and Gellerstedt, G. (2008) 2D Heteronuclear (1H-13C) Single Quantum Correlation (HSQC) NMR analysis of Norway spruce bark components. In: Characterization of Lignocellulosic Materials (eds T. Hu), pp 3–16. Blackwell Publishing Ltd.

Zheng, J. and Ramirez, V.D. (2000) Inhibition of mitochondrial proton F0F1- ATPase/ATP synthase by polyphenolic phytochemicals. British Journal of Pharmacology, 130, 1115–1123.