Vicki Huizu Guo Decker
Phenolics, Nitrogen, and Biotic Interactions
— A Study of Phenylpropanoid Metabolites and Gene Expression in the Leaves of Populus tremula.
Vicki Huizu Guo Decker
Fysiologisk Botanik
Umeå universitet, 901 87 Umeå www.umu.se
Umeå 2016
-Anyone who never made a mistake never tried anything new -- Albert Einstein
To my dear family
This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-590-2
Front cover by Vicki Huizu Guo Decker
Elektronisk version tillgänglig på http://umu.diva-portal.org/
Tryck/Printed by: KBC Servicecenter, Umeå University
iii
List of papers
I. Bandau F, Decker VHG, Gundale MJ, Albrectsen BR. 2015. Genotypic Tannin Levels in Populus tremula Impact the Way Nitrogen Enrichment Affects Growth and Allocation Responses for Some Traits and Not for Others. PLoS One 10(10).
II. Decker VGH, Bandau F, Gundale MJ, Cole CT, Albrectsen BR. 2016.
Aspen Phenylpropanoid Genes’ Expression Levels Correlate With Genets’ Tannin Richness and Vary Both in Responses to Soil Nitrogen and Associations With Phenolic Profiles. (accepted for publishing in Tree Physiology)
III. Decker VGH.*, Siddique A.*, Albrectsen BR. 2016. Specialist Leaf Beetles Enrich Diversity of Endophytic Fungi in Aspen Leaves, and Mask Intra-specific Host Specificity (Manuscript)
IV. Decker VGH, Agostinelli M, Chen SS, Cleary M, Witzell J, Albrectsen BR. 2016. Foliar fungal endophytes respond to nitrogen fertilization and herbivory in aspen genotypes (Manuscript)
* the authors contributed equally to the article.
Table of Contents
List of papers ... iii
Abstract ... v
Keywords ... v
Abbreviations ... vi
1. Introduction ... 1
1.1. Plant defense theories and Populus spp. ... 1
1.2. Populus defense-related phenolics ... 2
1.2.1. Condensed tannins... 2
1.2.2. Salicinoids ... 5
1.2.3. The biosynthesis of CTs and salicinoids ... 7
1.3. The interaction of environmental factors and defense phenolics ... 9
1.3.1. Soil nitrogen addition ... 9
1.3.2. Biotic interactions with insect herbivores and pathogens ... 9
1.3.3 Aspen interactions with associated endophytes ... 10
2. Objectives of this thesis ... 12
3. Results and discussion ... 13
3.1 Growth and allocation-related N effects ... 13
3.2. PPP- related N effects ... 15
3.3 Aspen plant- endophyte interactions ... 18
3.3.1. Endophytes association to host genotype ... 18
3.3.2. Endophytes - host genotype specificity disappeared with herbivory... 19
3.3.3. Fungal endophytes association to host genotypes within the TL chemotype21 4. Conclusions ... 24
4. Acknowledgments ... 25
5. References ... 27
Abstract
European aspen (Populus tremula) is a fast growing tree species, rich in phenolic compounds. Defense theories suggest that soil nitrogen greatly influence plant allocation to growth and defense; however, the allocation priorities are not well understood. Further, although foliar phenolic compounds are considered defensive, specialist organisms may positively associate with and alter them. There are two classes of phenolics in aspen, condensed tannins (CTs) and salicinoids. They are likely to shape the interactions of the many organisms, for example, herbivorous insects and endophytic fungi and three-way interactions among host genotype, specialist herbivore and endophytic fungi relationships could be greatly altered by aspen geno- and chemotypes’ responses to soil nitrogen.
Firstly, I focused on the allocation of carbon to growth and defense in aspen genotypes with varied tannin content in response to nutrient addition. Nitrogen promoted plant growth and suppressed foliar CT levels. At the molecular level expression of genes of the phenylpropanoid pathway (PPP) decreased under low additions of N (equivalent to 15 kg/ha), whereas genes at the beginning and at the end of the pathway increased in response to high levels of N (~150 kg/ha). Aspens high in CTs displayed consistently stronger PPP gene expressions compared to CT-low aspens, and correlations between PPP genes and phenolic products varied with tannin content, as an effect of leaf age, in response to N enrichment, and individually with genotype. More negative correlations (indicative of allocation trade-offs) between PPP gene expressions and phenolic products were found in aspen genets with low tannin levels compared to aspens with inherently high tannin levels.
Secondly, I studied the connection between foliar phenolic compounds and endophytic fungi in the presence and absence of a specialist herbivorous beetle (Chrysomela tremula) and as an effect of soil nitrogen addition. Richness and abundance of fungal endophytes associated with aspen genotypes and phenolic profile, however this specificity disappeared in the presence of the leaf beetles. Herbivory both enhanced endophyte richness and abundance in the leaves and it also increased in response to nitrogen addition..
Keywords: Populus tremula, condensed tannins, salicinoids, phenolics, genetic variation, soil nitrogen fertilization, herbivory
Abbreviations
CT: Condensed tannins
PPP: phenylpropanoid biosynthetic pathway A: simulated ambient condition
D: simulated depositional condition F: simulated fertilizational condition N: nitrogen fertilization
H: herbivore insects
NH: combined treatment of nitrogen fertlization and herbivore insects PAL, Phenylalanine ammonia lyase;
C4H, cinnamate 4-hydroxylase;
4CL, 4-coumarate:CoA ligase;
CHS, chalcone synthase;
CHI, chalcone isomerase;
F3H, flavanone 3-hydroxylase;
DFR, dihydroflavonol reductase;
ANS, anthocyanidin synthase;
LAR, leucoanthocyanindin reductase;
ANR, anthocyanidin reductase;
1
1. Introduction
1.1. Plant defense theories and Populus spp.
Forests cover over a quarter of the terrestrial surface on the Earth and provide essential elements to humanity like clean air, water, and resources for food and energies. Besides providing livelihoods for humans and habitats for animals, forests also offer such as watershed protection, prevent soil erosion and mitigate climate change. During millions of years’ evolution, long-lived trees can accumulate relatively high levels of secondary metabolites mainly carbon-based phenolics, which are related to their defense.
There are two most prominent theories to explain nutrient allocations to defense-related secondary metabolites: the growth differentiation balance hypothesis (GDBH, Herms and Mattson 1992), and the protein competition model (PCM, Jones and Hartley 1999). GDBH states in nutrient-limited conditions, plant growth is limited and the carbon-resource surplus from the photosynthesis can be diverted to produce such as the defense phenolics. In PCM, the defense phenolics and growth- required proteins compete for a common precursor e.g. phenylalanine. As a consequence, when nutrients are limiting, resource partitioning is in favor to the phenolic accumulation.
Poplar (Populus) species are widely distributed and abundant in the northern hemisphere. They are valued as pioneer species habituating many species of dependent flora and fauna. Due to characteristic growth properties, fully-sequenced genome with a relatively small size (comparing to conifers), and easy to be propagated and manipulated at the lab, poplars are used as model species to study defense mechanisms in perennial tree, as well as interactions of host-microorganisms and host- herbivores.
Aspen (P. tremula, Euroasian aspen) shares many of the same ecological, morphological and physiological traits with its closely related sister species quaking aspen (P. tremuloides, e.g. Lindroth et al., Whitham et al., Robinson et al., 2012), although differences exist for example realted to evolutionary history (De Carvalho et al., 2010; Bernhardsson & Ingvarsson, 2012; Cole et al., 2016).
P. tremula has its own ecological value as well. For example, currently forestry practice in Scandinavia retains aspens in clear-cut zones to promote biodiversity (Gustafsson et al., 2013), as aspen is used for a common forage resource providing food for insects and mammals, and habituating a suite of epiphytes, endophytes and arthropods ((Kuusinen, 1996; Suominen et al., 2003; Hedenas et al., 2007; Myking et al., 2011).
The establishment of the Swedish Aspen Collection (SwAsp, 2004) provided a platform for such studies (Luquez et al., 2008). For example, it provides plant materials to study aspen genetic variation in relationship to pathogen, insect herbivores, and arthropod activities, along geographic structures (Albrectsen et al., 2010b, Robinson et al., 2012), from the genetic structure in their defense genes to metabolome (Bernhardsson & Ingvarsson, 2012; Bernhardsson et al., 2013), and genotype-based plant chemistry within P. tremula (Keefover-Ring et al., 2014).
1.2. Populus defense-related phenolics
The genus Populus shows diverse genetic variations in defense-related phenolics, which are also affected by environmental factors such as light intensity and herbivore stresses to varied degrees (Philippe & Bohlmann, 2007; Chen et al., 2009; Boeckler et al., 2011). Among those phenolics, condensed tannins (CTs, proanthocyanidins) and salicinoids are of great importance. Together CTs and salicinoids account for a large portion of dry mass across Salicaceae species. for example, salicinoids and CTs can comprise up to 30 % of plant dry mass in field-grown aspens, up to 15.8% in hybrid willow seedlings, 20% in field mature willow trees, 12% in outdoor hybrid aspen saplings, and 35% in cottonwoods (Hansen et al., 2006; Rehill et al., 2006; Häikiö et al., 2009; Orians et al., 2010;
Robinson et al., 2012). Defense-related phenolics have been intensively studied with regards to their bioactivities, in order to explain outbreaks or occurrence of a range of herbivores (Donaldson et al., 2006), as well as associations between host trees and microorganisms (Bailey et al., 2005; Albrectsen et al., 2010a).
1.2.1. Condensed tannins
Condensed tannins (proanthocyanidins) are the most common group of tannins (Salminen & Karonen, 2011) and flavan-3-ols are the basic building units of CTs (Barbehenn & Peter Constabel, 2011). In plants, CTs are formed as oligomers (two to ten monomer units) or polymers (>10monomer units) (Salminen & Karonen, 2011). Common types of CTs are procyanidins (PC) and prodelphinidins (PD), which are consist of monomeric (+)-catechin or (-)-epicatechin units, and (+)-gallocatechin or (-)- epigallocatechin units, respectively (Fig. 1). The differences in the ratio of PD and PC subunits, degree of polymerization and configuration of the polymers determine the bioactivity of the compounds (Ayres et al., 1997). CTs are ubiquitously distributed over plant tissues (from root to shoot). Within leaves CTs are commonly found in the epidermal or sub-epidermal layer (Kao et al., 2002; Dixon et al., 2005; Lepiniec et al., 2006; Randriamanana et al., 2014).
3 The biochemical activities of tannins range from beneficial antioxidants to damaging prooxidants and toxins. Tannins act as antioxidants by scavenging free radicals, or reduce other oxidized compounds, and form relatively stable semiquinone radicals, although few studies have demonstrated antioxidant effects of tannins in insect herbivores (Barbehenn & Constabel, 2011).
Fig. 1. Structures of the common (a–d) monomeric building units of condensed tannins, and an example of the polymeric condensed tannins (right) with molecular mass of over 23 900 g mol-1. Figure is adapted from Salminen and Karonen (2011).
It is still debating where CTs are localized, and it is largely unknown how they are transported and polymerized, especially in tree species. For herbaceous plant like Arabidopsis, it is generally believed that CTs are localized to cell walls in the coats of mature seeds. In the nuclei of seed coat endothelial cells, CT biosynthesis genes are activated by CT regulatory transcription factors and CT biosynthesis proteins are translocated to the cytosolic side of the ER for synthesis of epicatechin and anthocyanins.
Later these CT building units (epicatechin and anthocyanins) are transported into the vacuole (the main storage place of CTs), ER, or other cell compartments by means of binding to CT transporters, through membrane vesicle trafficking or by glycosylation. The condensation of CT biosynthetic units may happen in vacuoles through enzymatic process or facilitated by the acidic vacuolar conditions (Fig. 2, Zhao et al., 2010). Recently, it has been reported that tannin polymerization occurs inside the tannosome, a chloroplast-derived organelle, in vascular plants (Brillouet et al., 2013). The tannosome shuttle transfers through the cytoplasm towards the vacuole where it binds to the tonoplast. The incorporated shuttle-tonoplast then aggregates into tannin accretions that are stored in the vacuole, facilitated by various cross-membrane transporters (Fig. 3, Brillouet et al., 2014a).
Fig. 2. Model for the transport and polymerization of condensed tannins in a plant cell. The synthetic enzymes translocated to the cytosolic side of ER for the biosynthesis of CT monomers. CT monomers then are transported to vacuole or other compartment by transporter proteins, vesicle traffiking, or glycosylation. CTs are polymerized in the vacuole.
Zhao et al. (2010).
Fig. 3. An illustration of tannin polymerization (in the tennosomes, arrowheads), transport (tannin shuttles, sh), and its accretion (ta) in the vacuole (Brillouet et al., 2013)
5 1.2.2. Salicinoids
Salicinoids are phenolic glycosides that have long been known for their pharmaceutical importance as anti-inflammatory agents. Nowadays, salicinoids have also received extensive attention for their ecological roles in many plant–herbivore studies conducted on Salicaceae species, where they have been implicated as toxins and deterrents to a number of insect and mammalian herbivore species (see e.g. Boeckler et al., 2011). Poplar, aspen, and willow species share many salicinoid compounds, however their chemistry profiles can also be used to distinguish between Salicaceous trees, both inter- (Orians & Fritz, 1995; Caseys et al., 2012) and intra- specifically (Lindroth & Hwang, 1996b; Abreu et al., 2011; Keefover-Ring, et al., 2014b).
Aspen (P. tremula) displays an unexpected diversity of salicinoids, though this group consisting of barely more than 20 compounds, ranging from simple structured salicin to higher order compounds such as 2’O-cinnamoyl-salicortin (Abreu et al., 2011; Keefover-Ring et al., 2014, Fig. 4). In addition to salicortin, tremulacin, salicin, and tremuloidin (the signature salicinoids of P. tremuloides, Lindroth and Hwang 1996), some salicinids in P. tremula were previously reported only for Salix spp.
(Nichols-Orians et al., 1992; Ruuhola & Julkunen-Tiitto, 2003).
Fig. 4. Structural relationship of 19 salicinoids found in the foliage of to the loading plot (Fig. 2b). 1= new compounds for P. tremula, 2= compounds with two isomers present but the conformation of the P. tremula from the SwAsp collection grouped similar cinnamoyl group double bond is ambiguous. 3=2’-(E)- and 2’-(Z)-cinnamoylsalicortin (Keefover-Ring et al., 2014).
The core structure of the salicinoid compounds consists of simple phenolics (such as salicyl alcohol) and β-D-glucopyranose moieties, with an ether linkage between the phenolic hydroxyl group and the anomeric C- atom of the glucose. The complex salicinoids are not stable if not handled appropriately because their ester bonds are susceptible to chemical and enzymatic hydrolysis (Lindroth and Pajutee, 1987), or if the metabolic breakdown occurs. The latter contributes to the toxicity of salicinoids (Clausen et al., 1989). As a consequence, molecules of those complex salicinoids will break down to simpler ones, and mostly the simplest salicin. However, due to lack of free phenol groups or free electrons, unlike flavonoids or CTs in the Salicaceae, most salicinoids cannot undergo typical anti- oxidative reactions. Despite the intensive focus on salicinoids, no genes or enzymes involved in salicinoid biosynthesis have been identified (Boeckler et al., 2011), although in vivo isotope-labeled and cell culture feeding experiments of salicinoid precursors provided information of salicinoid formation. In cell cultures, the glycosyltransferases may function in glycosylation of simple phenolics, and protein transporters that localized in tonoplast may facilitate such glycosides into the vacuole (Payyavula et al., 2009; Babst et al., 2010). Recently, a study on vanilla fruit suggested a phenyl glycoside accumulated in the phenyloplast, a newly found chloroplast-derived organelle formed in the inner volume of chloroplasts, after chloroplast generate loculi between thylakoid membranes filled with the glycoside (Brillouet et al., 2014b). However we don’t know whether salicinoids may accrete in the same organelle.
Salicinoids divide P. tremula into three chemical phenotypes (hereafter chemotypes, Abreu et al., 2011, Keefover-Ring et al., 2014). These chemotypes were mainly defined by the presence of high amounts of salicinoids with either cinnamoyl moieties (CN), 2’-acetyl moieties (AC), or very low amounts of either (TL, Fig. 5 ).
7 Fig. 5. Characterization of the Swedish Aspen Collection (SwAsp) based on their salicinoid profiles. Figure adapted from Keefover-Ring et al., 2014.
Salicinoids have been reported to have negative impact on the performance of many generalist herbivores (e.g. Hemming and Lindroth, 2000; Osier and Lindroth, 2001; Albrectsen et al., 2010a, Robinson et al., 2012). However, as for some of the specialist herbivores, salicinoids are not so
“toxic”, and they can even sequester salicinoids for their own defense, such as the leaf beetle Chrysomela tremula (Axelsson et al., 2011; Hjalten et al., 2012). The different reaction of herbivores towards salicinoids indicates salicinoids play double- role in mediating plant–herbivore interactions.
1.2.3. The biosynthesis of CTs and salicinoids
Both salicinoids and CTs are derivatives from the phenylpropanoid pathway (PPP, Fig 6, Constabel &
Lindroth, 2010). The entry-point enzyme in PPP is L-phenylalanine ammonia-lyase (PAL) catalyzing phenylalanine, which is the end derivative from the primary Shikimic pathway. The PPP hydroxycinnamate-derived monolignols are essential for cell wall formation. Chalcone synthases (CHSs) are type III polyketide synthases with two independent active sites that catalyze a series of decarboxylation, condensation and cyclization reactions. Chalcone isomerases (CHIs) catalyze the stereospecific isomerization of chalcones into their corresponding flavanones. The CT branch starts with the conversion of flavanones to dihydroflavonols, catalyzed by ‘flavone 3-hydroxylase’ (F3H), a plant membrane-associated cytochrome P450 enzyme. Hydroxylation of flavonoid skeletons by NADPH-dependent cytochrome P450 monooxygenases is important in the biosynthesis of complex flavonoids, as the prenylation enhances flavonoids’ lipophilicity and membrane permeability which sequentially increased their antibacterial and antifungal activities (Sohn et al., 2004). Anthocyanidin synthases (ANSs), are members of the family of 2-oxoglutarate-iron-dependent oxygenases
(Naoumkina et al., 2010) and catalyze the oxidation of the flavonoid ‘C-ring’ (Naoumkina et al., 2010).
Fig. 6. The scheme of phenylpropanoid biosynthetic pathway of Populus, based on Lindroth and Constabel 2010. PAL, Phenylalanine ammonia lyase; C4H, cinnamate 4-hydroxylase;
4CL, 4-coumarate:CoA ligase; CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone 3-hydroxylase; DFR, dihydroflavonol reductase; ANS, anthocyanidin synthase;
LAR, leucoanthocyanindin reductase; ANR, anthocyanidin reductase.
Phenolic compounds are synthesized by all plants, and the genome-wide characterization and analysis of PPP genes regulating defense metabolism have been reviewed from various perspectives in different plant species (Tsai et al., 2006; Hamberger et al., 2007; Chen et al., 2009; Hao et al., 2011).
Compared to the annual herbaceous plant Arabidopsis, Populus spp. produce more abundant phenolic secondary metabolites with more diverse structures, related to defense, which agrees with the fact that the gene families involved in the downstream CT metabolism are much larger in Populus, although genes responsible for PPP upstream steps are much more conserved (Tsai et al., 2006). With the high abundance and diversification of phenolic compounds in all tissues typically for the Salicaceous, the PPP appears to play a major role in chemical ecology across this family.
9 1.3. The interaction of environmental factors and defense phenolics
1.3.1. Soil nitrogen addition
According to plant defense theories, resource availability determines relative allocation between growth and defense. Both GDBH or PCM state that the production of defense phenolics is in favor when resource is limiting, and growth is in favor when it comes to resource-rich conditions. Therefore the effect of soil nitrogen addition is one of the most crucial components in understanding how plants determine their allocation to defense.
Diverse N forms exist in the soil, ranging from simple inorganic N compounds such as NH4+
and NO3-
to polymeric N forms such as proteins (Paungfoo-Lonhienne et al., 2008). In Sweden, the annual N deposition amounts are 1-3 kg N ha-1 y-1 in the north and 10-15 kg N ha-1 y-1 in the south, respectively, mainly due to anthropogenic activities (Gundale et al., 2011). In Northern Sweden, the usual dose of N applied today during a fertilization event is up to 150 kg N ha-1 y-1 (Hedwall et al., 2013). Although organic N also affects plant growth and ecosystem process, in this thesis I only focus on the inorganic N as it is commonly used as chemical fertilizer.
Soil nutrient addition affects plant performance by affecting plant functional traits related to growth and defense. It has been well known that fertilization promotes plant growth. As for plant defense- related chemical traits such as salicinoids and CTs, soil nitrogen may affect the C partitioning between them, and the effect depends on genetic factors. For example, a study on two P. fremontii × angustifolia backcross lines with contrasting rates of CT accrual reported that in nitrogen-limited condition, salicinoids increased only in the low-CT, fast-growth line (FG), but not in high-CT, slow- growth line (SG). This study further suggests accumulation of starch in source leaves and CTs in sink leaves of SG may both contribute to the adaptation of slow-growth line in nutrient-poor habitats (Harding et al., 2013).
1.3.2. Biotic interactions with insect herbivores and pathogens
In their natural habitats, Salicaceae encounter a variety of mammalian herbivores browsing bark and shoots, especially in winter when other food sources are scarce. Populus also hosts various types of insect herbivores, including leaf feeders like Lymantria dispar, boring insects like Saperda calcarata, and sap-sucking insects like Philaenus spumarius (Chen et al., 2009, Boeckler et al., 2011). In addition, phytopathogens infect Populus including fungi such as Melampsora species, bacteria such as Xanthomonas popular), and viruses such as Populus mosaic virus (Chen et al., 2009).
Tannins protect plants from generalist insects as feeding deterrents and high concentration of tannins is responsible for the negative effects of tannins on insects. They function as deterrents to herbivores by co-precipitating with proteins to make plant biomass less digestible (Stevens & Lindroth, 2005;
Barbehenn & Constabel, 2011). In contrast, for specialist insects, tannins can actually act as feeding stimulants (Schultz, 1989), although few dose–response studies have been conducted about the effect of foliar tannins on insect herbivores (Barbehenn & Constabel, 2011). In fact, CTs can have little impact on insect herbivore performance as well. For example, they are neither deterrents nor toxins to L. dispar on aspen genotypes with varied CT concentration (Osier et al., 2000). Levels of CTs may also relate to pathogen infection. For example, the incidence of Venturia moreletti is negatively associated with the constitutive concentration of CTs in P. tremuloides (Holeski et al., 2009), and M.
medusae induces CT accumulation in the infected poplar leaves (Miranda et al., 2007).
Salicinoids increase in toxicity with greater molecular complexity, and certain salicinoids may also contribute to polyphenol oxidase-based defenses of poplars (Lindroth & Weisbrod, 1991; Haruta et al., 2001; Constabel & Lindroth, 2010). Salicinoids in high concentration result in the delayed development and reduced weights of non-adapted insects (Osier and Lindroth, 2006), whereas salicinoid-adapted species can sequester salicinoid compounds for their own defense (Lindroth &
Pajutee, 1987). For example, leaf beetles (Chrysomelidae) are capable of converting salicin to salicylaldehyde, which is then stored in glandular reservoirs and secreted upon attack by natural enemies like predators ants and shown antimicrobial properties (Pasteels, 1993; Opitz & Müller, 2009). As for phytopathogens, salicinoids can inhibit fungal spore germination (Flores & Hubbes, 1979; Boeckler et al., 2011).
Genetic variation in defense chemistry influences herbivore performance in aspen, such as the variation of CTs among SwAsp genotypes is important in shaping the composition of arthropod community (Robinson et al., 2012). Foliar CTs also influence the distribution of V. moreletii and the CT effect varied among quaking aspen (P. tremuloides) genotypes in a field study (Holeski et al., 2009). Genotypic specific salicinoids are also responsible for the resistance of quaking aspen to gypsy moth (L. dispar).
1.3.3 Aspen interactions with associated endophytes
All plants appear to host fungal endophytes (Rodriguez et al., 2009). Although endophytes are asymptomatic, they can act in plants as pathogens, neutral’, or mutualists. When they act as mutualists, their presence increase host fitness by improving plant growth, pest resistance, and/or
11 negatively affect plant fitness (Ridout & Newcombe, 2015). Unlike grass endophytes, tree endophytes are horizontally transmitted (Arnold 2007). Foliar endophytes exhibit various levels of diversity (e.g.
reviewed in Lamit et al. 2014), and endophyte communities are varied depending on spatial scales, inter- and intra-specifically (Zimmerman & Vitousek, 2012), and plant sexes (Elamo et al., 1999).
There is a dependency between host plants and endophytes, which affects fitness of host plants and thus they appear to be co-evolving (Higgins et al., 2007).
Phytochemistry is suggested to have negative correlation to the infection of twig fungal endophytes in hybrid cottonwoods (Bailey et al., 2005), while it is still largely unclear which plant traits and to which extend they may mediate the aspen- endophytes interaction. I asked whether the defense phenolics are responsible for the specificity of aspen genotype and fungal endophyte composition.
Thus, the assessment of such chemical traits will help to understand endophyte distributions, and modulation of endophyte species with tree genotype.
2. Objectives of this thesis
I studied PPP related effects of N, and consenquent effects on associated endophyte community, the richness and abundance. In order to achieve the goals, in the first two studies I selected aspen genotypes include CT-high and CT-low groups, which are characterized by their intrinsic ability of CT production; for the last two questions, I concentrated mainly on salicinoid chemotype, which are characterized by their salicinoid profile, and its relationship to endophyte communities. In four papers (thesis chapter I-IV), I asked the following questions:
Does CT-high and CT-low aspens differently respond to soil nitrogen addition, regarding to plant nutrient allocation between defense and growth? (Thesis chapter I)
Does CT-high and CT-low aspens differently respond to soil nitrogen addition, regarding to the gene expression and metabolism of the phenylpropanoid pathway? (Thesis chapter II)
Does host genotypic variation of foliar salicinoids affect aspen leaf fungal endophyte community, in the presence and absence of insect herbivores? (Thesis chapter III)
Does soil nitrogen fertilization affect aspen leaf fungal endophyte community, in the presence and absence of insect herbivores? Which chemical traits mediate the interactions between the host plant, insect herbivores, and the endophyte community? (Thesis chapter IV)
13
3. Results and discussion
- The variation of plant growth and defense related traits in young aspen plants were tested in response to two nitrogen addition levels, corresponding to atmospheric N deposition and industrial forest fertilization and the results are described in Paper I.
- In Paper II, PPP related N effects were tested, by the assessment of relative expression levels of PPP genes, and the correlation of the PPP gene expression and levels of PPP derived phenolics, in young and mature aspen leaves in response to the same nitrogen conditions as in Paper I.
- Besides, I studied aspen- leaf endophyte interactions in Paper III and IV. In these two papers, the ecological roles of host genetic variation in defense phenolics were studied, and the influences of environmental (both biotic and abiotic) factors in shaping foliar endophyte community were examined.
3.1 Growth and allocation-related N effects
Nitrogen addition promoted plant to grow more and produce less PPP phenolics (Fig. 7). Growth related traits including relative growth rate (RGR) and plant height that positively responded to the high dose of nitrogen addition, but not in the depositional N condition. CTs decreased with elevated soil N input, so did total C:N ratio. In addition to that, aspen growth-related and tissue chemistry traits varied in responses to soil nitrogen addition (Table 1).
Fig. 7. Plant growth traits and chemistry traits for high and low tannin aspens under three nitrogen conditions. White and hatched bars referred respectively to low and high tannin aspens. Trait values are the means (± s.e.) of pooled data. Different letters indicate Tukey’s test differences (α=0.05).
The results showed that plant responses to N were often weaker or sometimes absent when aspens grew in deposition N treatment, but effects reappeared in the fertilization condition. An overall negative relationship between foliar condensed tannins and plant growth indicates a trade-off between growth and defense, which supports plant defense hypotheses, and confirms results from several other studies (Osier & Lindroth, 2001; Stamp, 2003; Donaldson et al., 2006; Glynn et al., 2007)
15 Table 1. ANOVA summary of aspen growth and tissue chemistry traits of P. tremula in response to nitrogen (N), genotype (GT), and tannin group (Tgr). Df = 150, F-values are followed by significance levels: ***<0.001<**<0.01<*<0.05.
3.2. PPP- related N effects
Soil nitrogen enrichment also affected the Phenylpropanoid biosynthesis pathway (PPP), by affecting the PPP gene expression and PPP phenolic concentrations (Table 2). When N addition simulated the atmospheric deposition condition, PPP gene expression reduced. High N doses on the other hand restored the expression of genes at the beginning and end of the PPP pathway. The relative expression levels of PPP genes were consistently stronger in high CT aspens (Fig. 8). However, the changes in PPP gene expression were not reflected in pools of tannin precursors or final tannin products, and correlations between PPP genes and phenolic products also varied between young and mature leaves (Table 2, Fig. 8). With increased nitrogen, more negative correlations (indicative of allocation trade- offs) were detected between the PPP gene expression and phenolic products in low CT aspens than in high- CT ones (Fig. 8). This tannin-related expression dynamics suggests that, in addition to defense, relative tannin levels may also be indicative of intraspecific variations in the way aspen genets respond to soil fertility.
Table 2. Summary table of ANOVA models of effects on PPP gene expresison and levels of pooled phenolics in aspen leaves of intrinsic tannin production (CTgrp) and nitrogen addition (Ntrt) when adjusted for within-tannin group genet differences (GT[CTgrp]). *** p < 0.001; **
p < 0.01; * p < 0.05; n.s. p > 0.05.
The different responses of tannin groups to nitrogen addition might suggest a varied carbon partitioning in response to N addition between high and low CT plants, and spatial and temporal differences might affect the coordination of PPP genes to the PPP phenolics. The leaf sink strength of CT accrual might drive the separation of these two tannin groups, as described by Harding et al., (2013).
In this project, both young and mature leaves were selected for different purposes. Young leaves are frequently analyzed for the PPP gene expression (e.g. Tsai et al., 2006; Sjödin et al., 2009), and mature leaves are commonly used in plant physiology and chemical ecology studies (e.g. Orians and Fritz 1995; Ruuhola and Julkunen- Tiitto 2000; Cole et al., 2016). Thus the phenolic analyses on both young and mature leaves provided a chance to compare pooled foliar phenolics and search for signs of limiting factors or allocation trade-offs.
17 Fig. 8. The relationship of PPP genes and metabolic pools across N treatment between young and mature aspen leaves from high and low tannin genets. Upper panel: the relative expression of key genes in PPP and the Anova summary of individual genes in response to nitrogen (Ntrt), tannin group (CTgrp), and the nested genotypic (GT[CTgrp]) effects. Lower panel: the correlations of genes and selected PPP phenolic pools.
The results of high and low tannin genotypes differed in the responses to soil N enrichment suggested that aspens might evolve to different strategies to adapt to the environment. When nutrient is limited, CT productions become costly so low tannin plants could prioritize growth. However damages by pathogens and herbivores is more likely to occur in poor soil nutrient conditions, and in this case high tannin genotypes may gain greater benefit (Albrectsen et al., 2009; Holeski et al., 2009; Robinson, K.
M. et al., 2012). This study illustrated how plants could respond to genetic variability, which providing several insights into understanding how plants respond to environmental change.
.
3.3 Aspen plant- endophyte interactions
Fungal endophytes ubiquitously exist in plants with high diverse and unknown ecological functions, which marks them important “hidden players” in host-environment interactions. In this thesis, I compare the diversity, abundance, and competitive interactions of culturable endophytic fungi in Aspen leaves (P. tremula), in the presence and absence of a specialist herbivore (Chrysomela tremula), and use “endophyte study I” referring to the results. I further compare fungal endophyte communities from plants that received nitrogen fertilization, herbivore, and the two combined treatments. I use “endophyte study II” to distinguish from the former part.
3.3.1. Endophytes association to host genotype
The fungal isolates were strongly related to salicinoid composition and to genotypic identity. Most of the isolates were associated with tremuloides-like (TL) chemotype (Fig. 9), whereas the highest abundance was associated with acetyl-salicinoid chemotype (Fig. 9). The salicinoid structural complexity might determine the chemotype -- morphotype associations in aspen, because moieties defining salicinoids determined their toxicity and palatability (Lindroth et al., 1987; Reichardt et al., 1990; Boeckler et al., 2011; Keefover-Ring, Ken et al., 2014). This study might be the first to report the interactive relationships between endophyte and host salicinoid profile in aspen.
19 Fig. 9. Relationship between generic salicinoid content and the number of morphotypes (fungal richness) isolated from a genets. Upper panel: The generic salicinoid values are summed information from Table 1. Chemotype groups are indicated by symbols: Diamonds = TL, Triangles= AC, Circles=CIN. Lower panel:
The Aspen (P. tremula) genets of this study represent the three main chemotype groups detected in the Swedish Aspen collection, SwAsp based on salicinoid content.
3.3.2. Endophytes - host genotype specificity disappeared with herbivory
Bipartite graphs were used to visualize interactions between the endophyte composition and host plants. Compared to control plants (-beetles), the presence of herbivores resulted in greatly elevated numbers of endophyte morphotypes, and highly enhanced interactions between host plants and endophyte communities (Fig. 10). This increase in endophyte richness and abundance had been supported by the results of Simpsons index: 0.618 and 0.563 respective for control plant genotype and chemotype, 0.339 and 0.326 respective for genotype and chemotype of beetle damaged plants. In
addition, the strong relationship of endophyte and genotype (chemotype) specificity disappeared in the presence of beetles (Fig. 10).
Fig. 10. Bipartite graphs of relationships between the experimental SwAsp genets and the endophyte community that they were associated with, respectively in absence (left panel) and presence (right panel) of Chrysomela tremula leaf beetles. The host plants are represented by genotype (above) or by chemotype group (below): AC= acetyl;
CN=cinnamoyl-; and TL= tremuloides-like. Thicknesses of lines that connect genets (or chemotype groups) with morphotypes indicate the abundance with a morphotype occurred in the samples.
Quite a few studies found herbivore activities increased endophyte fungal infections for varied tree species. For example, Faeth and Hammon (1996) found the amount of mining activity by sedentary insect herbivores on oak leaves was positively associated with increased colonization by fungal endophytes. They suggested that mining activity might facilitate fungal spore germination and hyphal penetration into the leaf or by altering leaf phytochemistry. Albrectsen et al. (2010a) fount the presence of endophytic fungi in Aspen clones was related to field damage by herbivores and the pathogen Venturia tremula. A negative association was found in two separate surveys between Aureobasidium sp. and herbivore damage, but no evidence that endophyte presence was related to a history of Venturia symptoms.
21 3.3.3. Fungal endophytes association to host genotypes within the TL chemotype
As results from paper III (section 3.3.2 in this thesis) suggested, leaf endophytes showed specific association to varied salicinoid chemotype. Therefore, aspens within the same TL chemotype were expected to remove the salicinoid effect, which might result in the appearance of other interactive responses.
In total 30 morphotypes were isolated from leaves of control plants, plants that received herbivory, N fertilization, and the combined treatments (Table 3). The nitrogen enrichment and herbivore presence both increased the number of morphotypes (Table 3). The morphotypes as their presence were further divided to common morphs that present across treatment and rare morphs that only associated to certain treatments.
Table 3. The abundance and frequency of endophyte morphotypes listed according to treatment (H=herbivory;N=N addition; HN= combined treatment; C=control).
The network between all morph endophytes and host plants were interactive across treatment (Fig. 11, all morphotypes). However the network of treatment-specific rare morphotypes and host plants showed the modulation of host plants and certain endophytes. In addition, N, H and NH treatment enhanced interactions of the endophyte composition and host plants, especially in N treated plants there was the most interactive relationship of the network (Fig. 11, rare morphotypes).
Fig. 11. Bipartite network graphs showing links between aspen genotype and associated endophytic morphs under four experimental settings: left for all morphs and right for the remaining morphs when 11 common morphs had been excluded. The links between aspen genotype and the endophytic morph that were isolated from its leaves are indicated with lines and their thickness corresponds to abundance of the endophyte. The length of the bars indicates relative differences in genotype- endophyte associations.
23 Multivariate analyses of foliar metabolites showed amino acids and total phenolics were two groups strongly associated to fertilized and non-fertilized plants. Mixed linear models revealed that total phenolics involved in shaping the endophyte composition together with herbivory, nitrogen, and their interactive effects (Table 4).
Table 4. Responses of fungal endophytes to plant traits and treatments using mixed linear models. Df(n, dn) for all morphotypes model: 30, 49 and rare morphotypes: 19, 60; Residuals:
78. H: herbivory, N: nitrogen fertilization, TP: total phenolics.
These results suggested that the effects of herbivory and nitrogen fertilization in aspen leaf endophytes composition could be mediated by defense-related phenolics. Out of the expectation, CTs, of the suggested anti-fungal activities, did not influence endophyte composition, although levels of CTs were higher in non-fertilized plants. Furthermore, as only in rare morphs increased interactions were found in plants, it might indicate common morphotypes that distributed in all treatment could mask the host specificity in response to herbivores and nitrogen addition.
4. Conclusions
To conclude, PPP related N effects play important roles in aspen growth and nutrient allocation to defense related phenolics. The N enrichment not only affects the PPP derivatives, but also at the transcriptional level of PPP genes. The correlations of PPP gene expression and PPP metabolism varied so as the relationship of CT products and CT precursors between these two groups, which might suggest high and low tannin plants had different rate of substrate flux of PPP, related to the difference in sink strength of CT accrual between these two groups. All in all it provides insights how aspen plants adapt to different environmental surroundings.
To conclude for this study, the endophyte network becomes more complex in response to beetle activity and possible fertilization regimes. The association specificity between host plants and endophyte communities disappears in the presence of herbivores meanwhile the interactions of endophyte community with host plants became stronger. It suggests complex and dynamic three-way interactions and further suggests the study of endophytes should consider the specific environmental conditions. Foliar defense phenolics could mediate the interactions of endophyte community to hosts in varied abiotic and biotic environmental conditions.
Words by the end
Like in Kungfu Panda movies, A-Po realized the true meaning of martial arts was not only fist Kungfu, it was the harmony and balance in the spirit, which helped him to defeat enemies. There seems be some similarity with studying plant responses to environmental factors -- I realized actually plants do the same as A-Po, although so far no detectable method to know how they think – if they think of course. Using so complex and delicate mechanisms, plants always try to optimize themselves to a dynamic balance between defense and growth, as individuals, or with their associates. This is my, as a philosophy doctor candidate, contribution to philosophy
25
4. Acknowledgments
I would like to give my deepest gratitude to my principle supervisor Benedicte Albrectsen. Benedicte, beyond as my mentor and guide me into the fantastic research world hand by hand, I learned way more from you and I have to say it is my luck to be your student. I admire your wisdom and sharp views in both scientific work and daily life. I tried to follow your pedagogy and by the end I was nominated as “good teacher” by my students. You are always so supportive whenever and no matter what problems I consulted you, shared my happiness and my sorrow. I can’t forget my first cross- country skiing tour with you, and so as you invited me to concert and social events… It was because of you that I could focus and enjoy research! I am so thankful for your accompany during the past six years and I wish we could find more collaboration opportunities in the future.
I like to thank my co-supervisor Stefan Jansson, for your time and effort (and importantly the financial support) in helping me with my project, writing articles and for the good personal discussions we had. I would like to thank Dr. Johanna Witzell, Dr. Chris Cole, Dr. Kathryn Robinson, Dr. Carolina B., and Dr. Ken K. (KKR), for all your professional help in the lab and giving me nice advices in my presentations, posters, and article writing. Dr. Michael Gundale, Dr. Michelle Cleary, Abu, Sylvia and Marta A., thank you for the nice input of our manuscripts. I would also like to mention Slim and my reference group members Xiao-Ru, Totte, and Nat, who guided me through my PhD education during those years. All your evaluations and suggestions in our annual meetings brought me to the final step of my study. I also like to give my special thanks to my project students during the past years, for your help in the fungal isolation and plant sample preparation, and UPSC personnel especially Janne, Rebecca and previously Jenny and Karin, for helping me to sort out all the regulations and forms necessary with topics ranging from maternity leave to my contract prolongation.
Franziska, you and I have been closely worked together for nearly six years, and there were so many great things happened that we shared together. I remembered our first semi-field experiment with I propagated all aspens alone in the noisy tissue culture room and you the whole summer maintenance under scorching sun. I am glad that we studied this project from different aspects and because of that, I broadened my view in plant responses to environmental conditions and it is hard without your collaboration! You are the first author of the first article in my thesis and it makes the thesis more completed. And of course, you are one of my best friends. Melis K., Erik E., Christoffer J., Ionna A., Karen K., your friendship means so much to me! I am so lucky to have your accompany and after saying congratulations to the recently “title upgraded” Dr. Melis K., Dr. Franziska B., and Dr. Erik E., here I wish Mr. Christoffer J. best of luck in your phd project progress. Ilka A., Kerstin R., Xu J.,
David S., Chanaka, Jihua and many other current or previous colleagues at the department and from the SSF networks, I thank you for the contribution to such a great working atmosphere!
I’d like to thank the financers of this project: the Swedish Foundation for Strategic Research (SSF), the Department of plant physiology (UmU), Gunnar and Ruth Björkman's fund for botanical research in Norrland, and Kempe travel grant to cover my expenses for conferences.
Finally, I close my acknowledgement list with thanking and family:
Mum and dad: I would have never made this far without you who unconditionally support me and love me. My husband Daniel, being with you is the best thing happened in my life and I hope we will be together until the end of time. Pia, mamma och pappas lilla sötis, thanks for you, mamma is full of energy to fight for a better life. The little one on the way, it is great that you spend the same amount of time here when mamma writing her thesis in these dark and quiet nights. Thank you Margareta, Thomas, Celia, Lars, Gunnar, Lillamor… for all your love and support, and interested in knowing what actually I am doing.
27
5. References
Abreu IND, Ahnlund M, Moritz T, Albrectsen BR. 2011. UHPLC-ESI/TOFMS Determination of Salicylate-like Phenolic Gycosides in Populus tremula Leaves. Journal of Chemical Ecology 37: 857-870.
Albrectsen BR, Bjorken L, Varad A, Hagner A, Wedin M, Karlsson J, Jansson S. 2010a.
Endophytic fungi in European aspen (Populus tremula) leaves-diversity, detection, and a suggested correlation with herbivory resistance. Fungal Diversity 41(1): 17-28.
Albrectsen BR, Witzell J, Robinson KM, Wulff S, Luquez VMC, Agren R, Jansson S. 2010b.
Large scale geographic clines of parasite damage to Populus tremula L. Ecography 33(3):
483-493.
Arnold AE. 2007. Understanding the diversity of foliar endophytic fungi: progress, challenges, and frontiers. Fungal Biology Reviews 21(2): 51-66.
Axelsson EP, Hjalten J, Whitham TG, Julkunen-Tiitto R, Pilate G, Wennstrom A. 2011. Leaf ontogeny interacts with Bt modification to affect innate resistance in GM aspens.
Chemoecology 21(3): 161-169.
Ayres M, Clausen T, MacLean SJ, Redman A, Reichardt P. 1997. Diversity of structure and antiherbivore activity in condensed tannins. Ecology 78: 1696–1712.
Babst BA, Harding SA, Tsai C-J. 2010. Biosynthesis of Phenolic Glycosides from Phenylpropanoid and Benzenoid Precursors in Populus. Journal of Chemical Ecology 36(3): 286-297.
Bailey JK, Deckert R, Schweitzer JA, Rehill BJ, Lindroth RL, Gehring C, Whitham TG. 2005.
Host plant genetics affect hidden ecological players: links among Populus, condensed tannins, and fungal endophyte infection. Canadian Journal of Botany-Revue Canadienne De Botanique 83(4): 356-361.
Barbehenn RV, Constabel CP. 2011. Tannins in plant–herbivore interactions. Phytochemistry 72(13): 1551-1565.
Barbehenn RV, Peter Constabel C. 2011. Tannins in plant–herbivore interactions. Phytochemistry 72(13): 1551-1565.
Bernhardsson C, Ingvarsson PK. 2012. Geographical structure and adaptive population differentiation in herbivore defence genes in European aspen (Populus tremula L., Salicaceae). Molecular Ecology 21(9): 2197-2207.
Bernhardsson C, Robinson KM, Abreu IN, Jansson S, Albrectsen BR, Ingvarsson PK. 2013.
Geographic structure in metabolome and herbivore community co‐occurs with genetic structure in plant defence genes. Ecology Letters 16(6): 791-798.
Boeckler GA, Gershenzon J, Unsicker SB. 2011. Phenolic glycosides of the Salicaceae and their role as anti-herbivore defenses. Phytochemistry 72(13): 1497-1509.
Brillouet J-M, Romieu C, Lartaud M, Jublanc E, Torregrosa L, Cazevieille C. 2014a. Formation of vacuolar tannin deposits in the chlorophyllous organs of Tracheophyta: from shuttles to accretions. Protoplasma 251(6): 1387-1393.
Brillouet J-M, Romieu C, Schoefs B, Solymosi K, Cheynier V, Fulcrand H, Verdeil J-L, Conéjéro G. 2013. The tannosome is an organelle forming condensed tannins in the chlorophyllous organs of Tracheophyta. Annals of Botany: mct168.
Brillouet J-M, Verdeil J-L, Odoux E, Lartaud M, Grisoni M, Conéjéro G. 2014b. Phenol homeostasis is ensured in vanilla fruit by storage under solid form in a new chloroplast- derived organelle, the phenyloplast. Journal of experimental botany 65(9): 2427-2435.
Caseys C, Glauser G, Stölting KN, Christe C, Albrectsen BR, Lexer C. 2012. Effects of interspecific recombination on functional traits in trees revealed by metabolomics and genotyping-by-resequencing. Plant Ecology & Diversity 5(4): 457-471.
Chagnon PL, U'Ren JM, Miadlikowska J, Lutzoni F, Arnold AE. 2016. Interaction type influences ecological network structure more than local abiotic conditions: evidence from endophytic and endolichenic fungi at a continental scale. Oecologia 180(1): 181-191.
Chen F, Liu C-J, Tschaplinski TJ, Zhao N. 2009. Genomics of Secondary Metabolism in Populus:
Interactions with Biotic and Abiotic Environments. Critical Reviews in Plant Sciences 28(5):
375-392.
Clausen TP, Reichardt PB, Bryant JP, Werner RA, Post K, Frisby K. 1989. Chemical model for short-term induction in quaking aspen (Populus tremuloides) foliage against herbivores.
Journal of Chemical Ecology 15(9): 2335-2346.
Cole CT, Stevens MT, Anderson JE, Lindroth RL. 2016. Heterozygosity, gender, and the growth- defense trade-off in quaking aspen. Oecologia 181(2): 381-390.
Constabel CP, Lindroth RL 2010. The impact of genomics on advances in herbivore defense and secondary metabolism in populus. In: al. SJe ed. Genetics and genomics of populus, 279- 305.
Dixon RA, Xie DY, Sharma SB. 2005. Proanthocyanidins - a final frontier in flavonoid research?
New Phytologist 165(1): 9-28.
Donaldson JR, Kruger EL, Lindroth RL. 2006. Competition- and resource-mediated tradeoffs between growth and defensive chemistry in trembling aspen (Populus tremuloides). New Phytologist 169(3): 561-570.
Elamo P, Helander ML, Saloniemi I, Neuvonen S. 1999. Birch family and environmental conditions affect endophytic fungi in leaves. Oecologia 118(2): 151-156.
Flores G, Hubbes M. 1979. Phytoalexin production by Aspen (Populus tremuloides Michx.) in response to infection by Hypoxylon mammatum (Wahl.) Mill and Alternaria spp. European Journal of Forest Pathology 9(5): 280-288.
Glynn C, Herms DA, Orians CM, Hansen RC, Larsson S. 2007. Testing the growth-differentiation balance hypothesis: dynamic responses of willows to nutrient availability. New Phytologist 176(3): 623-634.
Gundale MJ, Deluca TH, Nordin A. 2011. Bryophytes attenuate anthropogenic nitrogen inputs in boreal forests. Global Change Biology 17(8): 2743-2753.
Gustafsson L, Fedrowitz K, Hazell P. 2013. Survival and vitality of a macrolichen 14 years after transplantation on aspen trees retained at clearcutting. Forest Ecology and Management 291:
436-441.
Hamberger B, Ellis M, Friedmann M, de Azevedo Souza C, Barbazuk B, Douglas CJ. 2007.
Genome-wide analyses of phenylpropanoid-related genes in Populus trichocarpa, Arabidopsis thaliana, and Oryza sativa: the Populus lignin toolbox and conservation and
