uef.fi
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND Dissertations in Forestry and Natural Sciences
ISBN 978-952-61-2655-5 ISSN 1798-5668
Dissertations in Forestry and Natural Sciences
DISSERTATIONS | KATRI NISSINEN | DOES SEX OR GROWTH FORM MATTER? | No 291
KATRI NISSINEN
DOES SEX OR GROWTH FORM MATTER?
A study of Populus tremula L., Salix myrsinifolia Salisb. and Salix repens L. facing ontogenetic and climatic changes
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND
The growth pattern of a plant species may affect its response to climate change. Also, female and male plants of dioecious species may differ in their growth and defence. This thesis provides new information of age- and sex-related changes in a tree, Populus tremula
L., in a small-sized shrub, Salix repens L., and in a tall shrub, Salix myrsinifolia Salisb., under changing climate conditions and widens
our understanding of the impacts of climate change on boreal ecosystems.
KATRI NISSINEN
DOES SEX OR GROWTH FORM MATTER?
A STUDY OF POPULUS TREMULA L., SALIX MYRSINIFOLIA SALISB. AND SALIX REPENS L. FACING ONTOGENETIC AND
CLIMATIC CHANGES
Katri Nissinen
DOES SEX OR GROWTH FORM MATTER?
A STUDY OF POPULUS TREMULA L., SALIX MYRSINIFOLIA SALISB. AND SALIX REPENS L. FACING ONTOGENETIC AND
CLIMATIC CHANGES
Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences
No 291
University of Eastern Finland Joensuu
2017
Academic dissertation
To be presented by permission of the Faculty of Science and Forestry for public examination in the Auditorium N100 in the Natura Building
at the University of Eastern Finland, Joensuu, on November, 24, 2017, at 12 o’clock noon
Grano Oy Jyväskylä, 2017 Editor: Matti Vornanen
Distribution: University of Eastern Finland / Sales of publications www.uef.fi/kirjasto
ISBN: 978-952-61-2655-5 (nid.) ISBN: 978-952-61-2656-2 (PDF)
ISSNL: 1798-5668 ISSN: 1798-5668 ISSN: 1798-5676 (PDF)
Author’s address: Katri Nissinen
University of Eastern Finland
Depart. of Environmental and Biological Sciences P.O. Box 111
80101 JOENSUU, FINLAND email: katri.nissinen@uef.fi
Supervisors: Professor Riitta Julkunen-Tiitto, Ph.D.
University of Eastern Finland
Depart. of Environmental and Biological Sciences P.O. Box 111
80101 JOENSUU, FINLAND email: riitta.julkunen-tiitto@uef.fi
Professor Line Nybakken, Ph.D.
Norwegian University of Life Sciences
Depart. Faculty of Environmental Sciences and Natural Resource Management
P.O. Box 5003 1432 ÅS, NORWAY
email: line.nybakken@nmbu.no
Senior Researcher Egbert Beuker, Ph.D.
Natural Resources Institute Finland
Depart. of Environmental and Biological Sciences Finlandiantie 18
58450 PUNKAHARJU, FINLAND email: egbert.beuker@luke.fi
Senior Researcher Virpi Virjamo, Ph.D.
University of Eastern Finland
Depart. of Environmental and Biological Sciences P.O. Box 111
80101 JOENSUU, FINLAND email: virpi.virjamo@uef.fi
Associate Professor Lauri Mehtätalo, Ph.D.
University of Eastern Finland School of Computing
P.O. Box 111
80101 JOENSUU, FINLAND email: lauri.mehtatalo@uef.fi
Reviewers: Professor Teemu Teeri, Ph.D University of Helsinki
Depart. of Agricultural Sciences P.O. Box 27
00014 HELSINKI, FINLAND email: teemu.teeri@helsinki.fi
Professor Hely Häggman, Ph.D University of Oulu
Depart. of Biology P.O. Box 8000
90014 OULU, FINLAND email: hely.haggman@oulu.fi
Opponent: Associate Professor Johanna Witzell, Ph.D Swedish University of Agricultural Sciences Southern Swedish Forest Research Centre P.O. Box 52
23053 ALNARP, SWEDEN email: johanna.witzell@slu.se
7 Nissinen, Katri
Does sex or growth form matter? A study of Populus tremula L., Salix myrsinifolia Salisb. and Salix repens L. facing ontogenetic and climatic changes
Joensuu: University of Eastern Finland, 2017 Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences 2017; 291 ISBN: 978-952-61-2655-5 (print)
ISSNL: 1798-5668 ISSN: 1798-5668
ISBN: 978-952-61-2656-2 (PDF) ISSN: 1798-5676 (PDF)
ABSTRACT
In this thesis, I studied the phenolic chemistry and growth of three different Salicaceae species in relation to three climate change factors, warming, CO2 and UVB, in addition to sex-related differences and ontogeny. Salix repens is a slow- growing dwarf shrub, Salix myrsinifolia is a fast-growing shrub and Populus tremula is a fast-growing tree. All the studies included several genotypes of each sex. The study of S. repens was conducted in a greenhouse, where the temperature was elevated by 3,5 °C above the ambient, and the CO2-level was raised to 720 ppm. The study of two-year old plants of P. tremula going through ontogenetic shifts was conducted in an open, computer-controlled experimental field, where the temperature was elevated by 1,8 °C and the UVB-radiation level by 30% above the ambient. The study of S. myrsinifolia was followed through ontogenetic shifts outdoors, during a 7-year period.
Of the three climate change factors studied, elevated temperature had the strongest effect. Warming increased the growth of both the slow-growing S. repens and the fast-growing P. tremula, as well as leaf salicylate concentrations in P.
tremula. Concentrations of flavonoids and phenolic acids decreased under elevated temperature in the leaves of both species, as did condensed tannins in P. tremula.
Warming suppressed the accumulative effects of enhanced CO2-levels on leaf flavonoids in S. repens and of UVB in P. tremula. In addition, the expression activity of a stress reactive, miR168, in young P. tremula seedlings responded to changes in UV and temperature.
S. myrsinifolia maintained its phenolic levels at a constant level over ontogeny from the sapling phase and after reaching the reproductive age, despite a slight variation possibly caused by yearly changes in climatic conditions in the field. This was in contrast to P. tremula, plants in the juvenile phase containing higher concentrations of simple-structured salicylates and phenolic acids in their leaves,
8
while plants in the adult vegetative phase had higher concentrations of more complex flavonoids and condensed tannins. This phase change started already during the second growing season. Interestingly, elevated temperature suppressed the ontogenetic changes in both flavonoid and condensed tannins levels in the leaves, and thereby prolonged the chemotypic juvenile phase. Furthermore, warming decreased rust infections, but increased herbivory damage in juvenile P.
tremula plants.
The slow-growing shrub, S. repens, grown in a greenhouse, showed sex-specific differences in both leaf defence levels and in growth. The fast-growing tree, P.
tremula, also showed strong sex-specific differences in its condensed tannins level in field conditions, whereas in the field-grown, intensively growing S. myrsinifolia, differences between sexes were rare and small. In all the studied species, the genotypic differences were pronounced, and in many cases they may have reduced possible sex-related differences. This high genotypic variation gives these different Salicaceae species, equally in each sex, a good ability to survive and adapt to a changing climate in the future.
In conclusion, both the fast-growing P. tremula and the slow-growing S. repens showed adaptation to climate change. However, S. myrsinifolia has the highest variability in growth as well as a constant level of phenolics. Therefore, of these studied species, S. myrsinifolia may have the best ability to cope with the expected sudden and unpredictable changes in future climate conditions and the concomitant increases in infestation.
Universal Decimal Classification: 581.19, 547.9, 582.681.1, 504.7
CAB Thesaurus: carbon dioxide; climate change; dioecy; flavonoids; genotypic variation;
growth; phenolic compounds; Populus tremula; Salicaceae; salicylates; Salix myrsinifolia;
Salix repens; sex differences; temperature; ultraviolet radiation
Muut asasanat: aspen; kaksikotisuus; UV-B radiation; willow
Yleinen suomalainen asiasanasto: fenoliset yhdisteet; flavonoidit; geneettinen muuntelu;
haapa; hanhenpaju; hiilidioksidi; ilmastonmuutokset; lämpeneminen; mustuvapaju; pajut;
salisylaatit; ultraviolettisäteily
9
ACKNOWLEDGEMENTS
My sincere gratitute is owed to my main supervisor, Professor Riitta Julkunen- Tiitto for her inspiring enthusiasm for research, her tireless guidance and her trust.
I am much obliged to Line Nybakken, for her valuable encouragement throughout this project and for her thorough guidance with the manuscripts, and to Virpi Virjamo, for her uplifting support and knowledgeable comments. I also want to express gratitude to Lauri Mehtätalo and Egbert Beuker, for their valuable guidance and for sharing their expertise.
I warmly thank all the other members of our research team, Tendry, Anu, Anneli, Paula, Unni and Norul, it has been a pleasure to work with you. I am grateful to Sinikka Sorsa, Hannele Hakulinen and Mervi Kupari for their valuable assistance in the laboratory and Matti Savinainen for his help in the experimental field. I wish to thank all other colleagues, co-authors and staff members, who have taken part in my research work. I also thank Rosemary Mackenzie for checking the English language of this thesis.
I wish to thank Professor Teemu Teeri and Professor Hely Häggman for their reviews and Associate Professor Johanna Witzell for accepting the task to act as my opponent in the public examination of this thesis.
The Academy of Finland is acknowledged for providing the funding (no.
267360) for this thesis and the Department of Environmental and Biological Sciences for providing the working facilities.
I am grateful to my family and friends for their support and care. My warmest thanks go to my husband, Reijo, for his endless help and encouragement, and Aleksi, Jesse, Eetu and Santeri; being your mother is the most important thing in my life.
Joensuu, 24th November 2017 Katri Nissinen
10
11
LIST OF ABBREVIATIONS
AGO1 ARGONAUTE 1 protein
C control, ambient conditions
CO2 carbon dioxide acid
Ct threshold value of qPCR
cDNA complementary DNA
GVA graphical vector analysis
HCH 6-hydroxy-2-cyclohexen-on-oyl
HPLC high pressure liquid chromatography
MeOH methanol
miR168 microRNA 168
miRNA microRNA
PCA principal component analysis
PCR polymerase chain reaction
ppm parts per million
R programming language
RH relative humidity
qRT-PCR quantitative reverse transcriptase PCR
U6 reference sequence in qPCR
UHPLC-DAD ultra-high pressure liquid chromatography with
photodiode array detection
UHPLC Q-TOF LC/MS UHPLC-quadrupole time-of flight liquid chromatography/mass spectrometer
UVA ultraviolet-A radiation (315–400 nm)
UVB ultraviolet-B radiation (280-315 nm)
12
13
LIST OF ORIGINAL PUBLICATIONS
This thesis is based on data presented in the following articles, referred to by the Roman Numerals I-III.
I Nissinen K, Nybakken L, Virjamo V, Julkunen-Tiitto R (2016). Slow-growing Salix repens (Salicaceae) benefits from changing climate. Environmental and Experimental Botany 128:59–68.
II Nissinen K, VirjamoV, RandriamananaT, Sobuj N, Sivadasan U, Mehtätalo L, Beuker E, Julkunen-Tiitto R, Nybakken L (2017). Responses of growth and leaf phenolics in European aspen (Populus tremula) to climate change during juvenile phase change. Canadian Journal of Forest Research, doi: 10.1139/cjfr- 2017-0188, in press.
III Nissinen K, Virjamo V, Mehtätalo L, Lavola A, Nybakken L, Julkunen-Tiitto R. A seven-year study of phenolic concentrations of the dioecious Salix myrsinifolia Salisb. Resubmitted.
The above publications have been included at the end of this thesis with their copyright holders’ permission.
In addition, the effects of enhanced temperature and UVB-level on the activity of miR168 were studied from Populus tremula L. and herbivory infestation of Salix myrsinifolia Salisb.
14
AUTHOR’S CONTRIBUTION
1) The author was responsible for the growth, photosynthesis and biomass measurements in addition to the chemical analyses in paper II.
2) The author took part in sampling and growth measurements in paper III and processed the chemistry data in papers I and III.
3) The author planned the miRNA experiment together with her main
supervisor and was responsible for the maintenance and sampling of miRNA plants and for implementing the miRNA analyses.
4) The author conducted statistical analyses and is the main author in all of the three papers (I, II, III).
15
CONTENTS
ABSTRACT ... 7
ACKNOWLEDGEMENTS... 9
1 INTRODUCTION ... 17
1.1 Salicaceae species ... 17
1.2 Sex-related differences ... 17
1.3 Phenolic compounds in Salicaceae species ... 18
1.4 Age-related changes in phenolics in Salicaceae species ... 21
1.5 Effect of climate change on plants ... 21
1.6 Infestation in Salicaceae species ... 22
1.7 Aims of the thesis ... 23
2 MATERIAL AND METHODS ... 25
2.1 Plant material ... 25
2.2 Experiments ... 25
2.3 Sampling and measurements of growth, leaf traits and infestation ... 30
2.4 Laboratory analyses ... 30
2.4.1 Phenolic analyses ... 30
2.4.2 miRNA analyses ... 31
2.5 Statistics ... 31
3 RESULTS AND DISCUSSION ... 33
3.1 Growth form differences in chemistry ... 33
3.2 The interactive effects of age and climate change factors on growth and leaf phenolics ... 36
3.3 Infestation... 39
3.4 Sex-related and genotypic differences ... 42
4 CONCLUSIONS ... 45
5 BIBLIOGRAPHY ... 47
6 APPENDICES ... 55
16
17
1 INTRODUCTION
1.1 SALICACEAE SPECIES
The Salicaceae family includes nearly 60 genera with over 1000 species (The Plant List 2013). Species belonging to Salix and Populus genera of the Salicaceae family grow wild in Finland. Salix species are distributed across most of the world’s continents, and the majority, over 400 species, are shrubs (Skvortsov 1999). Salix repens is widespread in northern Europe, and in Finland it is found mostly in the western parts. It is a slow-growing, low (height 0.1–0.5 m) shrub succeeding in many different habitats, from infertile heaths and sand dunes to peatland meadows and birch and pine forests (Hämet-Ahti et al. 1998, Skvortsov 1999). The rapid- growing Salix myrsinifolia is one of the most variable in its growth forms among the Salix species, ranging from shrubs to small trees (height 2–10 m). Both S. repens and S. myrsinifolia are adapted to a great variety of secondary habitats, resulting from human activities (Skvortsov 1999). Populus tremula belongs to the genus Populus (von Wühlisch 2009). It is a rapid-growing, early successional tree species, and one of the most widespread tree species in the world. In Finland, it succeeds in all kinds of different forest types (Hämet-Ahti et al. 1998). Neither the Populus nor the Salix species are of any great economical importance, although Salix species are used for biomass crops and environmental restoration (Aylott et al. 2008, Kuzovkina and Quigley 2005). Generally, the Populus and Salix species are an ecologically valuable tree and shrub species, important for the lives of hundreds of different mammal, bird and invertebrate species (von Wühlisch 2009).
1.2 SEX-RELATED DIFFERENCES
The Salicaceae species are dioecious. The sex ratio of S. repens and S. myrsinifolia, as in most of the Salix species, is sexually female-biased 2:1 (de Jong and van der Mejden 2004), whereas P. tremula is male-biased 2:1 (Worrel et al. 1999). The reasons for these differences in the sex ratios are still unsolved. They may result from the differences between the sexes in their adaptation to different natural habitats (Munné-Bosch 2015). According to Obeso (2002), females of dioecious species are expected to invest more of their resources in reproduction and synthesis of defence compounds, whereas males are expected to be more growth-oriented. Although Cornelissen and Stiling (2006) have found males of various plant species in general to be more susceptible to herbivore damage, the results from Populus and Salix
18
species are variable. In addition, we know very little about sex-related differences in plant vulnerability to different diseases or herbivory.
1.3 PHENOLIC COMPOUNDS IN SALICACEAE SPECIES
In general, Salicaceae species, such as Salix and Populus, contain large amounts of different phenolic compounds (e.g. Julkunen-Tiitto 1986) (Fig. 1). Many of these secondary metabolites function as protective compounds against herbivory, diseases or other stress factors, such as UVB-radiation (Barbehenn and Constabel 2011, Nybakken et al. 2012, Ruuhola et al. 2001, Seigler 1998).
Phenolic compounds are synthesized via the phenylpropanoid pathway (Seigler 1998) (Fig. 2). Regulation of the metabolism of different phenolic compounds is a complex network still under study, including synthesis and possible storage and degradation steps at different developmental stages of the plant, as well as several external environmental factors.
The group of phenolic acids is the simplest in structure. They belong to the so- called low molecular weight phenolics together with several other subgroups, e.g.
salicylates, phenolic glycosides and flavonoids. The most common phenolic acids in Salicaceae species are cinnamic and chlorogenic acids (Fig. 1) (e.g. Nybakken et al.
2012, Nybakken and Julkunen-Tiitto 2013, Randriamanana et al. 2015a). In addition to their role in defence against herbivory and UVB-radiation (Izaguirre et al. 2007, Randriamanana et al. 2015a, b), phenolic acids are used as precursors of other more complex-structured phenolic compounds, including salicylates (e.g. Maeda and Dudareva 2012).
Most of the species belonging to the Salicaceae family contain considerable amounts of salicylates, conjugated -D-glucosides of salicyl alcohol, though differing both qualitatively and quantitatively (Boeckler et al. 2011, Julkunen-Tiitto 1989b, Julkunen-Tiitto and Virjamo 2017). The simplest salicylate compound is salicin, whereas salicortin and tremulacin represent more complex-structured compounds (Fig. 1) (Ruuhola and Julkunen-Tiitto 2000). Salicylates have been proved to play an important role in plant defence against generalist herbivores (e.g.
Boeckler et al. 2011). Salicylates have anti-inflammatory effects in humans and they have been used as painkillers (Förster 2008).
19 HO
OH
OH
O O
HO OH
O O
OH OH
OH HO
HO O OH
OH OH OH HO O
OH
OH OHOH HO O
OH OH HO O
OH
OH OH
OH
OH OH OH
O
OH HO
OH O O O
OH
HO OH
OH OH OH
HO O
O
OH O O
HOHO
OH OH OH
O O OH
HO O HO
O
O OOH
O O O
OH
HOHO
O OOH OH O
condensed tannins quercetin 3-galactoside
(+)-catechin salicin
tremulacin salicortin
chlorogenic acid cinnamic acid
Figure 1. Chemical structures of some phenolic compounds and condensed tannins found in Salicaceae species
20
phenylalanine
cinnamic acids salicylates p-coumaroyl coA
flavonoids
condensed tannins
chlorogenic acids
Figure 2. A simplified chart of the leaf phenolics synthesis pathway in Salicaceae species
Flavonoids, sharing the precursors in the phenylpropanoid pathway with the chlorogenic acids, are generally more complex in structure than phenolic acids or salicylates (Fig. 1) (Seigler 1998, Brunetti et al. 2013). They consist of two phenyl rings and one heterocyclic ring. Chlorogenic acids can serve as a temporary storage for flavonoid synthesis (Seigler 1998). Many flavonoids, such as quercetin 3- galactoside, are acknowledged to have antioxidant properties, e.g. against reactive oxygen species and in photoprotection, and also to provide defence against herbivory and pathogens (Fig. 1) (e.g. Agati et al. 2013). Flavonoids also play an important role as regulators in cell growth and differentiation processes in plants (Agati et al. 2013, Ferreyra et al. 2010).
High molecular weight phenolic compounds are polymerized phenolics, such as condensed tannins (proanthocyanidins), which are polymers of flavonoid catechins (Fig. 1). Condensed tannins are assumed to be of structural importance, but they also play an important role in plants’ defence against herbivores (e.g. Barbehenn and Constabel 2011). Altogether, phenolic compounds can account for more than 20% of plant dry weight in Salicaceae species.
21
1.4 AGE-RELATED CHANGES IN PHENOLICS IN SALICACEAE SPECIES
The ontogenetic development of higher plants often includes morphological and chemical changes (Brunner & Nilsson 2004, Poethig 2003). The concentrations of phenolic compounds in boreal species are expected to be at a high level during the juvenile stages, and to decline after reaching the reproductive age. There is, however, variation between species, depending on differences, e.g. in morphology, herbivore pressure and defence traits (Barton and Koricheva 2010, Donaldson et al.
2006.
In trees from the genus Populus, the level of salicylates is high only during the juvenile stage (Randriamanana et al. 2015a). With maturation, they are replaced by other phenolic compounds, such as flavonoids and condensed tannins, as seen in P.
tremula and P. tremuloides (Donaldson et al. 2006, Julkunen-Tiitto 1986, Smith et al.
2011). In P. tremula, the developmental phase change is seen morphologically in the shape of the leaves. The leaf blades are heart-shaped with a sharp tip in the rapid- growing juvenile plants and round-tipped in vegetative adult-phase plants.
In Salix species, at the seedling phase, the concentrations of 2’cinnamoylsalicortin and salicortin in S. sericea (Andersson), as also the concentrations of condensed tannins in S. eriocephala (Michx.) increase with age (Fritz et al. 2001) and with above-ground growth in older seedlings (Orians et al.
2010). In S. pentandra L., the high level of salicylates decreases when seedling plants reach the sapling phase, with a concomitant increase in flavonoids and condensed tannins (Julkunen-Tiitto and Virjamo 2017). In S. myrsinifolia, the level of salicylates is also at a high level in mature plants (Julkunen-Tiitto 1986, Randriamanana et al.
2015b).
1.5 EFFECT OF CLIMATE CHANGE ON PLANTS
Sufficient levels of light, temperature, moisture and nutrients are prerequisites for plants to thrive. At the moment, the global temperatures are increasing, because of the rising atmospheric CO2 level (Intergovernmental Panel on Climate Change (IPCC) 2014). The effects of the temperature increase are regional, and in northern and Arctic regions, the rise in temperature is intensified by diminishing snow cover, which reflects a smaller portion of the incoming radiation (Bais et al. 2015).
At high latitudes, slightly elevated temperatures generally have a positive effect on all the growth parameters of plants through accelerated photosynthesis. In addition, CO2 enrichment has increased the accumulation of carbohydrates in plants and thereby increased their growth (e.g. Cole et al. 2010, Lavola et al. 2013).
Concentrations of phenolic compounds have generally been found to increase under enhanced CO2-levels, but to decrease under elevated temperatures
22
(Nybakken and Julkunen- Tiitto 2013, Paajanen et al. 2011, Zvereva and Kozlov 2006). This decrease may often be a result of dilution caused by increased growth (Nybakken et al. 2012, Randriamanana et al. 2015a, Zvereva and Kozlov 2006).
The already increased levels of temperature and CO2 are expected to rise even further. At the same time, the depleted ozone layer is recovering, as a result of the Montreal Protocol (IPCC 2014). However, the ensuing decrease in the amount of UVB fluctuates regionally, depending on changes in e.g. clouding and air pollutants (Bais et al. 2011, McKenzie et al. 2011). At northern latitudes, it is predicted that the UV-radiation level will be restored within the next 10 years (Bais et al. 2011).
Through recent research, it has been found, that UVB-radiation is generally not a stress factor under natural conditions, but functions as a regulator in plant growth and development (Ballaré et al. 2011, Bornman 2015).
Plant responses to different factors depend on their ontogenetic phase and genotypic differences. At the genetic level, most UVB-responsive genes are specific to a certain developmental phase and to a certain organ, and they are regulated when a certain threshold is met, rather than expressed proportional to the dose (Casati & Walbot 2004). MicroRNAs (miRNAs) are small (app. 20 nucleotides long), conserved, non-coding RNAs, that take part in specific genetic post-transcriptional down-regulation by degrading target mRNAs involved in the growth and development of plants (Bartel 2004). Even though miRNAs are conserved, they can have species-specific functions (Lu et al. 2008). In P. tremula, miR168 is a UVB responsive miRNA that is involved in plant developmental regulation through feedback regulation of its target ARGONAUTE1 (AGO1) gene (Jia et al. 2009, Vaucheret et al. 2006, Vaucheret et al. 2009). AGO1 is involved in regulation of different stress responses, and it is associated with the regulation of juvenile and adult phases in plant development (Wang et al. 2011).
1.6 INFESTATION IN SALICACEAE SPECIES
The Salicaceae species have gone through co-evolution with various insect herbivore and mammal species and also with pathogens, resulting in a combination of both physical (e.g. trichomes) and chemical protection mechanisms (Volf et al.
2015). For example, herbivory has induced the synthesis of 2’-cinnamoylsalicortin in young, undamaged leaves of S. sericea (Fields and Orians 2006). On the other hand, salicylate tremulacin increased the oviposition activity of a specialist insect herbivore, Euura lsiolepis (Roininen et al. 1999). Results have shown that generalist herbivores are more seriously affected by the total secondary chemistry of willows, whereas leaf nitrogen content and plant physical defence mechanisms have a stronger influence on specialists (Rubert-Nason 2015, Volf et al. 2015). In relation to climate, insect herbivore species have often benefitted from warmer conditions, e.g.
through acceleration of their developmental and growth rates (Bale et al. 2002,
23 Uelmen et al. 2016). More severe insect herbivore damage may appear in temperate and boreal regions, where the proportional rise of the temperature is higher than in the tropics, thus enabling colonization of these northern areas by southern insect species adapted to warmer climate conditions (Bale et al. 2002). In addition, the phenological synchrony between herbivore insect egg hatch and bud break may be affected by warming, which favours insect species that are capable of variability and mobility (e.g. Uelmen et al. 2016). Furthermore, though not having a direct impact on herbivore succession, enhanced UVB may indirectly improve plants’
defence against herbivores through chemicals produced via the same chemical synthesis routes (Caldwell et al. 2013).
In addition to herbivory, different fungal pathogens can severely harm the growth of Salicaceae species and their defence against other diseases (Heiska et al.
2007, Ramstedt 1999). For these fungi, e.g. the availability of a sufficient amount of moisture is essential (Helfer 2014). For example, infections of Melampsora-rust decreased under elevated temperature in both P. tremula and S. myrsinifolia (Nybakken et al. 2012, Randriamanana et al. 2015a). Rust infection increased the concentrations of salicortin, chlorogenic acid and (+)-catechin in leaves of S.
myrsinifolia (Hakulinen 1998, Hakulinen et al. 1999). In Populus, rust infections increased the synthesis of condensed tannins (Miranda et al. 2007).
1.7 AIMS OF THE THESIS
In nature, plant species face several simultaneous, fluctuating changes in their growth environment. In addition, the morphological and biochemical characteristics of the species, as well as the ongoing developmental phase changes, all have an effect on the responses to climate change. The overarching aim of this thesis was to study the effects of the ontogenetic phase and different climate change factors on three Salicaceae species of different phenotypes: a dwarf shrub S. repens, a tall shrub S. myrsinifolia, and a tree P. tremula.
In addition, we were interested in possible sex-related responses to these climate change effects and whether females and males of dioecious species respond differently to the environmental changes so that the fitness of one sex is improved compared to that of the other. If so, this could have an effect on the sex-bias and thereby on the overall survival of the species.
The following questions were addressed;
1. Do males and females of S. repens (I), S. myrsinifolia (III) and P. tremula (II), differ in their growth and defence parameters?
2. What are the effects of elevated temperature (I, II), CO2-level (I) and UVB- radiation (II) on S. repens (I) and P. tremula (II)?
3. Does elevated temperature or UVB-radiation affect the activity of miR168 in Populus tremula?
24
4. Do males and females of slow-growing S. repens (I) and fast-growing P.
tremula (II) respond differently to elevated temperature, CO2-level and UVB-radiation treatments?
5. How does ontogenetic development affect the leaf phenolics of S.
myrsinifolia (III) and P. tremula (II) and the phenolics response to elevated temperature and UVB-radiation in P. tremula (II)?
6. Are there any changes in the herbivory and Melampsora-rust severity of P.
tremula (II) and S. myrsinifolia (III) during ageing, and of P. tremula (II) under elevated temperature and UVB-radiation?
25
2 MATERIAL AND METHODS
2.1 PLANT MATERIAL
The S. repens and S. myrsinifolia plants used in experiments I and III were planted as cuttings, which were collected from plants growing naturally at different sites in western (S. repens) and eastern (S. myrsinifolia) Finland. The P. tremula plants used in experiment II were originally micro-propagated from the dormant axillary buds of adult trees collected from different sites in eastern and southern Finland. The collection sites for all three plant species used were located several kilometers apart to ensure that they were different genotypes. The plantlets used in the miR168 experiment originated from three P. tremula genotypes, two female genotypes and two male genotypes. They were micro-propagated as above, replanted in 5.0 l plastic pots (Teku EC24, Pöppelman, Germany) with two plantlets of the same genotype per pot, and then moved to the experimental field on June 15, 2013.
2.2 EXPERIMENTS
An overview of different experiments included in this thesis is presented in Table 1.
The S. repens greenhouse experiment (I), was conducted at the Mekrijärvi research station located in eastern Finland (Zhou et al. 2012). Cuttings of 12 different genotypes were placed in each of 16 greenhouse chambers (Fig. 3). Three different treatments were applied in these chambers: elevated temperature, elevated CO2- level, a combination of elevated temperature and CO2-level, in addition to control conditions. There were four replicates of each chamber type. In the temperature treatment, the chamber temperature was elevated by 3.5 °C compared to the ambient temperature measured outside. The control CO2-level was around 360 ppm and the enhanced level of it was around 720 ppm.
26
Table 1. Plant species, number of replicates, genotypes and individuals used and the different plant traits studied in addition to the types and duration of the experiments in articles I, II and III. T refers to elevated temperature, CO2 refers to enhanced CO2-level and UV refers to enhanced UVB- and UVA-radiation levels
Article I II III
Species Salix repens Populus tremula Salix myrsinifolia
Age of plants initially Two-year old
cuttings One-year old plants Two-year old cuttings
Type of experiment Greenhouse Field Field
Duration Two months
One growing season/five months
(the second year of the experiment)
Seven years
Treatments T, CO2 T, UV
Studied effects Sex, T, CO2 Sex, T, UV
Phase change Sex, Age
Number of replicates Four Six Seven
Number of genotypes
Five females Seven males
Six females Six males
Eight females Nine males Total number of
individuals
120/leaf phenolics 189/growth
432/leaf phenolics 1728/growth
153/leaf phenolics 401/growth
Plant traits investigated
Leaf phenolics Leaf dry weight
Leaf area Spesific leaf area
Height growth Diameter growth
Shoot biomass
Leaf phenolics Leaf dry weight
Leaf area Spesific leaf area
Height growth Diameter growth
Shoot biomass miR168 expression
Rust infection Herbivory damage
Chlorophyll content index Leaf gas exchange
Leaf phenolics Stem phenolics Leaf dry weight
Leaf area Spesific leaf area
Height growth Diameter growth
Flowering Rust infection Herbivory damage
27 Figure 3. Salix repens in one of the experimental chambers at the Mekrijärvi research station in eastern Finland (photograph Riitta Julkunen-Tiitto)
The P. tremula field experiment (II) was conducted in the botanical garden of Joensuu, Eastern Finland, Botania (Fig. 4) (Nybakken et al. 2012). In the experimental field, there were 36 different treatment plots with six different treatment and six replicates of each treatment: elevated temperature, elevated UVB- radiation, elevated UVA-radiation, combined treatments of elevated temperature with UVB-radiation, elevated temperature with UVA-radiation, and also control plots. Elevated temperature and UVB-levels were set and 2.0 °C and 30% above the ambient, respectively. Both temperature and UVB ambient levels were measured in the field with special sensors. The P. tremula plants used in this experiment had already been growing in the field for one growing season (Randriamanana et al.
2015a). There were in total 1728 plants, 4 replicate individuals of the 12 different genotypes in each of the 36 plots. The P. tremula plants for the miR168 experiment were grown in the same experimental field in pots on the southern sides of the elevated temperature and UVB treatment plots, their combination and control plots,
28
and in four replicates of each, for six weeks, and they were watered daily if necessary (Fig. 5).
(a)
(b)
Figure 4. a) Aerial view (fonecta.fi) of the Populus tremula plants in the elevated temperature and UV experimental field in the botanical garden in Joensuu, Eastern Finland. b) Populus tremula plants in the field in the second growing season
29 C
T
UVB
UVB T
C
T+UVA
T+UVB
C
T UVB
T+UVB UVA
T+UVB
T+UVB
T+UVB
T+UVB
C
C
C T
T
T UVB
UVB UVB
UVA
UVA
UVA
UVA
UVA
T+UVA
T+UVA
T+UVA
T+UVA T+UVA
Figure 5. Map of the Populus tremula experimental field in Botania, Joensuu, Eastern Finland. Squares represent different plots of ambient conditions (C), elevated temperature (T), enhanced UVB-radiation (UVB) and enhanced UVA-radiation (UVA) treatments and their combinations. Circles represent the places of the separate pot-planted Populus tremula plantlets used in miR168 experiment
In the seven-year field study (III), S. myrsinifolia 7–10 cuttings from 8 female genotypes and 9 male genotypes were planted randomly in rows in an abandoned hay field located in Luikonlahti, Kaavi, Eastern Finland in 2007 (Fig. 6).
Figure 6. The Salix myrsinifolia experimental field in Kaavi, Eastern Finland in 2010
30
2.3 SAMPLING AND MEASUREMENTS OF GROWTH, LEAF TRAITS AND INFESTATION
The height and diameter of the longest shoot were measured every week during experiment I, every third week during experiment II and once a year in the seven- year experiment III. At the end of the experiment (I, II) and at the end of the growing season of each study year (III), the youngest fully expanded leaf or leaves were sampled, dry-air dried (RH 10%) at room temperature and kept frozen (-20
°C) before chemical analyses. Leaves were collected from a total of 160 plant individuals in experiment I (from five female and five male genotypes from each of the 16 chambers), 432 plant individuals in experiment II (from six female and male genotypes from each of the 36 plots), and from 150–153 plant individuals each year (from eight female and nine male genotypes), in experiment III (Table 1).
The naturally occurring herbivory and rust damage in the leaves of S.
myrsinifolia and P. tremula plants were measured in both of the field experiments, I and III. The severity of Melampsora-rust infections was calculated by microscopic examination of the leaf blades (II, III), and herbivory damage was calculated as the ratio of eaten leaf area per total leaf area (cm2) measured from the sampled leaves (II, additional material from S. myrsinifolia experiment III) with a portable leaf area meter before chemical analyses. Herbivory damage to the leaves of S. myrsinifolia was measured in the years 2008, 2012 and 2013. In 2011, herbivory damage was measured only as the number of affected leaves.
In the miRNA experiment, the youngest fully expanded leaf from every plant was collected on July 26 between 11.30 am and 1 pm local time. The leaves were frozen in liquid nitrogen, placed in small plastic bags and stored at -80 °C until RNA was isolated.
2.4 LABORATORY ANALYSES
2.4.1 Phenolic analyses
Low molecular weight phenolic compounds from the dried leaves (I, II, III) were extracted with ice-cold methanol according to Nybakken et al. (2012), before analysis by ultra-high performance liquid chromatography (UHPLC-DAD) (I) and high performance liquid chromatography (HPLC) (II, III). Concentrations of different low molecular weight phenolic compounds were calculated according to commercial standards and identified using UHPLC Q-TOF LC/MS and comparison of retention times and ultraviolet spectra with the authentic compounds and existing literature. The concentrations of condensed tannins were quantified using the acid butanol assay method (Hagerman 2002) (II, III). Purified tannin extracts
31 from the leaves of P. tremula were used as standards for quantification (Randriamanana et al. 2014).
2.4.2 miRNA analyses
RNA for miRNA analyses was isolated using the mirVanaTM miRNA Isolation Kit, with phenol (Ambion, Inc., Austin, USA) and following the instructions provided with the kit. The amount and purity of the isolated RNA was checked using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Wilmington, USA) and separating it by 1% agarose gel electrophoresis.
A real-time quantitative reverse transcription-polymerase chain reaction (qRT- PCR) (LightCycler480, Roche, Switzerland) was performed using ExiLENT SYBR®
Green master mix, 2.5 ml, miRCURY LNATM Universal RT microRNA PCR, (Exiqon A/S, Vedbæk, Denmark) and the Universal cDNA synthesis kit II (Exiqon A/S, Vedbæk, Denmark). For each sample, we used a miR168-primer set (ath-miR168a LNATM, PCR primer set, UniRT, Exiqon A/S, Vedbæk, Denmark), with a target sequence (5'–3') UCGCUUGGUGCAGGUCGGGAA and U6 was used as an endogenous control (U6_Populus_t, Custom miRCURY LNATM, PCR primer set, UniRT, Exiqon A/S, Vedbæk, Denmark). Three technical replicates was performed in the RT-PCR for each sample. For calculating fold change values, the mean of three technical replicates was first calculated for each sample, for both 168miRNA and U6. Fold change values were calculated using the formula 2-Δ(ΔCt), where ΔCt was the difference between the Ct-values of U6 and 168miRNA, and Δ(ΔCt) the difference between each treated sample's ΔCt-value and the mean of the control samples’ (n=6) ΔCt-values.
2.5 STATISTICS
The analyses of treatment, sex and age effects on phenolics, leaf shape, growth, flowering, gas exchange parameters and infestation severity were conducted using the linear mixed effect model using SPSS (IBM® SPSS® Statistics 22 and 24 Armonk, NY, USA) (I, II) and R version 3.3.1 (R Core Team 2016) (II, III) with packages lme4 (Bates et al. 2015) and lmerTest (Kuznetsova et al. 2016) in RStudio 0.99.903 (© 2009–2016 RStudio, Inc., Boston, USA) (II, III). In addition, the effect of the treatments on the proportion of the sharp-tipped leaves was analysed by the Independent-Samples T Test using SPSS (II). In S. myrsinifolia, the number of flowering plants (III), plants with severe rust-infections (III), and plants attacked by herbivores, was analysed using a binary logistic mixed-effect model in R. The difference between the relative fold change values of miR168 of samples under UVB, T and UVB+T treatments and a control value 1 was analysed by the One
32
Sample T-test. The difference between treatments was analysed by One-Way Anova in SPSS (II). The fold change values were log10(x)-transformed before statistical analyses.
A graphic vector analysis (GVA) (Haase and Rose 1995) was carried out to study the effects of elevated temperature and CO2-level, and their interactions, on the phenolic group levels in the leaves of S. repens (I); elevated temperature, UV-levels and their interactions on the phenolic group levels in the leaves of P. tremula (II) and the effect of aging on the phenolic group levels in the leaves of S. myrsinifolia (III). A principal component analysis (PCA) was conducted using the SIMCA-P+- program, in order to obtain an overview of the relationships between the different genotypes in the total 7-year leaf phenolics data of S. myrsinifolia.
33
3 RESULTS AND DISCUSSION
3.1 GROWTH FORM DIFFERENCES IN CHEMISTRY
The fast-growing tree P. tremula contained 27, the slow-growing shrub S. repens 18, and the diverse S. myrsinifolia 31 different leaf low molecular weight phenolic compounds (I, II, III). Salicylates were the main group of low molecular weight phenolics in all of the three studied species (Fig. 7), while salicortin was the most abundant salicylate compound (I, II, III). One new salicylate compound was detected in P. tremula, HCH-salicin 2, where it is suggested that the HCH-group is attached to glucose (Fig. 8) (II). Two new flavonoid compounds were detected in S.
repens; a quercetin-malonyl-glucoside and a isorhamnetin-malonyl-glucoside derivative (Fig. 9) (I). Salix myrsinifolia had the highest concentration of phenolic acids (Fig. 7) (III), and chlorogenic acids were the main phenolic acid compounds in all the studied species (I, II, III).
Figure 7. Concentrations of different low molecular weight phenolic groups in leaves of Salix repens, Salix myrsinifolia and Populus tremula. Bars represent mean values +SE (n=4 for S.
repens, n=7 for S. myrsinifolia and n=6 for P. tremula)
34
O O HOHO
O OH OH
HOC O O
HCH-salicin 2
Figure 8. Potential chemical structure of HCH-salicin 2 in leaves of Populus tremula.
O O
OH O
OH CH3
O O
HO O
OH O O
HO OH
HO O
OH HO
OH O
OH
O O
HO O
O OH HO OH
(a)isorhamnetin-malonyl-glucoside (b)quercetin-malonyl-glucoside
Figure 9. Potential chemical structures of isorhamnetin-malonyl-glucoside (a) and quercetin- malonyl-glucoside (b) in leaves of Salix repens
Slow-growing species often constitutively keep concentrations of defence compounds at a high level (Coley et al. 1985, Tjoelker et al. 1998). In addition, they may not be able to compensate for the possible biomass loss through herbivory by growth, to the same extent as do fast-growing species such as S. myrsinifolia and P.
tremula (Barton and Koricheva 2010). Such high constitutive defence was seen in the slow-growing dwarf-shrub S. repens (I), which contained two times the amount of salicylates in the leaves compared to the fast-growing tree P. tremula (II) and to the tall shrub S. myrsinifolia (III) (Fig. 7). Salix repens (I) also had more of the low molecular weight salicin than P. tremula (II) and S. myrsinifolia (III). Small-sized salicylates are cheaper to biosynthesize and reuse through degradation and new synthesis than are large-sized molecules, such as, for example, condensed tannins
35 (e.g. Ruuhola et al. 2003, Babst et al. 2010). In S. repens, salicortin concentrations were high, together with tremulacin (I). High levels of these salicylates may indicate their dominant role in constitutive protection against generalist herbivores.
The effects of size and the growth pattern on phenolic levels were also seen in a comparison of Salix daphnoides Vill., Salix purpurea L. and S. pentandra. S. pentandra grew the highest of these three species, and it had the smallest amount of stem phenolics (Förster et al. 2008, 2010). In addition, the small-sized S. lapponum L., was eaten less by hares than was the taller Salix species (Tahvanainen et al. 1985).
By contrast, fast-growing species, with their morphological and biochemical plasticity, are able to compensate for the biomass loss by accelerated growth increment, even after severe herbivory damage, and they may rely more on induced defence than do slow-growing plants (Barton and Koricheva 2010). Both S.
myrsinifolia and P. tremula have a strong regrowth ability compared to the slow- growing S. repens. Furthermore, as a tree, P. tremula may be more oriented towards growth than the shrub S. myrsinifolia, which has a variable growth pattern. P.
tremula (II) had only a small amount of simply-structured salicin compared to S.
myrsinifolia (III). In addition, the overall proportion of simpler-structured phenolic acids was higher in S. myrsinifolia (19%) (III) than in P. tremula (10%) (II). Of the different low molecular weight phenolic groups in the three studied species (I, II, III), flavonoids showed the greatest qualitative diversity. Populus tremula had the highest rate of flavonoids and it also had more of the high molecular weight condensed tannins than did S. myrsinifolia (Fig. 10) (II, III).
Figure 10. Concentrations of condensed tannins in leaves of Salix myrsinifolia and Populus tremula. Bars represent mean values +SE (n=6 for P. tremula and n=7 for S. myrsinifolia)
36
The phylogeny of the Salicaceae family, and especially within the Salix species, is still largely unresolved. Salix, Chamaetia and Vetrix are the three subgenera of the genus Salix (Skvortsov 1999). The subgenus Salix is the most primitive of these, and the most evolved Salix species, such as S. myrsinifolia, belong to the subgenus Vetrix.
Based on morphological differences, Salix species are supposed to have developed from Populus-like species (Skvortsov 1999). Salix species, for example, have uniform bud scales, a smaller number of stamens and the ability for intense hybridization and forming hermaphrodite catkins.
Though no clear correlation between morphology and phenolics has been found in Salix species (Julkunen-Tiitto 1989a, Skvortsov 1999), the differences in leaf phenolics may point towards P. tremula being at an earlier evolutionary phase than S. myrsinifolia and S. repens. In P. tremula, the high level of salicylates during the sapling phase only, is later replaced by a high level of flavonoids and condensed tannins (II) (Randriamanana et al. 2015a). These compounds are synthesized further down the phenylpropanoid pathway than are salicylates, and they are more complex in structure (e.g. Seigler 1998). This pattern of P. tremula may be an ancestral characteristic, which is also seen in some Salix species, as in the tree-like S.
pentandra (Julkunen-Tiitto 1989b). In addition, both P. tremula (II) and S. repens (I) had higher amounts of the complex-structured salicylate compound, tremulacin, which was very rare in S. myrsinifolia (III). The appearance of the complex- structured tremulacin may be an ancestral characteristic in P. tremula, which has emerged again in S. repens (e.g. Julkunen-Tiitto 1989b).
3.2 THE INTERACTIVE EFFECTS OF AGE AND CLIMATE CHANGE FACTORS ON GROWTH AND LEAF PHENOLICS
Salix myrsinifolia maintained a high constant defence level in all of the different leaf phenolic compounds with increasing age and across different ontogenetic stages, from juvenile to mature, and after reaching the reproductive phase (Appendix 1, III). On the other hand, in P. tremula, the concentration of salicylates was higher in the leaves of plants at an ontogenetically earlier phase compared to the plants at later phase (II). In the group of phenolic acids, especially the cinnamic acids were at a high level in the leaves at the earlier phase (II), which is understandable, since cinnamic acids are the precursors of salicylates (e.g. Babst et al. 2010). Salicylates were at a high level only during the first growing season in P. tremula (Randriamanana et al. 2015a), and they began to be replaced by more complex condensed tannins already during the second growing season (II). In the concentrations of other flavonoids detected, with the exception of (+)-catechin and condensed tannins, significant changes were towards lower amounts in P. tremula plants at later ontogenetic phases (II).
Of the three studied climate change factors, elevated temperature had the strongest impact on both leaf phenolics and growth parameters (experiments I and
37 III). This is in accordance with studies of other northern tree species, e.g. Betula pendula Roth. (Lavola et al. 2013). All three studied Salicaceae species responded to warmer conditions with increased growth (I, II, III). Height, diameter and biomass growth as well as leaf area increased under elevated temperature in the fast- growing tree P. tremula (II), as earlier seen, e.g. for S. myrsinifolia (Nybakken et al.
2012) (II). The biomass of P. tremula almost doubled under elevated temperature compared to under ambient temperature conditions (II). Although slow-growing species are not expected to be able to respond fast to environmental changes (Coley et al. 1985), the slow-growing shrub S. repens increased in height, diameter and biomass growth in the greenhouse (I), as did also the fast-growing S. myrsinifolia (Paajanen et al. 2011) and B. pendula (Lavola et al. 2013). In S. repens, the highest increase in biomass was achieved under the combination of enhanced temperature and CO2-level (14–38%) (I), which was also seen for P. cathayana Rehder (Zhao et al.
2012). This may be a result of accelerated photosynthesis (Curtis et al. 2000, Jia et al.
2010, Kellomäki and Wang 2001). The benefit of enhanced CO2-level for photosynthesis may be useful for plants only after a certain temperature threshold level, when it is warm enough for stomata to open. A strong effect of higher temperature on growth was also seen in experiment III, as the leaves of S.
myrsinifolia were largest and thickest during the warmest summers. Enhanced UV- radiation, on the other hand, led to a decrease in the height and diameter growth of P. tremula (II). This is in accordance with results on P. cathayana and B. pendula (Tegelberg. et al. 2001, Xu et al. 2010). However, these effects were cancelled out by enhanced temperature (II).
The concentration of salicylates increased under elevated temperature in the leaves of the fast-growing P. tremula, tremulacin being the most abundant (II). On the contrary, elevated temperature caused dilution of the concentrations of flavonoids and phenolic acids as a result of growth increase in both of the ontogenetic phases, as found in earlier studies of S. myrsinifolia (Nybakken et al.
2012). Warming also cancelled the effect of the leaf ontogenetic phase on concentrations of (+)-catechin and condensed tannins in P. tremula trees (II). On the other hand, in the leaves of P. tremula, the strong effect of the ontogenetic shift on the concentration of salicylates was not altered by elevated temperature, though warming did increase the concentration of some individual salicylate compounds (II).
Elevated temperature and elevated CO2-level had opposite effects on the concentrations of salicylate and flavonoid compounds in the slow-growing S. repens (I). CO2 increased the most abundant flavonoids, quercetin 3-galactoside and quercetin-malonyl-glucoside, while elevated temperature decreased their concentrations. According to the carbon nutrient balance (CNB) and the growth differentiation balance (GDB) hypotheses (Bryant et al. 1983, Herms and Mattson 1992), this increase of flavonoids was to be expected in a slow-growing species prioritizing defence under nutrient-rich growing conditions with increased
38
resources. Elevated temperature had a stronger effect than did CO2 in S. repens, cancelling out the increasing effect of enhanced CO2-level on leaf phenolic concentrations, as earlier seen in B. pendula (Lavola et al. 2013). This shows that under favorable circumstances, a slow-growing species may also be able to favour growth before defence. Under enhanced CO2-levels, flavonoids increased and chlorogenic acids decreased at the same time, which is explained by the shared precursors in the phenylpropanoid pathway (Dixon et al. 2013, Gramazio et al.
2014).
The effect of enhanced UV-radiation on flavonoids and phenolic acids weakened with increasing age, when comparing the leaf phenolics of the second growing season (II), to those of the first growing season of P. tremula (Randriamanana et al.
2015a). In an earlier study of S. myrsinifolia, this weakened effect of UV was seen in growth parameters only after the second growing season (Nybakken et al. 2012, Randriamanana et al. 2015b), whereas in B. pendula the weakening effect of UV was not seen until the third year (Tegelberg et al. 2001). In addition, the effect of elevated temperature decreased with age. This shows that the plants may be able to adapt to both the increased UVB and temperature levels with increasing age.
However, each species may have a certain optimum temperature threshold limit, which the adaptation ability cannot fall under or exceed.
The response to climate change factors was also seen at the genetic level, as shown by the activity of miR168 in the leaves of the four-month-old P. tremula plantlets. The activity of miR168 was significantly higher under all climate change treatments; elevated temperature, UVB-radiation and their combination, compared with the control plants (Fig. 11). The level of miR168 activity did not differ between the treatments (F=2.510 and p=0.105, df 2), but the response to UVB tended to be stronger than the response to T and UVB+T. This implies a diminishing effect of T on the effect of UVB, as the miR168 activity response to single T and to the combined UVB+T treatments seemed to be the same. This corresponds to the responses seen for height and diameter growth and to accumulation of flavonoids in two-year old P. tremula. The activity of miR168 is linked to the regulation of multiple miRNA functions through the feedback action of AGO1 protein (Jia et al.
2009, Vaucheret et al. 2006, Vaucheret et al. 2009). In addition to the regulation of the ontogenetic phases, AGO1 protein might be involved in gene regulation when plants are acclimatizing to stressful conditions in normal plant development.
39 Figure 11. Relative expression values of miR168 in leaves of young Populus tremula plants grown under enhanced UVB-radiation (UVB), elevated temperature (T) and their combination (UVB+T). C refers to ambient conditions. Statistically significant differences (p≤0.01) between plants grown under ambient conditions (C) (expression value=1) and plants grown under different treatments are marked with the letters a and b. Bars represent mean values +SE
3.3 INFESTATION
In both S. myrsinifolia and P. tremula, the total low molecular weight phenolics level correlated negatively with rust severity (Appendix 2). In P. tremula, also total salicylates and phenolic acids correlated negatively with rust severity, whereas in S.
myrsinifolia, luteolins and single chlorogenic acid derivatives showed the strongest negative correlations with rust. Concentrations of (+)-catechin correlated positively with rust in both willow species (Appendix 2). The same increase in (+)-catechins and chlorogenic acids was found in some genotypes of S. myrsinifolia in a shorter inoculation experiment (Hakulinen 1998, Hakulinen et al. 1999). Salicin concentrations increased in S. myrsinifolia in the years in which herbivory damage was severe (Fig. 12, III). It may be that salicin can be induced in the fast-growing S.
myrsinifolia, either by increased synthesis or by degradation of the more complex salicylates, such as salicortin.
40
Figure 12. The number of plant individuals attacked by herbivores (a) and the eaten area (%) of affected leaves (b) within the studied female (black bars) and male (grey bars) genotypes of Salix myrsinifolia. In 2011 herbivory damage was measured only as the number of affected plants. Bars represent mean values +SE (n=8 for females and n=9 for males)
There was no significant change in the rust severity in S. myrsinifolia during ageing (III). Rust fungi need sufficient moisture to thrive (e.g. Helfer 2014), as was shown by the low levels of rust frequencies in leaves of S. myrsinifolia during warm (r2=-0.432, p<0.01) and dry (r2=0.345, p<0.01) summers. The levels were clearly higher during rainy summers (Appendix 2b). In addition, the temperature treatment reduced rust severity in the leaves of P. tremula, but only in plants at the earlier ontogenetic phase (II). These plants also grew higher, so that the tips of the plants were nearer to the heaters, causing drier and warmer conditions at the plant tip, compared to the slowergrowing plants at a later ontogenetic phase. Populus
41 tremula plants at the earlier ontogenetic phase had clearly higher concentrations of salicylates than did the plants at a later phase, which may also restrict the viability of rust fungi.
Herbivore damage did not show any significant trend with increasing age in S.
myrsinifolia (Fig. 12). Warming increased the damage caused by herbivores in two- year old P. tremula leaves (II) and in S. myrsinifolia during years 2012–2013 (Fig. 12), when the mean summer temperatures were higher than in 2008 (III). This may result from the decreased concentrations of flavonoids, phenolic acids and condensed tannins, as well as from higher nitrogen levels due to accelerated assimilation rates in the leaves of plants grown under elevated temperature (Ikonen 2002, Randriamanana et al. 2015a, Volf et al. 2015). Furthermore, the leaf areas of P.
tremula were larger under elevated temperature (II), and the amount of the rust pustules was lower, making them more palatable to herbivores.
In S. triandra L., a high leaf phenolics level correlated negatively with Melampsora amygdalinae rust severity, which was in contrast to the positive correlation with the egg-laying activity of a specialized insect herbivore Pontania triandrae (Hjältén et al. 2007). Also at genetic level, responses to herbivory and rust infection have been found to be opposite to each other in Populus (Miranda 2007).
By contrast, the leaves of S. myrsinifolia were severely damaged by both rust and herbivory in 2012 (Fig. 12, II).
S. myrsinifolia showed possible late induction capacity against herbivory and rust (Fig. 12, III). Levels of salicortin, chlorogenic acids and cinnamic acids increased one year after severe rust and herbivore attacks (Fig. 12, III, Appendix 1).
Severe damage may have caused a regeneration process in S. myrsinifolia, thereby increasing the synthesis of defence compounds. The regulation mechanisms of defence are also affected by the traits of herbivory. High levels of salicortin and tremulacin in P. tremuloides have reduced the amount of damage caused by a generalist gypsy moth (Lymantria dispar) (Donaldson and Lindroth 2007, Rubert- Nason et al. 2015), but increased the damage caused by a specialist leaf beetle Phratora vitellineae in P. tremula x P. tremuloides (Kosonen et al. 2012). Future climatic conditions may favour generalist species, which have a higher capacity for mobility than do specialist species (Uelmen et al. 2016).
In both field experiments (II and III), herbivores had unlimited possibilities to choose between plant individuals and individual leaves within plants. In the field, there are many different variables affecting infestation at the same time, including variation in light, moisture, temperature, and nutrients, in addition to other plant, invertebrate and vertebrate species and micro-organisms (e.g. Niinemets 2010). All these factors may have affected the lack of significant results from these field experiments.
42
3.4 SEX-RELATED AND GENOTYPIC DIFFERENCES
The slow-growing S. repens (I) showed sex-related differences in all growth parameters in the greenhouse, while such differences were rare in S. myrsinifolia (III) and P. tremula (II) in the field. The males of S. repens grew clearly faster than the females as regards most of the measured parameters (I). Under the stable growing conditions in the greenhouse, the effects of the studied factors may be clearer compared to those under the more fluctuating conditions in the field experiments.
For example, in a greenhouse experiment, females of S. myrsinifolia showed stronger growth than did males (Nybakken and Julkunen-Tiitto 2013). The number of replicates of the treatments, genotypes, as well as plant individuals within genotypes, may all have affected the level of singificance.
There were clear sex-specific differences in leaf phenolics in both S. repens (I) and in P. tremula (II), while in S. myrsinifolia (III) there were almost none. Males of S.
repens had more salicylates, while females had more flavonoids and phenolic acids (I). In P. tremula, males had clearly higher concentrations of (+)-catechins and condensed tannins, and this sex-related difference increased with age, from the first to the second growing season (II). A tendency towards sex-related differences in catechins and tannins was also seen in S. myrsinifolia, but the differences between genotypes were too pronounced to allow significant sex-related differences to emerge (III). In S. myrsinifolia, only one chlorogenic acid in the leaves was more abundant in females.
Although the sexes of S. repens did differ in terms of growth and leaf phenolics parameters, there were no significant sex-specific responses to elevated temperature or CO2-level (I). This is in contrast to an earlier study of the shrub S.
arctica Pall., where females were affected only by enhanced CO2 (Jones et al. 1999), and a study of P. cathayana, where males grew better under a combined treatment of enhanced temperature and CO2. In P. tremula, males had higher proportions of sharp-tipped leaves (which are typical of juvenile plants) under elevated temperature (II). This may be a sign that P. tremula males are more responsive to elevated temperature than are females. On the other hand, concentrations of condensed tannins were higher in P. tremula males under all the treatments, indicating a stronger effect of genotype than of temperature treatment (II). UV- treatments had only minor effects on leaf phenolics in P. tremula. Consequently, no sex-related differences were detected. This is in contrast to earlier findings of e.g.
Randriamanana et al. (2015), in which females were more responsive to UVB than males. Xu et al. (2010) found males of P. cathayana to be most tolerant to UVB- radiation. Neither S. myrsinifolia nor P. tremula plants showed significant sex- specific differences in their susceptibility to Melampsora-rust or herbivory in the field experiments. This is comparable to earlier field studies of S. myrsinifolia and P.
tremula (Nybakken et al. 2012, Randriamanana 2015a, b), although a meta-analysis
