Circumpolar variation in antibrowsing defense in tundra dwarf
birches
Elin Lindén
Degree Thesis in Ecology 60 ECTS
Master’s Level
Report passed: 13 January 2017
Supervisor: Johan Olofsson & Mariska te Beest
Abstract
The shrub encroachment (shrubification) currently happening in tundra ecosystems can result
in increased greenhouse gas emissions. Shrubification is in turn generally explained to be
driven by increased temperature, but there is regional variation in shrub encroachment that
cannot be solely explained by climate. Instead, herbivory is proposed as a key factor since
browsing has been shown to regulate density of shrubs in the tundra. Furthermore, regional
variation in anti-browsing defense, i.e. various deterrent and/or toxic compounds, has been
hypothesized to control the herbivory pressure. Dwarf birches are present and often dominant
throughout the low arctic. They can be divided into two functional groups based on their antibrowsing defense, i.e. resinous and non-resinous birches. This study investigated the variation
in anti-browsing defense within and among different taxa of dwarf birches and the two
functional groups. We also examined if these differences in anti-browsing defense affects the
level of invertebrate damage in dwarf birch. We found that although there were clear
differences in terpenes between resinous and non-resinous shrubs, neither functional groups
nor taxa are sufficient to understand the circumpolar variation in defense compounds.
Moreover, the variation in chemical anti-browsing defense had no clear effect on the level of
invertebrate damage, indicating that many other factors than food quality regulate the
abundance and importance of herbivores. This study does, for the first time, reveal the
circumpolar variation in anti-browsing defense in dwarf birches, which will be vital for a
mechanistic understanding of the greening of the arctic in the future.
Keywords
Anti-browsing defense, dwarf birch, herbivory, plant secondary metabolites, tundra
Table of contents
1 Introduction ............................................................. 1
1.1 Background ..................................................................................... 1
1.2 Herbivory impact ............................................................................ 1
1.3 Anti-browsing defense by plant secondary metabolites (PSMs) ....... 1
1.4 Defense in tundra dwarf birches ..................................................... 2
1.5 Can variation in anti-browsing defense explain shrubification
patterns? .............................................................................................. 2
1.6 Aim and hypotheses ........................................................................ 3
2 Method and materials .............................................. 3
2.1 Study area and sampling ................................................................. 3
2.1.1 Study area ............................................................................................ 3
2.1.2 Sampling and sample selection ............................................................ 4
2.2 Analyses.......................................................................................... 4
2.2.1 Targeted metabolite profiling .............................................................. 4
2.2.2 Traditional condensed tannin analysis ................................................ 5
2.2.3 Plant traits........................................................................................... 5
2.3 Climate data.................................................................................... 5
2.4 Statistical analyses .......................................................................... 6
3 Results ..................................................................... 6
3.1 Circumpolar pattern in anti-herbivory defense in shrub birches ..... 6
3.1.1 General pattern .................................................................................... 6
3.2 Testing interspecific variation ........................................................ 7
3.2.1 Secondary metabolites ......................................................................... 7
3.2.2 Other traits .......................................................................................... 9
3.3 Invertebrate herbivory.................................................................... 9
3.4 Traditional analytical methods, butanol acid assay ....................... 10
4 Discussion and conclusions ................................... 10
4.1 Large variation in anti-browsing defense ...................................... 10
4.2 No optimization of anti-browsing defense ...................................... 11
4.3 Herbivory ..................................................................................... 12
4.4 Analyzing plant secondary metabolites ......................................... 12
4.5 Conclusions .................................................................................. 13
5 References ............................................................. 15
1 Introduction
1.1 Background
In arctic tundra ecosystems, there is an ongoing shrub encroachment (shrubification) believed
to lead to a large number of negative effects on the ecosystem. Due to change in albedo (Sturm
et al. 2005), nutrient cycling and carbon balance (Weintraub and Schimel 2005; Natali et al.
2011) an increase in shrub cover on the tundra can increase greenhouse gas emission resulting
in a positive feedback loop between global warming and shrubification (Chapin et al. 2005;
Swann et al. 2010; Zhang et al. 2013). Furthermore, there are concerns that an increased
amount of shrubs would lead to diversity loss in the tundra plant community by invading and
possibly outcompete other tundra vegetation types (Post and Pedersen 2008). Shrub
encroachment in tundra ecosystems is generally explained by a close connection to global
warming (Walker et al. 2006). That shrubs indeed do benefit from warming have been shown
in numerous studies many of them including the genus Betula (Post and Pedersen 2008;
Christie et al. 2015). There is however a strong regional variation in tundra shrub response to
warming that cannot solely be explained by warming (Elmendorf et al. 2012; Xu et al. 2013).
In order to display this, Myers-Smith et al. (2015) carried out a circumpolar study showing a
greater sensitivity to summer temperatures in European tundra shrubs than in North
American ones. The European shrubs also had a positive sensitivity, i.e. increased growth,
while increased temperature rather had a negative or no effect on North American shrub
growth. This pattern was, however, not completely homogeneous and exceptions were found
in both regions.
1.2 Herbivory impact
One factor that can contribute to explaining global shrubification patterns is herbivory
(Olofsson et al. 2009; Myers-Smith et al. 2011; Christie et al. 2015). Large herbivores such as
reindeer, sheep and muskoxen, especially in high densities, are known to prevent shrub
expansion in many systems throughout the arctic by browsing (Olofsson et al. 2001, 2009,
2013; den Herder and Niemelä 2003; Post and Pedersen 2008; Eysteinsson 2009; Hofgaard
et al. 2010; Speed et al. 2010, 2011). Also, smaller herbivores such as invertebrates are known
to degrade tundra vegetation by severe defoliation during so called mass occurrences or
outbreak years (Jepsen et al. 2008; Kaukonen et al. 2013). However, there are also examples
of high densities of herbivores having transitory negative or even absent effect on growth and
spread in dwarf shrubs (Crête and Doucet 1998; Tremblay et al. 2012). It thus seems like shrubs
can respond differently to the same treatment, in this case browsing. This could be due to
varying sensitivity to browsing or that herbivores differ in their preference towards the shrubs.
Both these explanations could be connected to plant palatability or how well a plant is defended
against herbivory.
1.3 Anti-browsing defense by plant secondary metabolites (PSMs)
Plants have different ways of defending themselves against herbivores by reducing their
palatability. Nutrient quality and lignin content are two examples, but nearly all woody plants
also develop a separate anti-browsing defense containing chemical compounds (Kramer and
Kozlowski 1979) with a direct deterrent effect on herbivores (Bryant et al. 1991). This chemical
type of defense is generally composed of a variety of plant secondary metabolites (PSMs) and
has been recognized for several decades (Fraenkel 1959). PSMs are compounds that rather
than contributing to growth and development are specified for plants to survive in their
environment. Triterpenes and phenolic compounds, such as tannins, are carbon based PSMs
known to defend plants against herbivores and pathogens by decreasing palatability or even
act as toxins towards enemies (Reichardt et al. 1984; Robbins 2001; Mclean et al. 2009; Forbey
et al. 2011). Even within different PSM groups there is a wide diversity of compounds that can
act as deterrents separately but are also shown to have interactive effects (Cates 1996;
Gershenzon et al. 2012). How plants chemical anti-browsing defense is composed should
therefore connect to the extent at which they are consumed.
1
1.4 Defense in tundra dwarf birches
Dwarf birches are widely abundant in the tundra and can be divided into two functional groups
based on their difference in secondary metabolite content: Resinous birches and non-resinous
birches.
Resinous birches, such as B. glandulosa and B. nana ssp. exilis, are often considered to be the
better defended ones. Their twigs (current annual growth) are densely covered with resin
glands producing a resin rich in dammarane triterpenes as toxic papyriferic acid and 3-0malyonylebetulafolientrioloxide I (Risenhoover et al. 1985; Reichardt et al. 1987; Mclean et al.
2009; Forbey et al. 2011). Deterred feeding in mammals has been shown in experimental
studies where resin was added to otherwise palatable food (Bryant 1981; Reichardt et al. 1984;
Williams et al. 1992). Also in the field, a variety of mammalian herbivores within the tundra
biome have showed lower preference or even avoidance towards resinous birches (Reichardt
et al. 1984; Risenhoover 1989; White and Lawler 2002; DeAngelis et al. 2015). Insects, as well
as mammalian herbivores, have been recorded to have feeding preferences against resin
birches (McLean and Jensen 1985). Insect herbivory, both the so-called background herbivory
(herbivory at normal levels) and mass outbreaks, is expected to increase with increased
temperature (Barrio et al. 2016; Birkemoe et al. 2016). Therefore, variation in anti-browsing
defense should play a role also for the future pattern of insect herbivory. Although being well
defended against herbivores, browsing of resinous birches does occur, most commonly in B.
glandulosa (Crête and Doucet 1998; Tremblay et al. 2012; DeAngelis et al. 2015).
Non-resinous birches, on the other hand, are considered less defended with a defense
dominated by condensed tannins (Julkunen-Tiitto 1996; Graglia et al. 2001). Rather than
being toxic, condensed tannins can affect herbivores’ capacity to take up protein and/or have
an oxidative effect (Julkunen-Tiito et al. 1996). Although being defended by less effective
substances there are studies proposing that non-resinous birches invest more carbon in
phenolic compounds (i.e. tannins and flavonoids) than resinous birches do (Graglia et al.
2001). Since resinous birches have triterpenes and not tannins as their main defense
compound, this could mean that the dwarf birches invest their carbon in the best defense they
have available. Nevertheless, it is said that the strongest vegetation response to herbivory is
found within the non-resinous birch range (Fennoscandia, Iceland and Greenland), while the
effect often is less or absent in all in areas dominated by resinous birches (Canada, Alaska, East
Siberia) (Bryant et al. 2014).
The origin of these differences is not completely known, but there are theories proposing that
B. nana in previously glaciated areas, like Fennoscandia, Iceland and Greenland, have a lower
genetic diversity (Hewitt 1996; Alsos et al. 2002). This would be due to a more limited gene
bank to start with in combination with difficulties for sexual reproduction because of the ice
cover (Loveless et al. 1984; Bayer 1991; Bauert 1996; Taberlet 1998). In addition, natural
selection driven by herbivores eating the less bad plants can have enhanced regional
differentiation in anti-browsing defense (O'Reilly-Wapstra et al. 2012).
1.5 Can variation in anti-browsing defense explain shrubification
patterns?
It does seem like different kinds of birches differ in their chemical anti-browsing defense and
therefore also in their sensibility to browsing. In fact, Bryant et al. (2014) have presented a
hypothesis suggesting that the capacity of herbivores to suppress warming-induced increases
in Betula shrubs can be decided by the nature of their defense, i.e. if they are resinous or nonresinous. Furthermore, they propose that not only herbivory but also anti-browsing defense
should be included when modeling vegetation responses in the arctic as a response to a
temperature increase. In order to develop viable predictions of the actual effect of antibrowsing defense on regional patterns in shrub response to global warming, more knowledge
of the spatial variation in anti-browsing defense is needed and solid data are crucial.
2
1.6 Aim and hypotheses
The aim of this study is to get an improved spatial resolution of the trait variation within tundra
Betula shrubs, with a special focus on anti-browsing defense compounds. During one growth
season, extensive sampling across the circumpolar tundra vegetation has been carried out in
order to create a database to facilitate regional comparisons. This unique dataset is used to
address the questions: 1) Focusing on plant chemical defense, what is the level of trait variation
in shrub birches in tundra vegetation? and 2) Does intraspecific trait variation in dwarf birch
affect circumpolar patterns in background insect herbivory in the arctic tundra via chemical
defense?
To address these questions, the following hypotheses will be tested:
1. Anti-browsing defense composition in tundra dwarf birch is connected to what
functional group, i.e. resinous and non-resinous birches, they belong to.
2. Anti-browsing defense composition in tundra dwarf birch is taxon dependent.
3. Tundra dwarf birch invests their carbon in order to optimize their main defense against
herbivores, meaning that resinous birches are higher in triterpene content and nonresinous birches are higher in tannin content.
4. The composition of anti-browsing defense in tundra dwarf birch affects the level of
insect herbivory. The better-defended resinous birches will be less damaged by insects.
2 Method and materials
2.1 Study area and sampling
2.1.1 Study area
The study area covers circumpolar tundra vegetation in the Northern hemisphere at latitudes
between 47.3 and 74.5 °N (figure 1).
Figure 1. The distribution of sampling locations of four dwarf birch taxa.
3
Figure 2. Betula makes 2 kinds of shoots, long shoots and short shoots. Long shoots elongate during the season,
with younger and smaller leaves towards the top. Short shoots do not elongate and are made up of a cluster of
usually 3-4 leaves that are all of the same age (Source: www.biolib.de).
2.1.2 Sampling and sample selection
In the summer of 2014, sampling of both resinous and non-resinous dwarf birch (Betula nana
ssp. nana, Betula nana ssp. exilis, Betula glandulosa and Betula pumila; Table 1) was carried
out at 164 sites across 55 locations in tundra vegetation in the northern hemisphere. For each
site (area of ~10 m radius), 10 individuals were chosen. For each individual, 10 long shoots and
50 short shoot leaves (figure 2) were sampled making a total of 100 long shoots and 500
additional leaves per sampling site. Sampled shoots were allowed to air dry. Some of the in
total 164 samples were sampled in experimental sites where sampling was carried out both in
control plots and in treatment plots (reindeer exclosures and thermokarsts). Also, some of the
samples came from forest vegetation rather than tundra vegetation. In order to eliminate
treatment effects and to isolate the study to tundra vegetation a subsample of 128 samples
across 41 locations was selected, excluding all reindeer exclosures, thermokarsts, and forest
sites.
2.2 Analyses
For the chemical analyses of metabolites (both methods) and nitrogen and carbon, only short
shoot leaves were used. This was to get as much phenological heterogeneity as possible
between the samples. The dried leaves were ground using a ball mill.
2.2.1 Targeted metabolite profiling
To quantify plant secondary metabolites connected to anti-browsing defense, a targeted liquid
chromatography- mass spectrometry (LC/MS) analysis was carried out. LC/MS is an
analytical chemistry technique combining physical separation methods of liquid
chromatography and further detection by mass spectrometry (MS). In LC, liquid at high
pressure is used to force samples through a column packed with particles chosen to separate
out certain kinds of compounds. MS is used to further separate different metabolites from each
other by ionizing them and sort the ions by their mass-to-charge ratio. Ten mg of dried leaves
were ground and extracted according to (Gullberg et al. 2004). About 2 µl of the extracts were
4
injected into a onto an Acquity UPLC HSST3 column (2.1 x 50 mm, 1.8 µm C18) at 40 °C. The
gradient elution was A (H2O, 0.1 % formic acid) and B (75/25 acetonitrile:2-propanol, 0.1 %
formic acid) as follows: 0.1 - 10 % B over 2 minutes, B was increased to 99 % over 5 minutes
and held at 99 % for 2 minutes; returning to 0.1 % for 0.3 – 0.9 minutes, the flow-rate being
0.5 mL min-1. The compounds were detected with an Agilent 6540 Q-TOF mass spectrometer
equipped with an electrospray ion source operating in negative ion mode. This study focused
on the secondary metabolite classes previously reported as being involved in plant antibrowsing defense as follows: triterpenes, condensed tannins, hydrolysable tannins, complex
tannins, flavonoids and chlorogenic acid. So, a targeted MS approach was used based on the
diagnostic fragments produced during LC/MS analysis of the predetermined class of
metabolites. From the analysis, 112 metabolites were identified and grouped for the statistical
analysis. Many of the metabolites within the classes are not cited in previous literature since
they have never before been identified in Betula.
2.2.2 Traditional condensed tannin analysis
To be able to compare the LC/MS technique to more traditional quantification methods,
condensed tannins were analyzed colorimetrically using acid butanol assay (Porter 1986).
Colorimetric assays induce color change of analyte using reagent whereupon quantification
can be carried out measuring the absorbance of the tinted samples. Each sample was extracted
twice in 70% acetone with 1% ascorbic acid. Twenty µl of the extracts were mixed with 800 µl
reagent (95% butanol and 5% HCl) and 180 µl Milli-Q water. To induce color change reaction
in tannins, the mixtures were heated in 95 °C for 50 minutes. Absorbance was measured using
spectrophotometry and condensed tannin concentration was calculated using a standard curve
(concentration range: 2 – 64 µg of procyanidin B/µl extract).
2.2.3 Plant traits
Nitrogen and carbon isotopes: Ground leaf samples were sent to UC Davis Stable Isotope
Facility, University of California, to be analyzed for d15N and d13C. Samples were analyzed
using an elemental analyzer interfaces to a continuous flow isotope ratio mass spectrometer
(IRMS). Also total nitrogen and total carbon content were quantified in this process.
Specific Leaf Area (SLA): For each sample, 10 leaves were weighed and scanned. In some cases
the leaves needed to be pressed before scanning. Leaf area was then calculated by analyzing
the scanned images with the ROI manager in ImageJ software (Schneider et al. 2012). True
Specific Leaf Area is measured as [fresh leaf area/dry mass], since the samples for this study
were already dried leaf characteristics was measured as [dry leaf area/dry mass].
Gland count: for each sample, 10 twigs were defoliated, and a 15 mm segment starting 20 mm
from the top of the twig was photographed with a digital camera. The diameter of the twig at
the beginning and end of all segments were measured to calculate the area of which the count
was carried out (approximated to a half of the total segment bark area). From the digital
pictures all visible resin glands were counted. The number of glands was divided by the
examined area to give gland/mm2, thereafter mean gland/mm2 was calculated for the 10
replicates.
Insect herbivory: In total, 100 leaves from both long and short shoots were examined for
invertebrate damage. In some cases samples needed to be pressed to more accurately execute
area estimates. Observed damage was noted as % of damaged area out of the total area of 100
leaves. Proportion (%) of leaves with at least one kind of invertebrate damage was also noted.
2.3 Climate data
Climate data, such as annual precipitation, winter temperature (mean temperature of January)
and summer temperature (mean temperature of July) were extracted from the WorldClim
database (Hijmans et al. 2005). Interpolated grid data with a resolution of 30 arc-seconds (~1
5
km) were imported to the GIS software ArcMAP version 10.4.1. (ESRI 2011), and based on GPS
co-ordinates from the sampling locations, climate data were extracted for each sample.
2.4 Statistical analyses
For the statistical analyses, chemical compounds were classified into six different groups of
compounds; triterpenes, condensed tannins, hydrolysable tannins, complex tannins,
flavonoids, and chlorogenic acid. The compounds within each group were summed and all
statistical analyses were carried out on group level. The two insect herbivory measurements
were strongly correlated to each other (Spearman, rho=0.852, p<0.001) meaning that samples
with high damage per leaf also had a higher number of damaged leaves. Since counts of
damaged leaves are less sensitive to human errors than estimating damaged leaf area, damaged
leaf count was chosen to represent insect herbivory in the statistical analyses.
Similarities and differences in the chemical anti-browsing defense throughout tundra
vegetation in the Northern hemisphere were explored using nonmetric multidimensional
scaling analysis (NMDS; Minchin 1987). The NMDS analysis was performed using the
metaMDS function of the vegan package. Explanatory environmental variables where added
to the NMDS using the envfit function in said package (Oksanen et al. 2015). The effect of taxa
on invertebrate herbivory was tested using taxon-specific multiple regression analyses
computed by lm models. Secondary metabolite groups, nitrogen and carbon content, and
environmental variables were set as explanatory variables. The invertebrate herbivory data
were found to not follow a normal distribution and was therefore loge-transformed to meet the
criteria for analyses with parametric models. In order to compare results from the LC/MS
analysis with the traditional butanol acid assay, Spearman’s rank correlation tests was carried
out. All statistical analyses were performed using R software (version 3.0.2) (R Core Team
2016).
3 Results
3.1 Circumpolar pattern in anti-herbivory defense in shrub birches
3.1.1 General pattern
The mean scores from the NMDS analysis (stress level=0.11) showed a gradient in triterpene
to tannin content among the samples. That gradient seemed to be related to the first axis
(NMDS1) with higher triterpene content at low NMDS1 values (figure 3). The second axis was
mainly explained by a shift from condensed tannin to hydrolysable tannin content with more
condensed tannins at high NMDS2 values (figure 3). The fist axis was somewhat explained by
winter temperature and annual precipitation. Moreover, resin gland counts as well as SLA was
higher, i.e. more glands and thinner leaves, at low NMDS1 values. Conversely, procyanidin
content analyzed with traditional methods showed to be higher at high NMDS1 values. The
traditional procyanidin measurements strangely enough fell out independently from the
condensed tannins analyzed with the metabolomics method in the NMDS plot (see more in
section 3.4). The second axis was explained by latitude, summer temperature and sampling
date. Percentage of invertebrate damaged leaves fell out orthogonal to the triterpene-to-tannin
gradient pointing towards no clear connection between level of herbivory and chemical
defense, but rather an effect of environment and taxon. Annual precipitation and d15N was
higher at high NMDS2 values (figure 3).
There was a differentiation between resinous birches (B. nana ssp. exilis and B. glandulosa)
and non-resinous birches (B. nana ssp. nana and B. pumila) (figure 3) induced by a higher
triterpene content in the resin birches. The difference was however continuous and better
explained as a gradual change. Within the resin birches, the two species were not significantly
different in anti-herbivory defense. For the non-resinous birches, B. pumila is clearly
6
0.6
0.4
Bp
Bp
0.2
0.0
NMDS2
Bg
Bg
Bp
Bg
c_tan
day
invd_leaf
c_acid
Bn
glands
triter
BgBg
Bg Bne
BgBg
Bne
BneBgBne
Bg
Bg
BgBg Bne
Bg
Bne
Bne
Bne
Bne
Bg
Bne
Bne
Bne
Bne Bne
Bne
Bne
Bne
Bne
Bne
Bne
Bg
Bne BneBne
Bg Bne
Bg Bg
Bne
Bne
Bn
Bne
-0.2
SLA
ann_perc
d15N
Bn
Bn
BnBn
Bn BnBn
Bn
Bn
Bn
Bn
Bn
BnBn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
BnBn
BnBnBn Bn
Bn
Bn
Bn
BnBn
Bn
Bn
Bn
Bn
Bn
Bn Bn
Bn
BnBn Bn
BnBn BnBn
Bn Bn Bn
Bn
Bn
Bn
Bn
Bn
temp_jan
compl_tan procyanidin
h_tan
flav
Bne
Bne
Bp
Bn
Bn
Bn
Bn
-0.4
latitude
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
NMDS1
Figure 3. Non-Metric Dimensional Scaling ordination showing differences in chemical defense composition in
samples of Betula glandulosa (orange), B. nana ssp. exilis (green), B. nana ssp. nana (blue) and B. pumila (black).
Mean scores of the compared compound groups (triterpenes, condensed tannins, hydrolysable tannins, complex
tannins, flavonoids and chlorogenic acid) are written out in black. Joint plots show environmental variables and
plant traits in red. NMDS stress =0.11.
separated from B. nana ssp. nana along NMDS2 (figure 3) showing a difference in tannin
composition. No general difference in tannin composition between the two functional groups
was however detected.
3.2 Testing interspecific variation
3.2.1 Secondary metabolites
Both differences and similarities in composition of chemical anti-browsing defense were found
for the four Betula taxa. Triterpenes were higher in the resinous B. glandulosa and B. nana
ssp. exilis than in the non-resinous B. nana ssp. nana and B. pumila (Table 1; figure 4a). Both
condensed and hydrolysable tannin content was at the same level for the non-resinous B. nana
ssp. nana as for the two resinous taxa. (Fig 4b,c). For B. pumila mean condensed tannin
content was at least 60% higher (Table 1; figure 4b) and hydrolysable tannin content at least
83% lower (Table 1; figure 4c) than in the other taxa. The most common defense compound in
B. pumila was clearly condensed tannins while the other three taxa were highest in
hydrolysable tannin content (figure 4b,c). Looking at the complex tannin content, it was higher
in B. nana ssp. nana compared with the other three taxa and lower in B. pumila than in the
two B. nana subspecies (Table 1; figure 4d). Also flavonoid content was higher in Betula nana
7
Table 1. One-way ANOVAs testing interspecific variation in plant secondary metabolites involved in anti-browsing
defense in four dwarf birch taxa (B. glandulosa, B. nana ssp. nana, B nana ssp. exilis, B. pumila). Significant values
are written in bold.
Source of variation
Triterpene
Condensed tannin
Hydrolysable tannin
Complex tannin
Flavonoid
Chlorogenic acid
Taxon
df
3
3
3
3
3
3
F
230
18.87
7.54
54.69
51.87
1.43
p
<0.001
<0.001
<0.001
<0.001
<0.001
0.237
Figure 4. Boxplots representing triterpene (a), condensed tannin (b), hydrolysable tannin (c), complex tannin (d)
and flavonoid (e) content as well as specific leaf area (f), resin gland count on current year twigs (g) and stable
isotope content as d15N (h) and d13C (i) in samples from four dwarf birch taxa (B. glandulosa, B. nana ssp. exilis,
B. nana ssp. nana and B. pumila). The boxplots contain of median (thick line), inner quartile (box) and outer
quartile (error bar). The inner and outer quartiles indicate the degree of dispersion within the samples, present
outliers can be seen as circles.
ssp. nana while it was lower in B. pumila than in any other taxa (Table 1; figure 4e). There was,
however, no difference in chlorogenic acid content between the four dwarf birch taxa (S2).
8
Table 2. One-way ANOVAs testing interspecific variation in plant traits in four dwarf birch taxa (B. glandulosa, B.
nana ssp. nana, B. nana ssp. exilis, B. pumila). Significant values are written in bold.
Source of variation:
Carbon
Nitrogen
d13C
d15N
SLA
Gland count
Invertebrate damage
Species
df
3
3
3
3
3
3
3
F
0.18
2.42
8.14
7.68
8.67
232
3.18
p
0.911
0.069
<0.001
<0.001
<0.001
<0.001
0.027
3.2.2 Other traits
There was no statistical difference in carbon or nitrogen content between the taxa (S3a,b).
Specific Leaf Area was higher in B. nana ssp. exilis than B. nana ssp. nana, meaning that the
resinous sub-species has thinner leaves than the non-resinous one (Table 2; figure 4f). Resin
gland count was significantly higher in B. glandulosa and B. nana ssp. exilis than in B. nana
ssp. nana and B. pumila (Table 2; figure 4g). B. nana ssp. exilis was lower in d15N than the
non-resinous birches but did not differ from B. glandulosa (Table 2; figure 4h) in this trait. B.
nana ssp. exilis and B. nana ssp. nana was lower in d13C than B. glandulosa. Additionally,
d13C was lower in B. nana ssp. exilis than in B. pumila (Table 2; figure 4i)
3.3 Invertebrate herbivory
Statistical difference was found between the four Betula taxa in number of invertebratedamaged leaves (Table 2), but the difference did, however, not survive the post hoc test.
However graphical examinations implies a difference between species (figure 5). When further
investigating the cause of variation in invertebrate-damaged leaves among the taxa, no clear
trend was found. Higher invertebrate damage in B. nana ssp. nana was explained to 33% by
lower nitrogen, flavonoid and complex tannin content, as well as higher summer temperature
Figure 5. Boxplot representing number of invertebrate damaged leaves out of 100 leaves in samples from four
dwarf birch taxa (B. glandulosa, B. nana ssp. exilis, B. nana ssp. nana and B. pumila). The boxplots contain of
median (thick line), inner quartile (box) and outer quartile (error bar). The inner and outer quartiles indicate the
degree of dispersion within the samples, present outliers can be seen as circles.
9
Table 3. Multiple linear regression model describing the effect of plant chemistry and environmental variables on
invertebrate damage in three dwarf taxa (B. glandulosa, B. nana ssp. exilis, B. nana ssp. nana).
Betula glandulosa
Betula nana ssp. exilis
Betula nana ssp. nana
β
β
t
p
β
t
p
0.81
3.58
<0.01
-0.445
-3.66
<0.001
-0.26
-2.15
<0.01
Flavonoid
-0.36
-3.10
0.035
Summer
temp.
0.23
2.06
0.04
0.36
3.06
<0.01
t
p
Nitrogen
Condensed
tannin
Complex
tannin
Triterpene
Annual prec.
0.91
0.96
0.90
4.46
4.16
4.60
<0.001
<0.01
-0.72
-3.21
<0.01
<0.001
Total model:
F
(3,13)=10.01
(2,16)=7.31
(5,59)=5.87
R2
0.698
0.477
0.332
p
<0.01
<0.01
<0.001
and annual precipitation (Table 3). For B. nana ssp. exilis 48% of the variation could be
explained by a higher nitrogen and lower triterpene content (Table 3). Moreover, an increase
in invertebrate damage in B. glandulosa could be explained to 69.8% by higher triterpene and
condensed tannin content and higher annual precipitation (Table 3).
3.4 Traditional analytical methods, butanol acid assay
In the NMDS analysis, the traditional procyanidin measurements showed to be independent
from the condensed tannins analyzed with the metabolomics method (figure 3). A correlation
test confirmed this independency (rho=0.091, p=0.31). When further comparing the
measurements from the butanol acid assay to all separate metabolites analyzed with the
LC/MS, they were correlated with many of the condensed tannins (rho=0.224-0.519) but also
uncorrelated or even negatively correlated with several of them (S1). Correlations were also
found to compounds from other metabolite groups such as hydrolysable tannins and
flavonoids (S1).
4 Discussion and conclusions
In this study we wanted to study the circumpolar variation in anti-browsing defense in tundra
dwarf birches. The results show a difference in anti-browsing defense between resinous and
non-resinous dwarf birches in terms of triterpene content but not in terms of tannin content.
Large variations in anti-browsing defense within the two functional groups, but also within
species and subspecies of the investigated tundra dwarf birches, were also found. No clear
connection was, however, found between chemical anti-browsing defense and insect herbivory
damage. Lastly, the comparison of the traditional analytical methods and our LC/MS analysis
showed poor correlations for many of the compounds that are said to be quantified by the
traditional method.
4.1 Large variation in anti-browsing defense
The first hypothesis of this study was that anti-browsing defense composition is connected to
if a birch is resinous or non-resinous. This connects to the hypothesis of Bryant et al. (2014)
10
meaning that the response of herbivory on warming-induced shrub increase in tundra regions
could be explained by dominance of resinous versus non-resinous birches. For their hypothesis
to be true there must be some kind of homogeneity within the two functional groups when it
comes to anti-browsing defense, i.e. my first hypothesis must be true. The results from this
study however show that it is not this straightforward. The resinous birches do indeed have
more triterpenes than non-resinous birches, but there are no unanimous differences in tannin
content between the two groups. Also, the variation within the two functional groups is large.
The two non-resinous birches B. pumila and B. nana ssp. nana are for example differentiated
in anti-browsing defense composition. Not only do they differ in condensed tannin and
hydrolysable tannin content, B. pumila also contains far less complex tannins and flavonoids
compared to B. nana ssp. nana. Even though the two resinous birch species do not differentiate
from each other there is still a significant variation in anti-browsing defense within the
functional group, most of it connected to differences in tannin content. It is thus not completely
straightforward to define the nature of the anti-browsing defense in dwarf birch vegetation by
functional group. Estimating regional response of Arctic birch shrubs with increased
temperature based on if resinous or non-resinous birches are dominating would therefore
cause large errors.
Moreover, the second hypothesis, that anti-browsing defense should be taxon dependent, is
proven wrong since there is generally large variations in defense composition also within all
the studied birch taxa. The NMDS analysis show scattered results for all four taxa, and when
looking at the content of the secondary metabolite groups on taxon level, the range is generally
large. Even generalizing vegetation response to herbivory on taxon level might therefore lead
to misinterpretations, at least when focusing on anti-herbivory defense.
The origin of the large variation in anti-browsing defense even on taxon level is to be further
investigated. There are studies showing that differences soil nutrient resources affects the
production of at least phenolic defense compounds, where an increase in plant available
nitrogen results in a decrease in chemical defense (Keinanen et al. 1999; Stark et al. 2015).
Moreover, the variation in genetic diversity throughout the Arctic with generally lower genetic
diversity in previously glaciated areas (Hewitt 1996; Alsos 2002) should not be neglected. More
studies are needed in order to fully understand the genetic diversity within the taxon and to
examine if anti-browsing defense diversity can be connected to genetic diversity. Furthermore,
this study did not investigate any spatial trends in the variation of anti-browsing defense. There
is still a possibility that anti-browsing defense can affect shrubification just not at a large
regional scale and not with the regards to functional group or taxa. Apart from anti-browsing
defense there are however other factor that could affect if variabilities in shrub expansion such
as nutrient and carbon resources and hydrological conditions. These factors should also be
taken into further consideration in future studies.
4.2 No optimization of anti-browsing defense
The third hypothesis of this study, that tundra dwarf birch invests their carbon in order to
optimize their main defense against herbivores, has been suggested before. Graglia et al. (2001)
found higher concentrations of phenolic compounds in non-resinous B. nana from Abisko
compared to resinous B. nana from Toolik. In contrast to this, and to hypothesis 3 of this study,
the results do not reveal any clear optimization of main chemical defense. B. nana ssp. nana
does not differ much from the resinous birches when it comes to how much condensed or
hydrolysable tannins they contain. In other words, just because tannins are the major
secondary metabolite involved in non-resinous birch defense they do not necessarily have to
contain more of those compounds. A defense optimization could be found in B. pumila; they
seem to be almost exclusively defended by condensed tannins. Although, the B. pumila
population is relatively small and isolated and does not contribute a lot to large-scale regional
patterns, it does provide important information of the complexity of anti-browsing defense in
Betula shrubs.
11
Although there is not necessarily a difference in phenolic compound content in tundra dwarf
birches, when including triterpenes there is a quantitative difference in carbon based
secondary metabolites. Furthermore, as Graglia et al. (2001) proposed there is no difference in
carbon content between the taxa. Resinous birches therefore seem to invest more of its carbon,
in total, in secondary metabolites than non-resinous birches. This gets even further support in
the SLA data. Resinous birches that invest more carbon for secondary metabolites have a
higher SLA, i.e. thinner leaves, whereas the less chemically defended Betula nana ssp. nana
has lower SLA. Leaf structure can also affect plant palatability. For example, higher lignin
content is considered to reduce plant palatability. Even though this study cannot confirm that
this is where the ‘left over’ carbon ends up it is definitely a possibility. This result does,
however, indicate a trade-off between investing in chemical defense or leaf structure.
4.3 Herbivory
We hypothesized that variations in anti-browsing defense would lead to variations in insect
herbivory damage where better-defended taxa would show less invertebrate damage. Despite
differences in anti-browsing defense among the taxa, no substantial difference was found in
background insect herbivory. Herbivores, both invertebrates and mammalians, can find ways
of coping with low quality food and there are examples of herbivore counter-adaptions and
offenses towards plant anti-browsing defense. Feeding choices (like keeping a mixed diet),
enzymatic metabolism of defense compounds and sequestration of such compounds (Karban
and Agrawal 2002; Sorensen and Dearing 2006) are some examples. Another factor that
obviously affects the level of herbivory damage is herbivore density. For example, our study
only included background herbivory. Thus, we cannot say if there are any connections between
what areas that are more or less affected during mass outbreaks and spatial variation in antibrowsing defense. When it comes to larger herbivores, such as reindeer, densities can be
affected by weather events diminishing their food supply. For example, in 2006 and 2013 so
called rain on snow events on the Yamal Peninsula led to massive winter mortalities in the
reindeer population in the region (Forbes et al. 2016).
That no effect of anti-browsing defense was found on invertebrate damage does however not
mean that anti-browsing defense has no effect on herbivores or herbivory levels at all. Without
chemical defense the shrubs would most certainly not survive, if not eaten by invertebrates or
larger herbivores, they would not be able to cope with fungi or other pathogens (Kramer and
Kozlowski 1979). Moreover, the fact that anti-browsing defense has evolved in such a variety
despite its high cost for the plant is strong support for its necessity. The result does, however,
indicate that other factors than just food quality regulate both the abundance and importance
of herbivores.
4.4 Analyzing plant secondary metabolites
Our results from the LC/MS analysis show that both Betula nana sub-species and Betula
glandulosa have more hydrolysable tannins than condensed tannins. This does not mean that
hydrolysable tannins should be considered as the major secondary metabolite in the nonresinous birch anti-browsing defense. Condensed tannins can still be the more effective of the
two tannins when it comes to defense against browsing. It does however tell us something
about the importance of incorporating new and better methods of analyzing ecological samples
in order to improve our knowledge. Previous literature state that the major secondary
metabolite involved in the defense in non-resinous birches is condensed tannins (JulkunenTiito et al. 1996). In the study by Graglia et al. (2001), condensed tannins were shown to be the
most abundant phenolic compound in both resinous and non-resinous Betula nana. Using
butanol acid assay for analyzing condensed tannins and HPLC (high performance light
chromatography) methods previously described (Julkunen-Tiitto 1989; Julkunen-Tiitto et al.
1996) for analyzing hydrolysable tannins they measured the condensed tannin content to be at
least 100 times higher than the hydrolysable tannins content. In our study, the comparison of
the LC/MS analysis and the traditional butanol acid assay showed that the results from the two
methods were more or less decoupled from each other. There was no correlation at all between
12
the procyanidin content measured with the butanol acid assay and the summed condensed
tannins from the LC/MS. When comparing each secondary metabolite to the procyanidin
content from the butanol acid assay several condensed tannins were not at all correlated
whereas some were correlated. Also, secondary metabolites from other classes, such as
flavonoids, which are precursors (building blocks) to condensed tannins, were at least as
correlated as some of the condensed tannins. There is, in other words, a big uncertainty in what
the traditional quantification methods can tell us about plant anti-browsing defense.
Incorporating the LC/MS metabolomics technique has not only resulted in more exact
quantifications of secondary compounds, it has also contributed to more detailed data
including identification of many secondary metabolites not previously found in Betula.
Moreover, this technique allows us to analyze all kinds of compounds with the same method
giving us the opportunity to compare relative differences between analytes. This technique
definitely opens up new doors for future phytochemical studies by providing high-resolution
data.
4.5 Conclusions
For the first time, this study unveils the circumpolar variation in anti-browsing defense in
dwarf birches. This information will be of great help for understanding what the variation
might mean to the shrubification in the Arctic. It can be concluded that neither functional
group nor taxon can provide a valid interpretation of the fate of shrub vegetation with
increased temperature when looking at anti-browsing defense. More studies including genetic
diversity, herbivore density and possible spatial effects of variation in anti-browsing defense
are, however, necessary in order to examine further factors regulating the impact of herbivores
on vegetation and vice versa.
13
Acknowledgements
First and foremost I want to thank Johan Olofsson and Mariska te Beest for trusting me with
this project. You have given me a lot of freedom to take my own decisions but also given me
both scientific and moral support when I needed it. I also want to thank Ilka Abreu and Thomas
Moritz at the Umeå Plant Science Center for a nice collaboration. A special thanks to Ilka for
carrying out the metabolomics analyses and for guiding me in the butanol acid assay. Lastly I
really must express my warmest gratitude to my wonderful friends. Thanks for cheering me
on, driving me, comforting me, letting me spend time in my “thesis-cave” and for occasionally
dragging me out of it. A special thanks to Cecilia Ribbefors, Madelene Fridell, Helena Dahlberg,
Petter Johansson and Emily Pickering Pedersen.
14
5 References
Alsos, I. G., Engelsskjøn, T., Brochmann, C. 2002. Conservation genetics and population
history of Betula nana, Vaccinium uliginosum and Campanula rotundifolia in Arctic
archipelago of Svalbard. Arctic, Anarctic and Alpine Research, 34: 408-418.
Barrio, I. C., Bueno, C. G. & Hik, D. S. 2016. Warming the tundra: reciprocal responses of
invertebrate herbivores and plants. Oikos, 125: 20-28.
Bauert, M. R. 1996. Genetic diversity and ecotypic differentiation in arctic and alpine
populations of Polygonum viviparum. Arctic and alpine research, 28: 190-195.
Bayer, R. J. 1991. Patterns of colonal diversity in geographically mariginal populations of
Annetaria rosea (Asteraceae, Inuleae) from sub-arctic Alaska and Yukon territory.
Botanical gazette, 152: 486-493.
Birkemoe, T., Bergmann, S., Hasle, T. E. & Klanderud, K. 2016. Experimental warming
increases herbivory by leaf-chewing insects in an alpine plant community. Ecology
and Evolution, 6: 6955-6962.
Bryant, J. P. 1981. Phytochemical Deterrence of Snowshoe Hare Browsing by Adventitious
Shoots of 4 Alaskan Trees. Science, 213: 889-890.
Bryant, J. P., Joly, K., Chapin, F. S., Deangelis, D. L. & Kielland, K. 2014. Can antibrowsing
defense regulate the spread of woody vegetation in arctic tundra? Ecography, 37:
204-211.
Bryant, J. P., Provenza, F. D., Pastor, J., Reichardt, P. B., Clausen, T. P. & Dutoit, J. T. 1991.
Interactions between Woody-Plants and Browsing Mammals Mediated by Secondary
Metabolites. Annual Review of Ecology and Systematics, 22: 431-446.
Cates, R. G. 1996. The role of mixtures and variation in the production of terpenoids in
conifer-insect-pathogen interactions. Phytochemical Diversity and Redundancy in
Ecological Interactions, 30: 179-216.
Chapin, F. S., Sturm, M., Serreze, M. C., Mcfadden, J. P., Key, J. R., Lloyd, A. H., Mcguire, A.
D., Rupp, T. S., Lynch, A. H., Schimel, J. P., Beringer, J., Chapman, W. L., Epstein, H.
E., Euskirchen, E. S., Hinzman, L. D., Jia, G., Ping, C. L., Tape, K. D., Thompson, C.
D. C., Walker, D. A. & Welker, J. M. 2005. Role of land-surface changes in Arctic
summer warming. Science, 310: 657-660.
Christie, K. S., Bryant, J. P., Gough, L., Ravolainen, V. T., Ruess, R. W. & Tape, K. D. 2015.
The Role of Vertebrate Herbivores in Regulating Shrub Expansion in the Arctic: A
Synthesis. Bioscience, 65: 1123-1133.
Crete, M. & Doucet, G. J. 1998. Persistent suppression in dwarf birth after release from heavy
summer browsing by caribou. Arctic and Alpine Research, 30: 126-132.
Deangelis, D. L., Bryant, J. P., Liu, R. S., Gourley, S. A., Krebs, C. J. & Reichardt, P. B. 2015. A
plant toxin mediated mechanism for the lag in snowshoe hare population recovery
following cyclic declines. Oikos, 124: 796-805.
Den Herder, M., Niemelä, P. 2003. effects of reindeer on the re-establishment of Betula
pubescens subsp. czerepanovii and Salix phylicifolia in a subarctic meadow. Rangifer,
23: 3-12.
Elmendorf, S. C., Henry, G. H. R., Hollister, R. D., Bjork, R. G., Boulanger-Lapointe, N.,
Cooper, E. J., Cornelissen, J. H. C., Day, T. A., Dorrepaal, E., Elumeeva, T. G., Gill, M.,
Gould, W. A., Harte, J., Hik, D. S., Hofgaard, A., Johnson, D. R., Johnstone, J. F.,
Jonsdottir, I. S., Jorgenson, J. C., Klanderud, K., Klein, J. A., Koh, S., Kudo, G., Lara,
M., Levesque, E., Magnusson, B., May, J. L., Mercado-Diaz, J. A., Michelsen, A.,
Molau, U., Myers-Smith, I. H., Oberbauer, S. F., Onipchenko, V. G., Rixen, C.,
Schmidt, N. M., Shaver, G. R., Spasojevic, M. J., Porhallsdottir, P. E., Tolvanen, A.,
Troxler, T., Tweedie, C. E., Villareal, S., Wahren, C. H., Walker, X., Webber, P. J.,
Welker, J. M. & Wipf, S. 2012. Plot-scale evidence of tundra vegetation change and
links to recent summer warming. Nature Climate Change, 2: 453-457.
Esri 2011. ArcGIS Desktop: Release 10. Redlands, CA: Environmental Systems Research
Institute.
15
Eysteinsson, T. 2009. Forestry in a treeless land. Iceland forest service. Midvangi 2-4, IS700, Egilstadir, Iceland.
Forbes, B. C., Kumpula, T., Meschtyb, N., Laptander, R., Macias-Fauria, M., Zetterberg, P.,
Verdonen, M., Skarin, A., Kim, K-L., Boisvert, L.N., Stroeve, A.B. 2016. Sea ice, rainon-snow and tundra reindeer nomadism in Arctic Russia. Biology Letters, 12.
Forbey, J. S., Pu, X. Z., Xu, D., Kielland, K. & Bryant, J. 2011. Inhibition of Snowshoe Hare
Succinate Dehydrogenase Activity as a Mechanism of Deterrence for Papyriferic Acid
in Birch. Journal of Chemical Ecology, 37: 1285-1293.
Fraenkel, G. S. 1959. The raison d'etre of secondary plant substances. Science, 129: 14991470.
Gershenzon, J., Fontana, A., Meike, B., Wittstock, U., Degenhardt, J. 2012. Mixtures of plant
secondary metabolites: metabolic origins and ecological benefits. In: GLENN R.
IASON, M. D. A. S. E. H. (ed.) The ecology of plant secondary metabolites: From
genes to global processes. New York: Cambridge Univeristy Press.
Graglia, E., Julkunen-Tiitto, R., Shaver, G. R., Schmidt, I. K., Jonasson, S. & Michelsen, A.
2001. Environmental control and intersite variations of phenolics in Betula nana in
tundra ecosystems. New Phytologist, 151: 227-236.
Gullberg, J., Jonsson, P., Nordstrom, A., Sjostrom, M. & Moritz, T. 2004. Design of
experiments: an efficient strategy to identify factors influencing extraction and
derivatization of Arabidopsis thaliana samples in metabolomic studies with gas
chromatography/mass spectrometry. Analytical Biochemistry, 331: 283-295.
Hewitt, G. M. 1996. Some genetic consequences of ice ages, and their role in divergence and
speciation. Biological Journal of the Linnean Society, 58: 247-276.
Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. 2005. Very high
resolution interpolated climate surfaces for global land areas. International Journal
of Climatology, 25: 1965-1978.
Hofgaard, A., Lokken, J. O., Dalen, L. & Hytteborn, H. 2010. Comparing warming and
grazing effects on birch growth in an alpine environment - a 10-year experiment.
Plant Ecology & Diversity, 3: 19-27.
Jepsen, J. U., Hagen, S. B., Ims, R. A. & Yoccoz, N. G. 2008. Climate change and outbreaks of
the geometrids Operophtera brumata and Epirrita autumnata in subarctic birch
forest: evidence of a recent outbreak range expansion. Journal of Animal Ecology, 77:
257-264.
Julkunen-Tiitto, R. 1989. Distribution of certain phenolics in Salix species (Salicaceae),
Joensuu: University of Joensuu.
Julkunen-Tiitto, R. R., M.; Bryant, J.; Sorsa, S.; Keinänen, M.; Sikanen, H. 1996. Chemical
diversity of several Betulaceae species: comparison of phenolics and terpenoids in
northern stems. Trees, 11: 16-22.
Karban, R., Agrawal, A. A. 2002. Herbivore offense. Annual Review of Ecology and
Systematics, 33: 641-664.
Kaukonen, M., Ruotsalainen, A. L., Wali, P. R., Mannisto, M. K., Setala, H., Saravesi, K.,
Huusko, K. & Markkola, A. 2013. Moth herbivory enhances resource turnover in
subarctic mountain birch forests? Ecology, 94: 267-272.
Keinanen, M., Julkunen-Tiitto, R., Mutikainen, P., Walls, M., Ovaska, J. & Vapaavuori, E.
1999. Trade-offs in phenolic metabolism of silver birch: Effects of fertilization,
defoliation, and genotype. Ecology, 80: 1970-1986.
Kramer, P. J., Kozlowski, T.T. 1979. Physiology of woody plants. , New York: Academic
press.
Loveless, M. D. a. H., J. L. 1984. Ecological determinants of genetic structure in plant
populations. Annual Review of Ecology and Systematics, 15: 65-95.
Mclean, S., Richards, S. M., Cover, S. L., Brandon, S., Davies, N. W., Bryant, J. P. & Clausen,
T. P. 2009. Papyriferic Acid, An Antifeedant Triterpene From Birch Trees, Inhibits
Succinate Dehydrogenase From Liver Mitochondria. Journal of Chemical Ecology,
35: 1252-1261.
Mclean, S., Jensen, S.J. 1985. Food plant selection by insect herbivores in Alascan arctic
tundra: the role of plant life form. Oikos, 44: 211-221.
16
Minchin, P. R. 1987. An Evaluation of the Relative Robustness of Techniques for Ecological
Ordination. Vegetatio, 69: 89-107.
Myers-Smith, I. H., Elmendorf, S. C., Beck, P. S. A., Wilmking, M., Hallinger, M., Blok, D.,
Tape, K. D., Rayback, S. A., Macias-Fauria, M., Forbes, B. C., Speed, J. D. M.,
Boulanger-Lapointe, N., Rixen, C., Levesque, E., Schmidt, N. M., Baittinger, C., Trant,
A. J., Hermanutz, L., Collier, L. S., Dawes, M. A., Lantz, T. C., Weijers, S., Jorgensen,
R. H., Buchwal, A., Buras, A., Naito, A. T., Ravolainen, V., Schaepman-Strub, G.,
Wheeler, J. A., Wipf, S., Guay, K. C., Hik, D. S. & Vellend, M. 2015. Climate sensitivity
of shrub growth across the tundra biome. Nature Climate Change, 5: 887-+.
Myers-Smith, I. H., Forbes, B. C., Wilmking, M., Hallinger, M., Lantz, T., Blok, D., Tape, K.
D., Macias-Fauria, M., Sass-Klaassen, U., Levesque, E., Boudreau, S., Ropars, P.,
Hermanutz, L., Trant, A., Collier, L. S., Weijers, S., Rozema, J., Rayback, S. A.,
Schmidt, N. M., Schaepman-Strub, G., Wipf, S., Rixen, C., Menard, C. B., Venn, S.,
Goetz, S., Andreu-Hayles, L., Elmendorf, S., Ravolainen, V., Welker, J., Grogan, P.,
Epstein, H. E. & Hik, D. S. 2011. Shrub expansion in tundra ecosystems: dynamics,
impacts and research priorities. Environmental Research Letters, 6.
Natali, S. M., Schuur, E. a. G., Trucco, C., Pries, C. E. H., Crummer, K. G. & Lopez, A. F. B.
2011. Effects of experimental warming of air, soil and permafrost on carbon balance
in Alaskan tundra. Global Change Biology, 17: 1394-1407.
O'reilly-Wapstra, J. M., McArthur, C., Potts, B.M. 2012. Natural selection for anti-herbivore
plant secondary metabolites: a Eucaluptys system. In: GLENN R. IASON, M. D. A. S.
E. H. (ed.) The ecology of plant secondary metabolites: From genes to global
processes. New York: Cambridge university press.
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R, O'hara, R.G. 2015. Vegan:
Community ecology package. R package version 2.3.0 http://CRAN.Rproject.org/package=vegan.
Olofsson, J., Kitti, H., Rautiainen, P., Stark, S. & Oksanen, L. 2001. Effects of summer grazing
by reindeer on composition of vegetation, productivity and nitrogen cycling.
Ecography, 24: 13-24.
Olofsson, J., Oksanen, L., Callaghan, T., Hulme, P. E., Oksanen, T. & Suominen, O. 2009.
Herbivores inhibit climate-driven shrub expansion on the tundra. Global Change
Biology, 15: 2681-2693.
Olofsson, J., Te Beest, M. & Ericson, L. 2013. Complex biotic interactions drive long-term
vegetation dynamics in a subarctic ecosystem. Philosophical Transactions of the
Royal Society B-Biological Sciences, 368.
Porter, L. J., Hirstich, L. N., Chan, B.G. 1986. The conversion of procyanidins and
prodelphinidins to cyanidin and delphinidin. Phytochemistry, 25: 223-230.
Post, E. & Pedersen, C. 2008. Opposing plant community responses to warming with and
without herbivores. Proceedings of the National Academy of Sciences of the United
States of America, 105: 12353-12358.
Reichardt, P. B., Bryant, J. P., Clausen, T. P. & Wieland, G. D. 1984. Defense of WinterDormant Alaska Paper Birch against Snowshoe Hares. Oecologia, 65: 58-69.
Reichardt, P. B., Green, T. P. & Chang, S. M. 1987. 3-O-Malonylbetulafolientriol Oxide-I from
Betula-Nana Subsp Exilis. Phytochemistry, 26: 855-856.
Risenhoover, K. L. 1989. Composition and Quality of Moose Winter Diets in Interior Alaska.
Journal of Wildlife Management, 53: 568-577.
Risenhoover, K. L., Renecker, L. A. & Morgantini, L. E. 1985. Effects of Secondary
Metabolites from Balsam to Poplar and Paper Birch on Cellulose Digestion. Journal
of Range Management, 38: 370-371.
Robbins, C. T. 2001. Wildlife feeding and nutrition. , San Diego: Academic press.
Schneider, C. a. R., W. S. & Eliceiri, K. W. 2012. NIH Image to ImageJ: 25 years of image
analysis. Nature methods, 9: 671-675.
Sorensen, J. S. a. D., M. D. 2006. Efflux transporters as a novel herbivore countermechanism
to plant chemical defenses. Journal och Chemical ecology, 32: 1181-1196.
Speed, J. D. M., Austrheim, G., Hester, A. J. & Mysterud, A. 2010. Experimental evidence for
herbivore limitation of the treeline. Ecology, 91: 3414-3420.
17
Speed, J. D. M., Austrheim, G., Hester, A. J. & Mysterud, A. 2011. Growth limitation of
mountain birch caused by sheep browsing at the altitudinal treeline. Forest Ecology
and Management, 261: 1344-1352.
Stark, S., Vaisanen, M., Ylanne, H., Julkunen-Tiitto, R. & Martz, F. 2015. Decreased phenolic
defence in dwarf birch (Betula nana) after warming in subarctic tundra. Polar
Biology, 38: 1993-2005.
Sturm, M., Schimel, J., Michaelson, G., Welker, J. M., Oberbauer, S. F., Liston, G. E.,
Fahnestock, J. & Romanovsky, V. E. 2005. Winter biological processes could help
convert arctic tundra to shrubland. Bioscience, 55: 17-26.
Swann, A. L., Fung, I. Y., Levis, S., Bonan, G. B. & Doney, S. C. 2010. Changes in Arctic
vegetation amplify high-latitude warming through the greenhouse effect. Proceedings
of the National Academy of Sciences of the United States of America, 107: 1295-1300.
Taberlet, P. 1998. Biodiversity at the intraspecific level: the comparative phylogeographic
approach. Journal och Biotechnology, 64: 91-100.
Team, R. C. 2016. R: A language and environment for statistical computing. R Foundation for
Statistical Computing. Vienna. Austria. ISBN 3-900051-07-0. URL http://www.Rproject.org/.
Tremblay, B., Levesque, E. & Boudreau, S. 2012. Recent expansion of erect shrubs in the Low
Arctic: evidence from Eastern Nunavik. Environmental Research Letters, 7.
Walker, M. D., Wahren, C. H., Hollister, R. D., Henry, G. H. R., Ahlquist, L. E., Alatalo, J. M.,
Bret-Harte, M. S., Calef, M. P., Callaghan, T. V., Carroll, A. B., Epstein, H. E.,
Jonsdottir, I. S., Klein, J. A., Magnusson, B., Molau, U., Oberbauer, S. F., Rewa, S. P.,
Robinson, C. H., Shaver, G. R., Suding, K. N., Thompson, C. C., Tolvanen, A., Totland,
O., Turner, P. L., Tweedie, C. E., Webber, P. J. & Wookey, P. A. 2006. Plant
community responses to experimental warming across the tundra biome. Proceedings
of the National Academy of Sciences of the United States of America, 103: 1342-1346.
Weintraub, M. N. & Schimel, J. P. 2005. Nitrogen cycling and the spread of shrubs control
changes in the carbon balance of Arctic Tundra ecosystems (vol 55, pg 408, 2005).
Bioscience, 55: 551-551.
White, R. G. & Lawler, J. P. 2002. Can methane suppression during digestion of woody and
leafy browse compensate for energy costs of detoxification of plant secondary
compounds? A test with muskoxen fed willows and birch. Comparative Biochemistry
and Physiology a-Molecular and Integrative Physiology, 133: 849-859.
Williams, D. E., Sinclair, A. R. E. & Andersen, R. J. 1992. Triterpene Constituents of the
Dwarf Birch, Betula-Glandulosa. Phytochemistry, 31: 2321-2324.
Xu, L., Myneni, R. B., Chapin, F. S., Callaghan, T. V., Pinzon, J. E., Tucker, C. J., Zhu, Z., Bi,
J., Ciais, P., Tommervik, H., Euskirchen, E. S., Forbes, B. C., Piao, S. L., Anderson, B.
T., Ganguly, S., Nemani, R. R., Goetz, S. J., Beck, P. S. A., Bunn, A. G., Cao, C. &
Stroeve, J. C. 2013. Temperature and vegetation seasonality diminishment over
northern lands. Nature Climate Change, 3: 581-586.
Zhang, W. X., Miller, P. A., Smith, B., Wania, R., Koenigk, T. & Doscher, R. 2013. Tundra
shrubification and tree-line advance amplify arctic climate warming: results from an
individual-based dynamic vegetation model. Environmental Research Letters, 8.
18
S1: Table of analyzed compounds and results from Spearman’s rank correlations testing for
correlation between each compound and quantifications of condensed tannins performed with
traditional butanol acid assay. All significant positive correlations are marked out in bold text.
Compound
Chlorogenic acid
Catechin-(4-alpha-8)-epigalloCatechin
Catechin-(4-alpha-8)-epigalloCatechin2
Catechin-(4-alpha-8)-epigalloCatechin3
EpigalloCatechin-8-C-ascorbyl-3-O-gallate
EpigalloCatechin-derived
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)5
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)6
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)8
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)11
(+)Catechin-(4-alpha-8)-(+)Catechin
(Proanthocyanidin B1)12
Catechin_derived
Catechin_derived_2
EpiCatechin(4-beta-8)]3-epiCatechin
(-)-EpigalloCatechin-3-O-p-coumaroate
(-)EpiCatechin-(4-beta-6)-(-)epiCatechin(4-beta-8)-(-)epiCatechin
Prodelphinidin B4
Prodelphinidin B44
Prodelphinidin B47
Prodelphinidin B49
Prodelphinidin B410
Acacetin
Acactin19
Apigenin
Aromadendrin-glucoside
Aromadendrin-glucoside_2
Aromadendrin-glucoside13
Aromadendrin-glucoside14
Catechin-7-O-xyloside
Chrysoeriol
Chrysoeriol_derived
Chrysoeriol17
Isorhamnetin
Isorhamnetin16
Kaempferol
Kaempferol_like
Kaempferol-3-O-glucuronide
Ladanein
Ladanein18
Group
Chlorogenic acid
Complex tannin (P)
Complex tannin (P)
Complex tannin (P)
Complex tannin (P)
Complex tannin (P)
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
Condensed tannin
Condensed tannin
Condensed tannin
(P)
Condensed tannin
(P)
Condensed tannin
Condensed tannin
Condensed tannin
Condensed tannin
Condensed tannin
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
rho
0.157
0.365
0.492
-0.412
0.4198
-0.290
0.367
p
0.045
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.102
0.196
0.241
0.002
-0.186
0.018
0.014
0.857
0.075
0.341
-0.486
-0.345
0.034
0.224
<0.001
<0.001
0.666
0.004
0.367
<0.001
0.519
0.383
-0.068
-0.407
-0.156
-0.122
0.173
0.194
-0.051
-0.133
-0.179
0.289
0.034
-0.238
0.259
0.119
-0.156
0.025
-0.116
0.323
-0.285
-0.113
0.007
<0.001
<0.001
0.39
<0.001
0.047
0.121
0.027
0.013
0.517
0.09
0.022
<0.001
0.664
0.002
<0.001
0.13
0.047
0.749
0.142
<0.001
<0.001
0.151
0.934
Ladanein20
Myricetin
Myricetin-3-O-galactoside
Myricetin-3-O-rhamnoside OR Quercetinrham
Myricetin-3-O-rhamnoside OR Quercetinrham15
Naringenin
Pentahydroxyflavone trimethyl ether
Pentahydroxyflavone trimethyl ether21
Quercetin
Quercetin-3-arabinoside
Quercetin-3-arabinoside_?
(Galloyl-HHDP-glucose) Corilagin
(Galloyl-HHDP-glucose) Corilagin25
1.2.3.4.6-penta-O-galloylglucose
1.2.6-tri-O-galloylglucose
1-beta-O-GalloylPedunculagin
1-beta-O-GalloylPedunculagin27
Casuarin
Casuarin26
Casuarin_derived
Digalloylglucose 1
Digalloylglucose 2
Digalloylglucose 3
Digalloylglucose 123
ellagic acid-2-O-beta-D-glucopyranoside
Gallic acid_conjugated
Pedunculagin
Pedunculagin22
Pedunculagin24
Tellimagrandin I
Tetragalloylglucose 1
Deacetylpapyriferic acid_2
Deacetylpapyriferic acid_230
Deacetylpapyriferic acid_231
BFTO
BFTO (m/z 646.392)
BFTO_conjugated
BFTO_conjugated (m/z 1043.66)
BFTO_conjugated (m/z 653.364)
BFTO_conjugated_629
BFTO28
BFTO29
compound 4_1
compound5
compound5_1
compound5_1 (m/z 1171.69)
Compund6
Compund7
Compund8
demalonyl PA
demalonyl PA_2
Flavonoid
Flavonoid
Flavonoid
Flavonoid
0.308
0.401
0.056
-0.121
<0.001
<0.001
0.477
0.125
Flavonoid
0.445
<0.001
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Flavonoid
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Hydrolysable tannin
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
-0.123
-0.142
0.319
-0.07
0.088
-0.096
0.227
0.123
0.149
-0.155
0.211
0.386
0.390
0.197
0.049
-0.057
-0.052
-0.124
-0.017
-0.047
-0.267
0.002
0.094
0.068
0.28
-0.143
-0.360
-0.412
-0.331
-0.32
-0.379
-0.281
0.341
-0.281
-0.191
-0.324
-0.342
-0.228
0.089
0.095
-0.35
-0.247
0.171
0.192
-0.384
-0.273
0.119
0.070
<0.001
0.375
0.263
0.274
0.004
0.119
0.058
0.048
0.007
<0.001
<0.001
0.012
0.536
0.47
0.509
0.116
0.831
0.548
<0.001
0.984
0.23
0.387
<0.001
0.068
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.145
<0.001
<0.001
0.003
0.258
0.225
<0.001
0.001
0.029
0.014
<0.001
<0.001
demalonyl PA_conjugated (m/z741.405)
Dihydroxy demalonyl PA_1
Dihydroxy demalonyl PA_2
Dihydroxy demalonyl PA_3
Hydroxy demalonyl PA_1
Hydroxy demalonyl PA_2
Hydroxy demalonyl PA (m/z 735.306)
Hydroxy demalonyl PA (?)
methyl papyriferate
Papyriferic acid
Papyriferic acid (m/z 1075.74)
Papyriferic acid_2
Papyriferic acid32
papyriferic acid_conjugated (m/Z 561.378)
papyriferic acid_conjugated (m/z 757.345)
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
Triterpene
-0.358
-0.37
-0.353
-0.353
-0.355
-0.287
-0.346
-0.336
0.015
-0.324
-0.323
-0.346
-0.324
0.022
-0.366
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.061
<0.001
<0.001
<0.001
<0.001
0.783
<0.001
S2: Chlorogenic acid
S2. Boxplot representing chlorogenic acid content in samples from four dwarf birch taxa (B. glandulosa, B. nana
ssp. exilis, B. nana ssp. nana and B. pumila).
S3 Carbon and nitrogen
S3. Boxplots representing carbon (a) and nitrogen (b) content in samples from four dwarf birch taxa (B. glandulosa,
B. nana ssp. exilis, B. nana ssp. nana and B. pumila).
Dept. of Ecology and Environmental Science (EMG)
S-901 87 Umeå, Sweden
Telephone +46 90 786 50 00
Text telephone +46 90 786 59 00
www.umu.se