Seasonal variations in monoterpene emissions from Scots pine
(
In this study, we analysed the relationships between needle monoterpene synthase activities, endogenous monoterpene storage pools and monoterpene emissions of needles in two consecutive years at a boreal forest site in Finland.
The results showed changes in the monoterpene synthase activity of needles,
linked to seasonality and needle ontogenesis, while the pool of stored
monoterpenes (about 0.5 % of dry weight) did not change considerably as a
function of needle aging. Monoterpene emissions did not correlate directly
with enzyme activity or the storage pool size. We observed notably high
plant-to-plant variation in the biosynthesis rates of individual
monoterpenes, which did not reflect the storage compound mixture. The enzyme
activity producing
This study emphasizes the seasonal, developmental and intraspecific variability of monoterpene biosynthesis and storage, and calls for more in-depth analyses to reveal how such complex interaction affects monoterpene emissions from pine needles in boreal forests.
The evergreen foliage of conifers needs to acclimate to severe stresses under boreal winter conditions, including low minimum temperatures, low light availability and repeated freeze–thaw cycles. This acclimation is manifested in both structural and metabolic adjustments of needles. Seasonal dynamics in many plant processes creates strong variations in metabolic pools that enable the needles to remain viable and retain their functional capacity when conditions improve (e.g. Porcar-Castell et al., 2008; Ensminger et al., 2004, 2006). The spring dehardening of coniferous trees is closely linked to physiological changes related to the onset of growth, whereas hardening in the autumn results from the gradual downregulation of cellular metabolism, largely triggered by changes in temperature and the light environment (Hänninen and Tanino, 2011). In addition to primary metabolism related to growth and development, the secondary metabolism of needles, including the synthesis of volatile compounds, also shows a seasonal pattern that reflects their physiological state (e.g. Fischbach et al., 2002; Jaakola and Hohtola, 2010).
In coniferous plant species, volatile terpenes are produced in all tissues (needles, sapwood, bark, roots) and stored either in specialized terpene storage structures, the resin ducts (Loreto and Schnitzler, 2010), or in non-specific storage pools, for example cell membranes (Niinemets and Reichstein, 2002, 2003a, b; Ormeño et al., 2011). The regulation of terpene biosynthesis in needles is a complex process controlled by the availability of carbon substrates, as well as by the energy status of the cell (energy and redox equivalents) and key regulatory enzyme activities (Bohlmann et al., 1998; Fischbach et al., 2002; Dudareva et al., 2004; Ghirardo et al., 2014; Wright et al., 2014). Conversely, the turnover rates of storage pools depend on environmental constraints (primarily temperature) and on physiological or stress-related processes (e.g. filling of the resin duct storage; herbivory-induced plant defence responses). However, understanding of synthesis and storage pool dynamics is rather limited. Labelling experiments under laboratory conditions have revealed that monoterpene biosynthesis in needles is closely linked to the incident photosynthetic carbon supply in many conifer species. It has been shown that 30 %–60 % of emitted monoterpenes originate from recently fixed carbon, comprising the so-called “de novo” emissions, as opposed to the emissions from permanently stored pools (Loreto et al., 2000b; Ghirardo et al., 2010, 2011, 2014).
Large seasonal variations in volatile organic compound (VOC) emission rates
and also in the blend of emitted compounds have often been reported from
coniferous trees such as Scots pine (
As previously shown in the case of high emissions from emerging foliage
(Aalto et al., 2014), phenology is an important driver for seasonal
monoterpene dynamics (e.g. Wiß et al., 2017), not only in deciduous plant
species but also in evergreens, which retain their foliage for several years.
In evergreens, the development of new buds and foliage occurring in spring is
characterized by conspicuously high emissions of monoterpenes, methanol and
some other VOCs (Aalto et al., 2014). However, long-term studies clarifying
the seasonality of monoterpene production and storage in evergreen foliage
are scarce, and their correlations with needle emissions have not yet been
studied under field conditions. Fischbach et al. (2002) detected strong
changes in holm oak (
The present work was designed to comprehensively examine the linkages and dynamics of seasonal monoterpene emissions with the corresponding in vitro enzyme activities and sizes of storage pools in Scots pine needles in situ in a boreal forest. Because needle development (flushing, maturation and gradual ageing) was anticipated to affect the production, storage pools and emissions of monoterpenes, we followed the same branches over two consecutive years to examine possible relationships between the developmental state of needles, monoterpene emission rates, storage pool sizes and monoterpene synthase activities.
The samples were collected from the SMEAR II measurement station (Station for
Measuring Ecosystem–Atmosphere Relations) in southern Finland, in a
50-year-old Scots pine forest during 18 months between winter 2009 and
summer 2010. The site is located in a managed boreal forest
(61
The height of dominant trees is 18 m, with a breast height diameter of
20 cm. The typical annual tree stem growth rate is 8 m
Details of the trees examined and the sampling.
The temporal patterns of monoterpene emissions, synthesis and pools were
analysed with a repeated sampling design from four Scots pine trees (#1 to
#4) located a few metres from each other. Needle samples were collected
from sun-exposed, healthy upper canopy branches of the trees, which were
accessible from a scaffolding tower. On each sampling occasion, six healthy
needle pairs from three branches (whorls 3 to 5, about 1–2 m below treetop)
were collected and pooled as one sample per tree. Samples from two needle age
classes, formed during summer 2008 (hereafter referred to as “2008
needles”) and 2009 (“2009 needles”), were collected separately and
wrapped in aluminium foil. The samples were immediately frozen in a container
of liquid nitrogen and kept at
Daily mean air temperature (solid line) and daily precipitation (dashed line) at the sampling site during the sampling period from February 2009 to July 2010. Stars indicate the sampling dates (not all the trees were sampled every time). Horizontal arrows represent the snow cover periods.
Emissions from all four trees were tested before the campaign in order to
examine whether they differed from each other regarding their emission spectrum (see
Bäck et al., 2012). Monoterpene emissions were measured in August 2009
from all four trees. Due to limited resources, this was the only time period
when all trees could be sampled at the same time. Tree #3 was monitored
more intensively: its emissions were measured a total of 15 times between
February 2009 and June 2010. The emission monitoring was always carried out
from the same upper canopy branch with a transparent FEP (fluorinated
ethylene propylene) foil-covered dynamic flow-through shoot chamber (volume
approximately 6 L, see Fig. 2a; further details, e.g., in Hakola et al.,
2006). The whole shoot, i.e. both 2008 and 2009 needles as well as woody
twigs, was enclosed in the chamber. The dry weight (DW) of the enclosed needles
was about 3 g (final DW was determined later by drying at 80
Emitted monoterpenes were collected for 60 min in adsorbent tubes filled
with Tenax-TA/Carbopack-B (200 and 130 mg, respectively), with flow rates of
approximately 4 L min
Needle and shoot length growth was measured from adjacent trees at the same site, and the data were utilized to determine the needle development. For the purpose of analysis, needle age (days) was set to zero on the day when needles attained 50 % of their final length. This was 27 June 2008 and 29 June 2009 for the 2008 and 2009 needles, respectively. Figure 2b–e present the growth of the shoot and needles in May (when the growth of shoots had just started, but new needles had not yet emerged), June (the growth of new needles had initiated), July (needles already existed and were used for sampling) and August (fully mature shoot and needles).
Monoterpenes stored within the needles (endogenous monoterpene contents) and
in vitro enzyme activities of monoterpene synthases (MTS) were analysed
from the sampled needles of four pines. The analysis of monoterpene storage
pools followed the methods introduced in Fischbach et al. (2000, 2002) and
further developed in Ghirardo et al. (2010). One millilitre of pentane was added as a
solvent to 50 mg of frozen needles. The following changes were made to the
methods: after pentane extraction of ground frozen samples, 1
Analysis of in vitro MTS activities was carried out as per the
description in Ghirardo et al. (2012). Briefly, proteins were extracted and
successively incubated for 60 min with non-polar, polydimethylsiloxane
(PDMS)-coated stir bars (twisters, film thickness of 0.5 mm, Gerstel GmbH)
together with the enzyme substrate geranyl diphosphate (GPP). Enzymatically
produced monoterpenes were trapped from the aqueous reaction solution by the
twisters, and the removal of the stir bars terminated the assays. After
rinsing in deionized water, the twisters were analysed using TD-GC-MS. Each
sample was analysed with three technical replicates. Enzyme activities were
only assessed for the samples with protein contents higher than
0.1 mg mL
A total of 12 different monoterpenes and their derivatives could be detected
from the extracts:
The adsorbent tubes were analysed in the laboratory using a thermal
desorption instrument (Perkin-Elmer TurboMatrix 650, Waltham, USA) attached
to a GC (Perkin-Elmer Clarus 600, Waltham, USA) with a DB-5MS (60 m,
0.25 mm, 1
The emission rates (
To explain the variance in monoterpene dynamics, a large amount of auxiliary
data from the same measurement site was employed. Pine foliage net carbon
assimilation and transpiration is continuously monitored at the site with
automated shoot enclosures, as described, e.g., in Altimir et al. (2002) and
Aalto et al. (2015). For this study, running averages were calculated of
daytime (PAR > 50
Air temperature at a height of 4.2 m was measured with a ventilated and
shielded Pt-100 sensor and soil temperature with thermistors (Philips KTY
81-110, Philips Semiconductors, Eindhoven, the Netherlands) at five locations.
The daily air temperature range was calculated as the difference between
daily minimum and maximum temperatures. The temperature sum was calculated as
the annual cumulative temperature sum of daily mean air temperatures
exceeding 5
The variables used in PCA.
Principal component analysis (PCA) was employed to assess whether the
variations in MTS activity, storage and emission could be attributed to
changes in physical and physiological conditions, such as seasonal changes in
the weather (spring–summer–autumn), needle ontogenesis (aging of the
needles) and physiological parameters (needle and shoot growth, net
The synthase activities of
We examined the effect of seasonality and needle age on monoterpene
synthase (MTS) activity, monoterpene storage pools and emissions in detail, and found
that MTS activities were highly dependent on needle age and season
(Figs. 3, 4). Specifically, the youngest needles, i.e. the needles born in
2009 and measured in their first summer and autumn, showed high in vitro
MTS activities (Figs. 3a, 4a). In particular,
Relationship between monoterpene storage pools (
The pool size of the endogenous (stored) monoterpenes remained much more
constant throughout the measurement period and was independent of the needle
age (Fig. 3b). The mean monoterpene storage pool sizes were
The highest monoterpene emission rates (up to 6 mg kg
Overall, neither the synthesis rates nor the emission rates directly correlated with the size of the monoterpene pools across the whole measured period (Fig. 4). All four major monoterpenes showed high variability in their associations between storage pools and synthase activities or emissions. For a given storage pool size, the MTS activities varied by several orders of magnitude, with highest MTS activities present in < 1-year-old needles (Fig. 4a–d).
Two-dimensional
Therefore, principal component analysis (PCA) was employed to examine the complex link between meteorological and physiological parameters that played a major role in the dynamic changes in MTS activities, storage pools and monoterpene emissions. For both years, the variation in the data could be fairly well described by two main factors: seasonality (principal component 1, PC1) and needle aging (principal component 2, PC2), which together accounted for 60 % (2009) and 54 % (2010) of the total variation in the dataset (Fig. 5). This was indicated by the clear separation in PC1 of summer (Fig. 5a–b, depicted in red), spring (in white) and autumn (in grey) samples for both 2008 and 2009 needle samples, and by the separation made by PC2 of 2008 needle samples (triangles) from 2009 needle samples (circles) (Fig. 5a–b). Meteorological and physiological data, in particular air temperatures, carbon assimilation, transpiration and needle growth rates, correlated positively with the monoterpene emission rates and negatively with MTS activities (Fig. 5c–d). Within the same year, the MTS activities were significantly and positively correlated with younger needles, and high monoterpene emissions with summer samples (Fig. 5a–d). In 2009, the endogenous monoterpene content was lower in young needles compared to the content in mature needles. Such a difference was absent in 2010, when both needle age classes were already mature, i.e. two and three years old, respectively.
Taken together, the multivariate data analysis revealed a seasonal and ontogenesis-related dependency of emissions, storage and biosynthesis of monoterpenes. Changes in both MTS activity and the monoterpene pool size were related to needle ontogenetic phases. These changes notably occurred during needle development and needle maturation, and were also affected by seasonality.
Aiming at analysing the different tree chemotypes, we took needle samples
twice from all the four trees that had initially been screened for the
monoterpene blend in the emissions, and analysed the MTS activity, storage
pools and emissions (only once) from the individual trees from the youngest
needle year class (when the needles were 2 and 9 months old). The MTS
activity varied between the trees: it was lowest in needles from tree #2
in both summer and winter, and highest in needles from tree #4 in spring
(> 4 times higher than in tree #2) (Fig. 6a–b). The storage
pool size was relatively stable across all trees and sampling times
(Fig. 6c–d). In all four trees,
The trees differed strongly in their monoterpene emission patterns (Fig. 6e).
All trees emitted
Although the seasonal variations in monoterpene emissions from evergreen trees are well documented (e.g. Komenda and Koppmann, 2002; Tarvainen et al., 2005; Hakola et al., 2006), the reasons for this variation are poorly understood. The common explanation includes the temperature response of emissions, due to the strong role of temperature in physical parameters such as volatility and diffusion (e.g. Guenther, 1997; Niinemets and Reichstein, 2003a, b). This indeed creates seasonal dynamics, which partially but not fully explain the observed emission rates. The present results showed that the potentials to synthetize monoterpene, i.e. enzyme activities, are strongly dependent on needle age and season. However, these monoterpene synthase activities did not correlate with needle emission rates, and neither did the storage pool dynamics, which were virtually constant throughout the field experiment.
As indicated by the PCA analysis, the interaction between monoterpene synthesis, storage and emissions is complex and dependent on climatic factors (season) and needle age. Therefore, changes in emission rates at the leaf scale are probably a consequence of much more complex mechanisms than a simple incident temperature proxy (Aalto et al., 2014). This signifies the need to understand the physiological drivers, in addition to the physico-chemical drivers behind emissions.
The possible relationships between monoterpene production, storage and
emissions were investigated in this study using simultaneous measurements of enzyme
activities, storage pools and emissions. The results indicate that
synthesis, emission rates and storage are mainly decoupled. For instance,
high MTS activities did not correspond to large storage pools or high
emission rates, while high emissions were not a result of large storage
pools. The likely reasons for this are, on one hand, the large
monoterpene storage pools in needles, and on the other hand, the disparity
and time lags between production and emissions. Here, we see an analogy with
VOC synthesis in
Our results demonstrate that the monoterpene storage pool can make up ca.
0.2 %–0.8 % of needle dry weight. These concentrations are in a similar
range to previous observations from other conifers (Lerdau et al., 1997;
Litvak and Monson, 1998; Litvak et al., 2002; Kännaste et al., 2013).
Since emissions were 3 orders of magnitude less than pool sizes, they are
unlikely to affect the pool sizes in the short to medium term. Similarly, the
effect of MTS activities on pool sizes is also quite minimal in mature
needles and in the short term; thus, it did not show up in any
correlations. It is known that incident monoterpene emissions simultaneously
originate from previously filled and specialized storage pools and from
de novo synthesis. In Scots pine, de novo synthesis can
contribute
The decoupling between monoterpene production and storage pools and
emissions was also qualitatively evident: the individual MTS activities did
not reflect the monoterpene composition found in the storage pool or in the
shoot emission. For instance,
The overall variation in the dataset was best correlated with the season and
needle age, as indicated in the PCA analysis. Young needles generally
displayed a larger emission capacity than older ones, reflected in their
higher MTS activities. Our data indicate that needles retain a high capacity
for monoterpene synthesis throughout their first full year of growth, but the
MTS activity later sharply declines. The monoterpene concentrations and the
relative proportions in Scots pine needles are known to change during the
first months of new needles (Thoss et al., 2007), and even though our data do
not cover the very first months of needle development, our results support
this observation. In particular, the
The stability of the storage pool suggests that the monoterpene storages are filled during the very first weeks or months after leaf emergence (Bernard-Dagan, 1988; Schönwitz et al., 1990). This is logical when considering that one of the main reasons for storing monoterpenes in needles is their protection from herbivory (Langenheim, 1994; Litvak and Monson, 1998; Loreto et al., 2000a), and the youngest needles are particularly susceptible to many insects feeding on the fresh, sugar-rich tissues. This is also in line with the higher MTS activity in the youngest needles.
Our data suggest that the turnover of the permanent storage is a very slow
process, and agree with the low rate of incorporation of
The time lag between monoterpene synthesis and emissions can probably range from minutes (de novo) to days (permanent pools). The MTS activities could possibly correlate with de novo emissions in the short term, but this cannot be addressed with our data, which were collected at approximate bi-weekly intervals. A much finer temporal resolution and labelling experiments should be used to analyse the relationships further. In addition, it is evident that the in vitro synthase activities we measured here reflect the maximum potential of the needle tissue to synthesize monoterpenes under optimal conditions of temperature, pH and saturated substrate availability. Thus, they may not represent in situ synthesis processes, and this is a possible reason for the observed decoupling of production and storage/emissions.
Together with the synthase activities, the monoterpene precursor (substrate)
availability is involved in monoterpene production. Monoterpene biosynthesis
relies on photosynthesis for the supply of carbon substrates, reductive
(reduced form of nicotinamide adenosine dinucleotide phosphate, NADPH) and
energetic equivalents (adenosine triphosphate, ATP) (Lichtenthaler et al.,
1997; Phillips et al., 2008). The impact of substrate availability on enzyme
reaction can vary largely, depending on the pool size of the substrate and
the velocity of the enzyme to catalyze its reaction. Isoprene production is
known to be under the control of both precursors and enzyme activities
(Rasulov et al., 2010). Theoretically, monoterpene emissions are expected to
be much more “sensitive” to enzyme activities than substrate limitation due
to the much lower Michaelis–Menten constant (
As we aimed at a measurement set-up with repeated samplings, the challenge was to keep the extremely delicate monoterpene storage pools as intact as possible and to avoid any induced emission responses (Ghirardo et al., 2010; Niinemets et al., 2011). However, the monoterpene synthase activities varied between and within trees, and some days with exceptionally high MTS activities and emissions were observed. They may be due to certain short-term processes such as transient responses to herbivory or mechanical stress. It is known that handling a pine shoot causes increased monoterpene emissions for a few days (Ruuskanen et al., 2005). In addition, the observed anomalous emission blend of tree #3 in September 2009 could have originated from a stress reaction, and such unexplained high emission peaks have also been reported earlier in similar measurement set-ups (e.g. Tarvainen et al., 2005). Nevertheless, no visible damage was observed in the needles prior to sampling, and the twigs were not mechanically injured.
Furthermore, we collected the first set of youngest needles in mid-July, at a time when shoot lignification had not been fully completed and needle elongation was still in progress (ca. 85 % of final length, see Fig. 2). In practice, sampling short shoots before their lignification is already advanced causes large wounds and ample resin flow from the wounded twig.
One caveat in the emission measurement method is that the enclosures contained needles from two older year classes, but did not contain the youngest (2009) needles. Thus, we cannot separate the effect of needle development and maturation from this dataset: almost all emission measurements were performed from already mature needles, and only a fraction of the emissions originated from < 1-year-old needles in the spring and summer of 2009. A part of the emissions may also have originated from non-needle sources, i.e. the woody tissues of twigs.
This study was the first known research to address the relationship between monoterpene synthase activities, storage pools and emissions in situ in a resin-storing conifer, Scots pine. Our results emphasize the seasonal dynamics and developmental and intraspecific variability in monoterpene biosynthesis and storage, and call for more studies to reveal their connections with emission rates. As the monoterpene emissions depend on physiological, structural and environmental factors as well as on plant chemotypes in a complex manner, future studies should focus more attention on describing the seasonality as well as the tree-to-tree variation, by using a sufficient number of measurements and plant individuals. Also, the availability of the substrate GDP and its potential impact on monoterpene emissions should preferably be included in following studies. To improve the mathematical models for an accurate prediction of monoterpene emissions from boreal regions, a better understanding of the complex linkage between all the factors controlling monoterpene emission is urgently needed.
The weather data measured at the SMEAR II station are
available on the following website:
The authors declare that they have no conflict of interest.
We thank Juho Aalto and Janne Korhonen for assisting with sample collection and data analysis. The SMEAR II staff are acknowledged for their maintenance of the measurements and infrastructure.
This work was supported by the Academy of Finland Centre of Excellence programme [1118615 and 272041], the Helsinki University Centre for Environment HENVI [470149021], the Nordic Centre of Excellence CRAICC and the COST Action FP0903 “Climate Change and Forest Mitigation and Adaptation in a Polluted Environment”. Edited by: Akihiko Ito Reviewed by: two anonymous referees