EFFECTS OF WARMING, OZONE EXPOSURE AND NITROGEN ADDITION ON SOIL MICROBIAL PROFILES IN RHIZOSPHERE AREA OF SCOTS PINE (PINUS SYLVESTERIS) SEEDLINGS
Muhammad Usman Rasheed Master of Science Thesis University of Eastern Finland Department of Environmental Science October, 2015
UNIVERSITY OF EASTERN FINLAND, Faculty of Science and Forestry
Department of Environmental Science
Muhammad Usman Rasheed: Effects of warming, ozone exposure and nitrogen addition on soil microbial profiles in rhizosphere area of Scots pine (Pinus sylvestris) seedlings
MSc thesis 54 pages
Supervisors: Anne Kasurinen (PhD), Promise A. Mpamah (MSc) and Toini Holopainen (Prof.) October, 2015
Keywords: warming, ozone, nitrogen deposition, rhizosphere, microflora, climate change, Scots pine
ABSTRACT
Increasing atmospheric temperature, tropospheric ozone (O3) concentration and nitrogen (N) deposition in the boreal forests are effecting the forest ecology. In this work, the effect of the above three factors, alone and in combination, on the below-ground microbial profiles in the Scots pine (Pinus sylvestris) rhizosphere have been studied. The experiment was set up in a field where the seedlings were placed in eight exposure plots. Half of the plots were exposed to elevated ozone concentrations (1.5x ambient O3) using the open air fumigation systems, and the rest of the plots were ambient air plots. These plots were further divided into subplots for warming treatments: in each plot one subplot was kept at elevated temperature using infrared heaters (air temperature elevation c.
+1°C, soil temperature elevation c. 0.8°C) and other subplots were kept at ambient temperature. Half of the trees in each subplot received nitrogen fertilization (N dose 120 kg/ha/a in 2011 and 2013), while rest were grown at prevailing soil N level. The microbial profiles were studied using phospholipid derived fatty acids (PLFAs) from soil samples collected in autumn 2013. The main microbial groups in the rhizosphere soil were gram-positive, gram-negative bacteria, Fungi(AMF-) (all fungi except AMF) and Fungi(AMF+) (only AMF). The results revealed that warming increased while ozone exposure decreased fungi(AMF-) relative abundance. Relative abundance of gram- negative bacteria decreased while that of fungi(AMF-) increased in response to increased nitrogen availability. Ozone exposure and warming showed significant interactive effect on fungi(AMF-) whereby they warming partly compensated for the effects of ozone. Results obtained in this experiment show that the below-ground microbial profiles are sensitive to the futuristic climatic conditions especially fungi; which it may effect the soil nutrient cycle.
ACKNOWLEDGEMENTS
This thesis work was conducted in Department of Environmental Science, Faculty of Science and Forestry, University of Eastern Finland, Kuopio campus. It was part of the CABI project 931050 and was funded by strategic funding by University of Eastern Finland and by the Academy of Finland (project nos. 272939 and 266532). The experimental exposures were conducted in the experimental field at Ruohoniemi near the University of Eastern Finland, Kuopio campus.
I would like to use this opportunity to express my gratitude to my supervisors Academy Research Fellow Dr Anne Kasurinen and Professor Toini Holopainen and for giving me the opportunity to work in their project, guiding me through each step of the thesis, their availability for discussion, advice and comments on my work and most importantly, being an inspiration for hard work and scientific approach. Dr Kasurinen has particularly helped me a lot in improving my writing skills. I would also like to thank my supervisor MSc Promise Mpamah for helping me through the PLFA extraction and analysis. I wish to thank Juhani Tarhanen for technical assistance in running the GC- MS. I also wish to thank the laboratory assistants and other staff members at the department of Environmental Science for helping me through the process of laboratory work. I would also like to extend gratitude to the library staff that was always there to provide technical support for data acquisition.
Finally, it is my pleasure to thank my family, especially my parents who, though are miles away, yet have been my greatest comfort, love and support. I owe whatever good I have achieved so far to them and at least, do humble acknowledgement to their efforts.
Kuopio, 2015
Muhammad Usman Rasheed
ABBREVIATIONS
AMF Arbuscular mycorrhizal fungi
AOT40 Accumulated dose over a threshold of 40 ppb
C Carbon
DW Dry weight EM Ectomycorrhiza GC Gas chromatography MS Mass spectrometry
N Nitrogen
O
3Ozone
PLFA Phospholipid fatty acids ppm Parts per million
ppb Parts per billion
CONTENTS
1. INTRODUCTION ... 6
2. BOREAL FORESTS AND CLIMATE CHANGE ... 8
2.1. Boreal forests ... 8
2.2. Microbial community structure in forest soils ... 8
2.3. Climate change factors ... 10
2.3.1. Warming climate ... 10
2.3.2. Warming and forest responses ... 11
2.3.3. Warming and soil microbes ... 12
2.3.4. Tropospheric ozone concentration changes ... 13
2.3.5. Ozone stress and forests ... 14
2.3.6 Ozone stress and soil microbes ... 14
2.3.7. Nitrogen deposition increase ... 16
2.3.8. Nitrogen deposition in boreal forests ... 17
2.3.9. Nitrogen deposition and soil microbes ... 17
2.3.10. Interactive effects of warming, ozone exposure and nitrogen deposition on boreal forests and soil microbes ... 18
2.4. Methods for characterizing soil microbial communities ... 18
3. AIMS OF THE STUDY ... 21
4. MATERIALS AND METHODS ... 22
4.1. The experimental design ... 22
4.2. Warming, ozone exposure and nitrogen addition in the plots ... 24
4.3. Sampling for PLFA analysis ... 26
4.3.1. PLFA extraction ... 26
4.4. Statistical analysis ... 27
5. RESULTS ... 29
5.1. Treatment effects on relative proportions of PLFAs and fungi:bacteria ratio ... 29
5.3. Treatment effects on PLFA concentrations in the soil ... 33
6. DISCUSSION ... 38
6.1. Fungi:bacteria-ratio changed due to increased N and warming ... 38
6.2. Nitrogen addition effects on fungi(AMF-) and gram-negative bacteria. ... 38
6.3. Warming effects on microbes can be modified by ozone ... 39
7. CONCLUSIONS ... 41
8. REFERENCES... 42
6 1. INTRODUCTION
Anthropogenic activities are leading to an increase in the global pollution. The most evident change, observable from the climatic trends, is global warming. The increased emission of nitrogenous oxides and hydrocarbons due to anthropogenic activities has also led to an increase in ozone (O3) concentrations in the troposphere (Brasseur et al., 1998). Ozone is not directly produced by anthropogenic activities but it is a product of the photochemical reaction between air pollutants (Fowler et al., 1999). The background ozone concentrations in the early twentieth century were around 20 ppb (Vingarzan, 2004), towards the mid twentieth century the concentrations reached almost 35 ppb in most effected regions (Cionni et al., 2011). The decadal mean between year 2000- 2009 shows regions in the northern hemisphere going over 50 ppb (Cionni et al., 2011). The rate of increase of tropospheric ozone concentration in the northern hemisphere has decrease since 1990s (Vingarzan, 2004) and based on the current emission scenario the near term (year 2050) tropospheric ozone concentrations are expected to be the same as, or slightly less, compared to the year 2000 values (IPCC, 2013, Kim et al., 2015).
Anthropogenic activities pose also another threat to the forest ecosystems: eutrophication. Nitrogen deposition in the soils have increased as a result of various anthropogenic activities (Galloway et al., 2008, Gundale et al., 2014). The deposition rates remain relatively low in the boreal forests compared to the industrialized regions, but the effects of nitrogen deposition are already visible in the forest vegetation (Dentener et al., 2006, Gundale et al., 2011). Recent findings showed that long lasting changes occur at low nitrogen (N) doses in the forest vegetation and it has been suggested that the critical load of nitrogen dose should be reduced to 6 Kg N ha-1 y-1 from 10-15 Kg N ha-1 y-1 (Nordin et al., 2005).
Soil microorganisms, comprising of bacteria, archaea and fungi, are largely responsible for the nutrient recycling processes between plants and soil. Forest soil microbial profiles can be sensitive to changes in the environment (Bossio and Scow, 1995, Waldrop and Firestone, 2006). It is expected that the increased temperature will have a stimulating effect on the plant biomass production as the increased temperature will result in enhanced availability of decomposed matter and consequently carbon dioxide (Briceno-Elizondo et al., 2006). Tropospheric ozone, on the other hand, is generally a stress factor for forests. It effects the primary plant production primarily by causing a loss of ribulose 1,5-bisphosphate carboxylase/oxygenase enzyme hence reducing the photosynthesis ability of plants (Wittig et al., 2007). Nitrogen addition acts as a fertilizer for plants, but deposition of nitrogen beyond the critical load (10-15 kg N ha-1 y-1 for Finland) will have rapid and long lasting effects on the forest ecology whereby the nitrophilic species will take over the habitat (Dirnböck et al., 2014). Though
7
single effects of all above factors have been studied to some extent, little is known about the interactive effects of these three factors combined.
The main aim of this study was to experimentally investigate how the soil microbial profiles beneath young Scots pine seedlings are effected by warming, increasing tropospheric O3 and N deposition alone and in combination. Samples for soil microbial analysis were collected after the third exposure year in 2013.
8 2. BOREAL FORESTS AND CLIMATE CHANGE 2.1. Boreal forests
Boreal forests form the largest terrestrial biome in the world (Dixon et al., 1994). They stretch between 45° to 70° North (METLA, 2013) comprising almost 1,200 million hectares over Alaska, Canada, Northern Europe and Russia, and represent a significant carbon pool (Olson et al., 1997, Conard et al., 2002, Janssens et al., 2003). Boreal forests are characterized by low temperatures, short growth periods, slow tree growth as well as low decomposition rates and thus high soil carbon content (Amundson, 2001, Bardgett, 2005). Usually, soil in boreal forests have a thick humus layer with an understory of mosses, dwarf shrubs and flowering plants (Fisher and Binkley, 2000). Boreal forests receive an average precipitation of 900 mm y-1 with low rates of evapotranspiration (Fisher and Binkley, 2000).
Almost 73% (23 million hectares) of Finnish land area is covered with boreal forests, Scots pine (Pinus sylvestris) being the most common (63.9%) tree species (METLA, 2013). Scots pine is found across almost the entire stretch of Finland with its northern growth limits close to 70° North (METLA, 2013).
2.2. Microbial community structure in forest soils
The soil microbial community comprises of a vast array of organisms that form a complex food web whose primary function is to recycle the litter entering the soil from above- and below-ground primary production of plants (Bardgett, 2005). The primary consumers in soil are bacteria and fungi, known as microflora (body width 0.3-20µm). DNA-based species identification of soil samples has revealed that at least ~104 different bacterial species belonging to 6300 taxa can be found in a cubic centimeter of boreal soil (Torsvik et al., 1990, Curtis et al., 2002, Torsvik et al., 2002).
Despite the large number of bacterial species, fungi dominate the soil microbial biomass in the undisturbed boreal forests (Bardgett, 2005, Högberg et al., 2007, Joergensen and Wichem, 2008). The mean biomass of soil bacteria in Scots pine forests has been reported to be 39 g DW m-2 while that of fungi to be around 120 g DW m-2 in a 14-month long sampling experiment from a 120-year old Scots pine stand in Sweden (Persson et al., 1980). Almost 95% of decomposition of organic matter in Scots pine forest soils is carried out primarily by bacteria and fungi (Persson et al., 1980). Both bacteria and fungi digest xenobiotic compounds (foreign chemical compounds e.g. complex carbohydrates) by secreting extracellular enzymes into the soil (Lavelle and Spain, 2001). Fungi are mostly saprophytic, some can be pathogenic or symbiotic and some are both saprophytic and symbiotic. Pathogenic fungi infect the roots and cause damage or even death to the plants (Lavelle
9
and Spain, 2001, Bardgett, 2005), whereas symbiotic fungi, such as mycorrhiza, infect roots and provide the plants with macronutrients like nitrogen and phosphorus in return for carbon (Bardgett, 2005). Saprophytic (litter decomposing, non-mycorrhizal) fungi are mostly found in the top soil while the lower layers of the soil are dominated by symbiotic mycorrhizal fungi (Lindahl et al., 2007).
Mycorrhizal infections are ubiquitous with almost 95% of vascular plants being associated with mycorrhizal fungi (Bardgett, 2005, Brundrett, 2009, Smith and Read, 2008). Arbuscular mycorrhizal fungi (AMF), the most common mycorrhiza type globally, are characterized by the inter/intra-cellular mycelia, arbuscules and spore producing extramatrical mycelium. In boreal forests the main mycorrhizal type is ectomycorrhiza (EM). Ectomycorrhiza are characterized by the formation of mantle on the host plant’s short roots, growth of hyphae between the root cortical cells (Hartig net) and growth of extramatrical mycelia (Figure 1a) (Peterson et al., 2004, Smith and Read, 2008). The extramatrical mycelium is site of absorption and translocation of nutrients and water from the soil beyond the reach of roots (Figure 1b), the Hartig net is the site for nutrient and water exchange in the roots, while the mantle provides physical protection and is also involved in storage of sugars and lipids (Peterson et al., 2004). Majority of EM fungi belong to basidiomycetes (Smith and Read, 2008), and the genera Pinus is among the ectomycorrhizal plants. (Smith and Read, 2008).
10
Figure 1. a) Diagram of ectomycorrhizal fungi along the longitudinal section of conifers (left half) and angiosperm (right half), ‘m’ represents the mantle, arrowheads point to the Hartig net and arrows point to the extra metrical mycelia in both the halves. b) Extramatrical mycelia of Suillus bovinus associated with Pinus sylvesteris. Images from (Peterson et al., 2004).
2.3. Climate change factors 2.3.1. Warming climate
Anthropogenic activities have caused unprecedented changes to the climate system. According to the IPCC fifth assessment report (IPCC, 2013) atmosphere and oceans have warmed, snow caps and glaciers have melted and the sea levels have risen, especially since the 1950s. The past three decades have been the warmest over the last 1400 years. Carbon dioxide, methane and nitrous oxide concentrations, as recorded in year 2011, are 391 ppm, 1803 ppb and 324 ppb, respectively, increase over the past century being unprecedented over the last 22000 years (IPCC, 2013). All this change is largely attributed to the anthropogenic activities, like burning of fossil fuels and change in land use.
The above mentioned greenhouse gases have a combined radiative forcing (defined as the variation in the net radiative flux at the tropopause) of 3.00 W m-2 leading to a global warming effect. The near
11
term (2016-2035) projected increase in the atmospheric temperatures is expected to be in the range 0.3-0.7ºC relative to the temperatures prevailing in the period 1986-2005 (IPCC, 2013). Warming is expected to be stronger in northern latitudes than in south and the short term projected (2016-2035) increase in atmospheric temperature in northern Europe is 1-2ºC relative to the temperatures prevailing in 1986-2005 (IPCC, 2013). For Finland the increase in the mean annual temperatures is expected to be even higher; where the current mean annual temperature for southern Finland is 4- 5°C, by the year 2080, according to three separate models, the mean annual temperatures can vary between 6-8°C with the coldest months having a mean temperature of -3°C or higher; and the temperatures are expected to increase even further by the end of 21st century (Jylhä et al., 2010, Jylhä et al., 2014).
2.3.2. Warming and forest responses
Net primary production in terrestrial ecosystems (including boreal Scots pine forests) is expected to increase with increasing temperature in the absence of water limitation (Briceno-Elizondo et al., 2006, Raich et al., 2006, Kellomäki et al., 2008, Chapin et al., 2009, Wu et al., 2011, Kasurinen et al., 2012).
Warming may also increase the decomposition rate of soil organic matter and it is expected that the carbon efflux due to soil respiration is greater than the carbon fixed by plants leading to a net reduction in soil carbon pools and give a positive feedback to warming (Raich and Schlesinger, 1992, Kirschbaum, 1995, Amundson, 2001, Rustad et al., 2001, Woodward et al., 2004, Raich et al., 2006).
Increased soil temperatures have also been reported to enhance nitrogen availability in soils, most likely as a consequence of increased microbial activity and nitrogen mineralization (Lukewelle and Wright, 1997, Rustad et al., 2001). Reduction in root growth in black spruce (Picea mariana) have also been shown as a consequence of soil warming (Bergner et al., 2004, Bronson et al., 2008).
Bergner et al. (2004) used open top chambers around the trees to realize warming effect while Bronson et al. (2008) used heating cables for soil warming. On the other hand, Kasurinen et al. (2012) have reported increase in root, stem and shoot biomass in silver birch (Betula pendula) in response to soil + air warming experiment using infra-red heaters above the tree canopy. Leppälammi- Kujansuu et al. (2013), have also reported increase in short roots in Norway spruce (Picea abies) and Scots pine after warming treatment using heating cables. A possible explanation to reduced root growth may be the indirect increase in nitrogen availability due to warming (Nadelhoffer, 2000, Hyvönen et al., 2006, Bronson et al., 2008), or overall increased respiration rates of the plants (Bergner et al., 2004, Vega-Frutis et al., 2014). Warming is often associated with increased evapotranspiration and drying of the soils with low water holding capacity (Kirschbaum, 1995, Davidson et al., 1998, Bronson et al., 2008). Low moisture content of soil has been reported to limit
12
tree growth in a study on Norway spruce from Sweden (Bergh et al., 2005). Increased evapotranspiration due to warming is expected to limit tree growth for Scots pine also, suggested by a modelling study, where this growth-limiting effect was more pronounced in Southern Finland than in Northern Finland (Kellomäki, 1995). Increasing temperatures may even lead to migration of temperate forest species into the boreal zone leading to expansion of mixed temperate-boreal forests and reduction in boreal forests (Fisichelli et al., 2014). Tang and Beckage (2010) have also modelled forest species migration in response to in response to warming.
2.3.3. Warming and soil microbes
Warming also induces changes in below-ground microbes. Fungi are reported to be more tolerant to low temperatures while the opposite is true for bacteria with the minimum temperatures for growth being -17.5°C and -12.1°C, respectively, in humus rich boreal soils from southern Sweden (Pietikäinen et al., 2005). Barcenas-Monero et al. (2009) have reported a shift in bacterial optimal growth temperatures as a result of warming (temperature increase was 10-35°C) in a laboratory experiment. They suggested that the observed change in optimal growth temperatures was due to a shift in bacterial diversity towards thermophilic species. Total fungal biomass measured as ergosterol (including EM, but not AM e.g. Olsson et al., 2003) has been reported to increase in response to temperature by Clemensen et al. (2006) after 14-years of warming (air temperature increase 2.8°C and soil temperature increase 0.4-0.6°C). Leppälammi-Kujansuu et al. (2013) have also shown a stimulating effect on ectomycorrhizal fungi after warming (air temperature increase 1-5°C) experiment. Ergosterol concentration in soil under silver birch have also been reported to increase in response to soil + air warming (Kasurinen et al., 2012). AMF growth is also reported to increase as a result of warming in a study based on the length of extraradical hyphae measurements (Heinemeyer and Fitter, 2004, Bunn et al., 2009, Vega-Frutis et al., 2014).
Allison and Treseder (2008) have reported a reduction in total microbial (bacterial and fungal) DNA content of soils by 50% after warming experiment (soil temperature increase 0.5°C) in boreal forests.
Berg et al. (1998) have also shown a decreasing effect on fungal hyphal length due to increasing soil temperature however their results show that there is no significant effect of warming on soil bacteria.
The reduction in microbial DNA in soil may be attributed to drying caused by warming (Berg et al., 1998, Davidson et al., 1998, Verburg et al., 1999, Gulledge and Schimel, 2000). Allison and Treseder (2008) also report increased fungal diversity in response to passive warming using greenhouses around the subject soil.
13 2.3.4. Tropospheric ozone concentration changes
The emission of carbon monoxide, nitrogenous oxides and hydrocarbons in the atmosphere have led to increase in tropospheric ozone concentrations from the preindustrial levels (Crutzen and Zimmermann, 1991, Fowler et al., 1999, Ehhalt, 2001, Vingarzan, 2004, Lamarque et al., 2011).
Ozone in troposphere is a cause of concern as it has a strong radiative force (0.35 W m-2) (IPCC, 2013), and it also adversely effects the vegetation (Wittig et al., 2007, Wittig et al., 2009, Matyssek et al., 2010a). Ozone is a secondary pollutant which means it is not directly released into the atmosphere as a result of anthropogenic activities, but it is instead a product of photochemical reaction between carbon monoxide, nitrogenous oxides and hydrocarbons (Crutzen and Zimmermann, 1991, Fowler et al., 1999, Ehhalt, 2001). The tropospheric ozone concentration shows spatial and temporal variation. The seasonal maxima occur in spring and early summer and the daily ozone concentrations peak close to midday in Northern Hemisphere (Vingarzan, 2004, Oltmans et al., 2006). Ozone concentration follows the spatial distribution of its chemical precursors in the atmosphere, and also increases with altitude reaching a maximum near the tropopause (Ehhalt, 2001, Cionni et al., 2011).
The early twentieth century records show that the tropospheric ozone concentration ranged around 20 ppb in central and southern Europe (Vingarzan, 2004). By the mid twentieth century the background O3 concentration reached 35 ppb and towards the end of twentieth century (year 1990), the background tropospheric ozone concentrations went beyond 40 ppb over large parts of Asia, Central Europe and North America with the alarming >70 ppb in some regions of North America, southern and central Europe, and South Asia; (Emberson and Ashmore, 2001, Akimoto, 2003, Cionni et al., 2011). The rate of increase of tropospheric ozone concentration decreased in 1990s (Vingarzan, 2004, IPCC, 2013) and the projections for the mid twenty first century based on the projected concentrations of greenhouse gases according to Representative Concentration Pathway (RCP) scenarios (RCP2.6, RCP4.5, RCP6.0) show that the tropospheric ozone concentrations are expected to fall globally below the year 2000 level, the change largely attributed to decrease in NOx emissions (Kim et al., 2015). Only according to the RCP8.5 scenario (based on projected methane concentration) ozone levels are expected to increase up to 7 ppb in Africa, South Asia, South America and Australia (Kim et al., 2015).
Studies from Finland suggest that the mean background ozone concentrations were less than 16 ppb in early 1900s, it increased to 25 ppb by mid twentieth century and it currently stands between 30-40 ppb during spring-early summer maximum (Laurila, 1999, Laurila et al., 2004, Ruoho-Airola et al., 2015).
14 2.3.5. Ozone stress and forests
In the changing climate where increasing temperature is expected to increase the plant primary production (Raich et al., 2006), ozone can significantly reduce the plant carbon dioxide uptake and primary production, and as a result reduce the potential of forests as carbon sinks (Karnosky et al., 2003, Sitch et al., 2007, Riikonen et al., 2009, Matyssek et al., 2010b, de Vries et al., 2014). Ozone’s phytotoxic effects include reduced stomatal conductance, and photosynthesis, advanced leaf senescence, changed carbon allocation leading to reduced root and shoot growth (Andersen, 2003, Grantz et al., 2006, Sitch et al., 2007, Wittig et al., 2007, Wittig et al., 2009). It has also been reported that ozone exposure causes thinning of the stems in mature trees leading to slender, heavy top tree structure (Pretzsch et al., 2010). The effects of ozone on plants are usually concentration dependent and the response towards ozone exposure also varies among different tree species (Wittig et al., 2007, Wittig et al., 2009). The effects may manifest in short-period or take a long time (memory effect) to manifest themselves (Bortier et al., 2000, Utriainen and Holopainen, 2001b, Ferretti et al., 2003, Vahala et al., 2003, Huttunen and Manninen, 2013). At high concentrations ozone can simulate pathogen attack due to formation of reactive oxygen species (Vahala et al., 2003). Ozone also stimulates immediate stomatal closure in Scots pine needles (Paoletti and Grulke, 2005) and may even lead to the loss of 2- to 3-year old needles in Scots pine trees (Augustaitis et al., 2007, Huttunen and Manninen, 2013).
Scots pine is considered to be more ozone sensitive compared to Norway spruce but it is less sensitive compared to deciduous tree species, like birch and aspen (Huttunen and Manninen, 2013). Manninen et al. (1998) report a stimulating effect of ozone on short roots after two years of 1.5 x ambient ozone exposure of 2-year-old nursery grown seedlings. Increase in the short root growth was also reported by Kasurinen et al. (1999) from an experiment on natural Scots pine stands in Eastern Finland.
Increase in short roots was observed after first year of 2 x ambient ozone exposure, however, stimulating effect of ozone disappeared after second year of exposure, and the overall effect of ozone on Scots pine short roots was shown to be statistically insignificant. Kainulainen et al. (2000) have also reported no significant effects on short roots of 3-year-old Scots pine after ozone exposure for three growing seasons. No significant effect on growth or biomass of 1-year-old hybrid aspen (Populus tremula L. x Populus tremuloides Michx) clones in response to three year ozone exposure (1.3-1.4x ambient ozone concentration) has been reported (Häikiö et al., 2009).
2.3.6 Ozone stress and soil microbes
Ozone is generally toxic to bacteria, and ozonation is an effective method for disinfection of drinking water (Lehtola et al., 2001) but tropospheric ozone is reported to have no direct effects on soil bacteria
15
or fungi (Kasurinen et al., 1999, Kainulainen et al., 2000) because it is degraded in the topsoil (Diaz et al., 1996, Aneja et al., 2007). There may also be indirect effects of ozone on underground microbes mediated through nutrient content of litter returning to soil (Aneja et al., 2007). Aneja et al. (2007) have shown an increase in starch content of the fallen leaves, however their results show no significant effects on the soil microbes. Kasurinen et al. (2006) have also reported decrease in boron, manganese, zinc, nitrogen and phosphorus content of leaf litter after ozone exposure of up to 60 ppb of silver birch trees from a three year exposure experiment.
On the other hand, Phillips et al. (2002) have shown that there is no significant effect of ozone on soil bacteria but they report a reduction in soil fungi after PLFA analysis of the soil samples from the rhizosphere of birch, aspen and maple trees (~2m high). Pritsch et al. (2009) have reported a decrease in bacterial signature PLFA concentrations in response to 2x ambient air ozone exposure to European beech (Fagus sylvatica L.), while their report shows no significant effect on fungal PLFA markers or ectomycorrhizal diversity upon visual examination of fine roots under stereomicroscope. Kanerva et al. (2008) have also reported decrease in overall soil microbial biomass in meadow soil, based on PLFA analysis, in response to 1.3 x ambient air ozone exposure.
Mycorrhiza are expected to be adversely effected by ozone exposure primarily due to decreased carbon allocation to roots (Manning, 1995, Andersen, 2003). Kainulainen et al. (2000), however, did not report significant decrease of Scots pine mycorrhizas after three-year ozone exposure. Kasurinen et al. (1999) showed a transient stimulation of mycorrhizal fungi under Scots pine trees in their three- year lasting ozone exposure experiment. Such transient stimulating effects of ozone has also been reported by Manninen et al. (1998) and Rantanen et al. (1994). Häikiö et al. (2009) reported a significant increase in ergosterol concentration, indicating an increase in mycorrhizal infection in response to ozone exposure in the rhizosphere of hybrid aspen (Populus tremula L. x Populus tremuloides Michx) clones. Exposure to higher than 2x ambient ozone concentration (55 ppb) have been shown to have strong decreasing effect on mycorrhizas under Scots pine trees (Perez-Soba et al., 1995). In another experiment 2x ambient air ozone exposure has been reported to alter mycorrhizal diversity under silver birch trees, shifting it towards more carbon demanding mycorrhizal assemblages (Kasurinen et al., 2005). Change in the biodiversity of ectomycorrhizal fungi in response to ozone exposure has also been reported by Edwards and Zak (2011) from a decade long exposure experiment on trembling aspen (Populus tremuloides Michx), paper birch (Betula papyrifera Marsh.) and sugar maple (Acer saccarum Marsh.).
16 2.3.7. Nitrogen deposition increase
In addition to greenhouse gases and tropospheric ozone concentration, nitrogen deposition rates have also increased globally, and especially over the Central Europe, North America and East Asia, mainly due to increased emissions of nitrogenous oxides from fossil fuel burning, fertilizer production and aggressive agricultural practices to meet the food demands of the increasing population (Aber et al., 1989, Dentener et al., 2006, Galloway et al., 2008, Reay et al., 2008). Nitrogen addition from air to soil generally stimulates the growth of vegetation (Gundale et al., 2014), although it may also lead to acidification of soil leading to poor nutrient uptake by the plants (Emmett, 1999, de Vries et al., 2009, Lukas et al., 2011, Lu et al., 2014). Forest fertilization for commercial purposes has also lead to changes in forest soil structure (Saarisalmi and Mälkönen, 2001). Forest fertilization may effect soil pH depending on the form of nitrogen being supplied. Urea may lead to an increase in soil pH as it supplies the basic NH3 that reacts with H+ ions while NO3- ion that may lead to the leaching of the cations leading to acidity (Aarnio and Martikainen, 1992, Aarnio and Martikainen, 1994, Aarnio et al., 1995).
The nitrogen deposition rates from year 2000 show a maximum deposition of 60 kg N ha-1 y-1 in Central Europe, North America and East Asia (Dentener et al., 2006, Galloway et al., 2008, Reay et al., 2008), whereas the nitrogen deposition rates have reached over 30 kg N ha-1 y-1 in South Asia, parts of Africa and South America (Dentener et al., 2006). The future projections (year 2030) using the current emissions scenario suggest that the nitrogen deposition rates are to remain constant or decrease over Central Europe and North America; and increase in East Asia (70 kg N ha-1 y-1) and in South Asia (40 kg N ha-1 y-1) (Dentener et al., 2006, Reay et al., 2008). The reactive nitrogenous species are produced largely in industrialized regions. However, the international trade of nitrogen containing products (including fertilizers and food products) plays a more significant role in its distribution across the globe compared to the aquatic and atmospheric means of nitrogen distribution (Galloway et al., 2008).
Nitrogen deposition rates in boreal forests are generally lower than those of the Central Europe (0-12 Kg N ha-1 y-1) (Dentener et al., 2006, Gundale et al., 2011). Reports from Finland vary between ≤ 1 to 7 Kg N ha-1 y-1 (Mustajärvi et al., 2008, Korhonen et al., 2013, Ruoho-Airola et al., 2015).
Mustajärvi et al. (2008) have reported an average of 2.8 kg N ha-1 y-1 across sixteen sites in Finland and also observed a decreasing trend in the rate of deposition particularly in Southern Finland over the period of 1998-2007. Korhonen et al. (2013) on the other hand, have recently reported a much higher value of 7 kg N ha-1 y-1 from Scots pine forest in Hyytiälä, Southern Finland. In addition, Ruoho-Airola et al. (2015) reported the five-year mean bulk N deposition from the Finnish subarctic
17
stations of Pallas, Sodankylä, Kevo and Oulanka to be below 1 Kg N ha-1 y-1. In Finland forest fertilization has also been practiced for increased wood production in selected forests where nitrogen (N) doses of up to 200 kg N ha-1 y-1 have been applied to the forest floors (Saarisalmi and Mälkönen, 2001).
2.3.8. Nitrogen deposition in boreal forests
Nitrogen deposition has a net fertilization effect that can lead to increased carbon sequestration (Emmett, 1999, de Vries et al., 2014, Gundale et al., 2014). It has been experimentally shown that the increase in carbon sequestration in forest soils resulting from the fertilization effect of nitrogen deposition may significantly absorb the carbon dioxide released into the atmosphere by anthropogenic activities (Högberg et al., 2006, Pregitzer et al., 2008). The biosphere on the other hand, is estimated to absorb only a minor amount of anthropogenic carbon dioxide emissions (Janssens et al., 2003, Gundale et al., 2014). It is also typical that the fine root biomass decreases in response to nitrogen deposition (Nadelhoffer, 2000). The understory vegetation in general, is more sensitive towards nitrogen deposition compared to the trees (Gundale et al., 2011, Gundale et al., 2014), and N deposition favors the growth of nitrophilous species like Pleurozium schreberi (Nordin et al., 2005, Suding et al., 2005, Hautier et al., 2009, Manninen et al., 2013, Dirnböck et al., 2014). Nitrogen deposition of up to 140 kg N ha-1 y-1 has been experimentally shown to increase growth in Scots pine, European aspen and hybrid aspen (Utriainen and Holopainen, 2001b, Häikiö et al., 2007). The effects of N addition in Scots pine also depends on the N source (Gruffman et al., 2014) .
2.3.9. Nitrogen deposition and soil microbes
In addition to having effects on above-ground biodiversity, N deposition also effects the below- ground biota and their activities. Nitrogen deposition increases the organic matter decomposition and nitrogen mineralization by fungal species in boreal forests and it is also observed to cause a difference in their substrate preference, but the response of fungi varies among species (Högberg et al., 2003, Chen and Högberg, 2006, Allison et al., 2009). Under high plant productivity, leaf litter delivers more nutrients to the saprotrophic fungi hence increasing their activity, community structure and reducing their diversity by favoring some species over the others (Högberg et al., 2003, Allison et al., 2007).
Over 90 kg N ha-1 y-1 nitrogen dose has been observed to reduce carbon delivery to mycorrhizal fungi by Scots pine and Norway spruce, leading to a decline in the symbiotic mycorrhizal fungi population (Högberg et al., 2003, Högberg et al., 2007, Blasko et al., 2013, Leppälammi-Kujansuu et al., 2013).
Over 90 kg N ha-1 y-1 nitrogen deposition is also associated with acidification which can lead to an increase in the bacterial biomass of the soils (Högberg et al., 2007).
18
2.3.10. Interactive effects of warming, ozone exposure and nitrogen deposition on boreal forests and soil microbes
Interactive effects of the above mentioned three climatic variables have not been studied extensively especially in relation to soil microbial profiles. The interactive effects of two factors are relatively easy to find in the literature, however, most of such studies focus on the trees, not the underground bacterial and fungal community composition. Generally it has been found that the harmful effects of ozone on the growth of trees are partly compensated by increasing nitrogen content of soil or warming (Häikiö et al., 2007, Riikonen et al., 2009, Mäenpää et al., 2011, Kasurinen et al., 2012). Positive stimulation of growth of Scots pine in response to N addition has been reported to be reduced in the presence of ozone exposure (Utriainen and Holopainen, 2001b). Utriainen and Holopainen (2001b) have reported the reduction in the growth stimulation, they also reported an increase in needle formation in response to ozone exposure of N-sufficient seedlings indicating a change in C allocation pattern in response to the combined exposure (N addition and ozone exposure). Utriainen and Holopainen (2001a) have studied the interactive effects of ozone and N availability on Norway spruce and they observed that ozone exposure did not modify the growth promoting effect of N availability, concluding that Norway spruce is tolerant to ozone exposure (1.5x ambient concentration). Kasurinen et al. (2012) have shown that ozone reduces the stimulating effects of warming on soil respiration in silver birch trees. It is reported that nitrogen addition further increases the soil carbon dioxide fluxes under warming by altering both the soil organic matter and the soil microbial community (Coucheney et al., 2013). Warming and nitrogen addition also increase the total fine root biomass, but warming does not compensate for the negative effects of nitrogen on ectomycorrhizas (Leppälammi-Kujansuu et al., 2013). Leppälammi-Kujansuu et al. (2013) have reported a decrease in the EM roots per basal area in response to warming + N addition treatment in Norway spruce; in the study, while fine roots increased in response to warming + N addition treatment, the number of EM roots remained constant indicating increased independence of roots upon mycorrhizas for nutrient uptake.
2.4. Methods for characterizing soil microbial communities
Traditionally the study of soil microbial communities has relied upon culture-dependent methods of community analysis (Hill et al., 2000). Due to the uncertainty and low recovery of soil microorganisms through culture-dependent techniques, the scientists have recently turned to the culture-independent methods of soil biodiversity analysis (Torsvik et al., 1990, Hill et al., 2000). The culture-independent techniques rely upon molecular markers directly extracted from the environmental samples, without the need of culturing. Currently the nucleic acids or PLFAs are used as molecular markers in the culture-independent community analysis (Hill et al., 2000).
19
The nucleic acid-based techniques use polymerase chain reaction (PCR) to generate copies of a desired gene of microbial communities. Most commonly, ribosomal DNA is amplified to study the microbial (bacterial, fungal and archaeal) community structure. Catabolic genes can also be targeted in the nucleic acid based techniques to study the presence of certain species in the target environment (Wilson et al., 1999, Malik et al., 2008). This technique has limitations as both storage methods of environmental samples prior to DNA isolation (Moran et al., 1993) and methods used for DNA extraction and/or amplification may alter the results (van Winzingerode et al., 1997). Also the usefulness of this method for eukaryotic biodiversity analysis is questionable given the more complex gene structure among them (Hill et al., 2000).
In the current MSc work, PLFA analysis is used to study the bacterial and fungal communities in the rhizosphere area of Scots pine seedlings. Phospholipids are major component of the membrane lipid bi-layer of the prokaryotic and eukaryotic organisms (except archaea) (Frostegård and Bååth, 1996, Zelles, 1999, Frostegård et al., 2011). PLFA analysis uses the fatty acids, attached to the phosphate and glycerol molecules in the phospholipids, as they are specific for major taxonomic groups (Table 1). PLFA analysis is a culture independent method used extensively to study the total microbial profiles in environmental samples, and since PLFAs are degraded rapidly after the cell death, it provides a projection of active communities in the samples (Hill et al., 2000, Papadopoulou et al., 2011). This method is, however, less specific as compared to the nucleic acid based technique as PLFAs can only give us information at taxonomic level and not at species level (Haack et al., 1994, Hill et al., 2000, Bardgett, 2005). Here we study the profiles of gram-positive and gram-negative bacteria, fungi other than AMF (from here on referred as fungi (AMF-)) and AM fungi (from here on referred as fungi(AMF+)) as they constitute the major taxa of soil microflora excluding archaea.
20
Table 1. Common fatty acid signatures adapted from (Hill et al., 2000, Ngosong et al., 2012).The first letter in the name of fatty acid compounds describes the molecular structure, i = iso, a = anti, cy = cyclic. The first number tells the number of carbon atoms in the fatty acid chain, the second number after the colon tells the number of double bonds, the ‘ω’ tells the double bond location.
The letter after the double bond location represents the molecular structure of the double bond, t = trans and c = cis. Me = methyl group.
Taxa Common fatty acid signatures
Common bacterial signatures i-15:0+, a-15:0+, 15:0. 16:0, 16:1ω5, 16:1ω9-, i-17:0+, a-17:0+, 18:1ω7t, 18:1ω5, i-19:0-, a-19:0-
Aerobes 16:1ω7, 16:1ω7t, 18:1ω7t
Anaerobes cy-17:0, cy-19:0
Sulfate-reducing bacteria 10Me16:0, i-17:1ω7, 17:1ω6
Methane-oxidizing bacteria 16:1ω8c, 16:1ω8t, 16:1ω5c, 18:1ω8c, 18:1ω8t, 18:1ω6c Barophilic/ psychrophilic
bacteria
20:5, 22:6
Cyanobacteria 18:2ω6
Protozoa 20:3ω6, 20:4ω6
Fungi(AMF-) 18:1ω9, 18:2ω6, 18:3ω6, 18:3ω3
Actinobacteria 10Me18:0
Microalgae 16:3ω3
Fungi(AMF+) 16:1 ω5t
+ sign shows gram-positive signature PLFAs and – sign shows gram-negative signature PLFAs (Frostegård and Bååth, 1996, Taipale et al., 2009, Tavi et al., 2010)
21 3. AIMS OF THE STUDY
In this work the effects of warming, increased tropospheric ozone concentration, and increased nitrogen deposition on soil bacterial and fungal communities beneath Scots pine trees were studied alone and in combination. One year old seedlings were exposed to the above abiotic factors, in a three-year lasting open-air exposure study (2011-2013), and soil samples for PLFA analysis were collected after the third exposure season. The following research questions are being tested:
1. Does warming alter the microbial profiles and/or fungi:bacteria-ratio in Scots pine rhizosphere?
2. Does ozone stress alters the microbial profiles and/or fungi:bacteria-ratio in Scots pine rhizosphere?
3. Does nitrogen deposition (addition) alter the microbial profiles and/or fungi:bacteria-ratio in Scots pine rhizosphere?
4. Do the exposures have interactive effects on the microbial profiles and/or fungi:bacteria-ratio in Scots pine rhizosphere?
22 4. MATERIALS AND METHODS
4.1. The experimental design
The experiment was conducted in Ruohoniemi open-air exposure field (62°53’N, 27°37’E, 80m a.s.l), located near the University of Eastern Finland, Kuopio campus (Figure 2a). A total of 480 Scots pine (Pinus sylvesteris) seedlings were purchased from Finnish Forest Research Institute (FFRI) (from 2014 onwards National Resources Institution Finland), Suonenjoki station in spring 2011. In Kuopio, seedlings were planted into 5 liter pots at the University of Eastern Finland, Research Garden, and were moved to the experimental plots in Ruohoniemi at the end of May 2011. The seedlings were planted into sand:peat mixture (2:1 vol:vol, quartz sand and peat Kekkilä White F6, NPK 16-4-17).
The seedlings had received an initial fertilizer dose of 77 kg N ha-1 y-1 before the start of experiment in summer 2010 and were one-year-old at the beginning of the experiment. The experimental group consisted of 288 of the 480 seedlings, and 192 of the remaining seedlings were used as side plants in the plots. The exposures lasted for three growing seasons (2011-2013).
The experimental set-up consisted of eight open-air exposure plots half of which were elevated O3
plots (1.5x ambient ozone) and the rest were controls (ambient ozone concentration). Each plot was further divided into elevated temperature (temperature elevation c. +1°C) and ambient temperature subplots (Figure 1 b-c), and half of the trees growing in each subplot were N fertilized N dose 120 kg N ha-1 y-1), and the rest served as N controls. Thus experimental treatments were: i) control (ambient temperature + ambient O3 + prevailing N concentration in soil), ii) elevated temperature alone, iii) elevated O3 alone, iv) elevated N alone, v) elevated temperature + elevated O3, vi) elevated temperature + N addition, vii) elevated O3 + N addition, viii) elevated temperature + elevated O3 + N addition. The seedlings were randomly distributed among the experimental plots, and the pots were submerged into the subplot soil. The experimental seedlings were placed in each subplot in three rows, six seedlings per row (3 rows x 6 trees = 18 trees per subplot), while six side trees were placed on each long side of the subplot.
23 a)
b) c)
Figure 2. (a) Bird’s eye view of Ruohoniemi open-air field near University of Eastern Finland, Kuopio campus, showing the open air exposure fields. Black arrows point to ambient O3
concentration plots and white arrows point to elevated O3 concentration plots (https://www.google.fi/maps/@62.8949663,27.6252838,180m/data=!3m1!1e3). (b) Ozone exposure plot and warming subplots (photo by Jarmo Holopainen). (c) Ambient ozone exposure plot and warming subplots (photo by Jarmo Holopainen). The photos (b) and (c) also show subplots from a parallel experiment on aspen in the same plot.
24
4.2. Warming, ozone exposure and nitrogen addition in the plots
The warming exposure system used in this experiment is described in Kasurinen et al., (2012).
Warming exposure was realized by using infrared heaters (model Comfortintra CIR 105-220, 230- 400 V, Frico AB, Partille, Sweden) placed above tree canopy in the middle of the subplots. In the ambient temperature subplots, a wooden bar of same size, shape and color was installed to mimic the shading effect of the heater (Figure 2b-c). The heaters were kept on for 24 hours day-1. A difference of 70 cm was maintained between the tree canopy and the heater. The target increase in the air temperature was 1°C above the ambient temperature. The soil temperature in each subplot was monitored using two T-type thermocouples at a depth of 5 cm. Air temperature and humidity were measured in the subplots using Vaisala sensors (Vaisala, Vantaa, Finland). The Vaisala sensor was installed inside a plastic rack to prevent direct heating from the IR waves and to protect from rain water. The mean monthly air and soil temperatures for the subplots were calculated and are presented in Table 2. The average increase in the air temperature was 0.8°C in 2011, 0.9°C in 2012 and 0.8°C in 2013, and the average increase in the soil temperature was 1.3°C in 2011 and 0.5°C in 2012 and 2013. The temperature exposures lasted from 6th June to 7th October in 2011 (124 days); 1st May to 30th September in 2012 (153 days) and in 2013 from 6th May to 25th October (173 days).
A free-air ozone exposure system was employed to realize the ozone exposure in the plots (Häikiö et al., 2007, Karnosky et al., 2007). Pure oxygen was used to produce ozone using the ozone generator (G21; Pacific Zone Technology Inc., Brentwood, California, USA). Ozone fumigation ran from 8 am to 10pm every day. Ozone was released into the air through the vertical vent pipes in upwind direction (Figure 2b). Ozone concentrations in the field were monitored using Dasibi 1008-RS ozone analyzer (Dasibin Environmental Corp., Glendale, California, USA). The target for ozone elevation was 1.5 x ambient ozone concentration. The mean ozone levels in ozone exposure subplots were 1.4 x in 2011, 1.5 x in 2012 and 1.6 x ambient concentration in 2013. The mean monthly ozone concentrations and corresponding AOT40 values over the three seasons are shown in Table 2. The ozone exposures lasted from 31st May to 2nd October in 2011 (125 days); 21st May to 30th September in 2012 (133 days) and in 2013 from 7th May to 30th September (147 days).
Nitrogen addition was done twice during the experiment. On both occasions the nitrogen dose was 120 kg N ha-1 y-1 (4,619g per pot), and the first fertilization was performed in June 2011 and the latter in May 2013. The fertilizer used was Yara Peatcare TM, Slow release 1, (N:P:K 9:3.5:5).
25
Table 2: The average O3 concentrations (based on 14h day-1 of O3 exposure data) and AOT 40 (accumulated dose over a threshold of 40 ppb) values in ambient O3 (AO) and elevated O3 (EO) plots over the exposure seasons 2011-2013. Monthly averages for ambient air temperature (AAT) and elevated air temperature (EAT) (24h day-1), for ambient soil temperature (AST) and elevated soil temperature (EST) (24h day-1), and for relative humidity (AT RH% and ET RH%) in ambient and elevated temperature subplots (all based on 24h day-1 measurement data).
Year Month AO
monthly Average (ppb)
EO monthly average (ppb)
AOT40 AO (ppmh)
AOT40 EO (ppmh)
AAT monthly average (°C)
EAT monthly average (°C)
AST monthly average (°C)
EST monthly average (°C)
AT RH (%)
ET RH (%)
2011 May 37.8 38.7
June 28.9 ± 6.1 39.8 ± 11.2 16.6±0.3 17.5±0.4 16.5±1.1 17.7±1.4 71.4±4.4 67.5±1.6 July 26.5 ± 5.3 39.6 ± 9.8 19.8±0.4 20.8±0.4 20.2±1.2 21.5±1.6 75.7±2.3 71.8±1.5 August 19.2 ± 5.8 29.5 ± 9.6 15.9±0.6 16.8±0.4 16.6±1.0 18.1±1.7 80.5±2.9 75.6±0.9 September 17.7 ± 4.4 27.9 ± 9.0 12.0±0.7 12.6±0.3 12.6±1.0 13.9±1.6 86.2±4.3 80.8±1.2 October 19.2 ± 2.2 33.6 ± 6.8 0.28 6.39 8.6±0.7 9.2±0.3 8.8±1.1 10.2±1.6 87.0±6.6 80.9±1.2 2012 May 33.3 ± 6.4 55.0 ± 12.0 10.4±0.7 10.9±0.9 10.2±0.9 10.7±1.0 62.5±4.5 62.5±1.5 June 31.6 ± 7.3 50.5 ± 15.2 14.2±0.8 15.0±0.5 14.8±1.3 15.7±1.3 73.5±2.0 69.7±1.1 July 4.9 ± 4.6 6.3 ± 5.2 17.8±0.8 18.9±0.4 17.7±1.2 18.1±1.2 77.4±2.0 72.4±1.3 August 19.2 ± 6.1 29.5 ± 9.3 15.2±0.9 16.4±0.3 15.6±1.2 15.8±1.0 79.8±2.2 73.4±1.3 September 20.6 ± 4.5 30.2 ± 9.3 0.45 9.87 10.5±0.7 11.5±0.4 10.8±1.3 11.1±1.2 84.8±2.0 79.6±1.3 2013 May 32.5 ± 6.6 52.5 ± 13.6 14.0±0.8 15.0±1.0 13.7±1.5 13.7±1.1 65.8±2.0 61.1±2.6 June 33.0 ± 5.4 51.8 ± 12.2 18.2±0.8 19.1±0.5 17.9±1.4 18.4±1.7 69.8±1.8 65.6±1.4 July 26.4 ± 5.4 42.7 ± 12.1 17.2±0.8 17.9±0.6 17.3±1.4 17.7±1.5 75.1±2.2 70.9±1.6 August 24.8 ± 5.9 39.2 ± 12.6 16.5±0.8 17.3±0.5 16.3±1.4 16.8±1.4 80.3±1.8 75.7±1.4 September 19.7 ± 4.9 26.7 ± 11.5 0.79 20.92 11.3±0.9 12.2±0.5 11.6±1.4 12.2±1.5 84.1±1.9 79.8±1.7
October 5.1±0.7 5.5±0.5 4.5±1.4 5.2±1.6 84.8±1.8 81.9±2.2
26 4.3. Sampling for PLFA analysis
The soil samples for PLFA were collected in early October 2013. Soil was separated from roots by gently shaking the seedlings mixed properly, and then c. 50-100 g of fresh soil was randomly collected around the soil pile. Soil was sampled from two seedlings per subplot separately, but later these two samples per subplot were mixed together to form a new pooled sample. Thus, in PLFA analysis there were 8 treatments x 4 replicates/treatment = 32 samples. The samples were kept frozen at -20°C until freeze-drying. The extraction of PLFAs was performed in April 2014.
4.3.1. PLFA extraction
All the reagents and solvents used were of analytical grade. Commercially available 37 mix fatty acid methyl esters (FAME, purchased from SUPELCO®, USA, Lot: 47885-U) were used as external standard. Dipentadecanoylphostatidylcholine (c15:0) (Larodan Fine Chemicals) was added as internal standard for quantification while extracting the fatty acids. The glassware and extraction tubes were all washed with water, baked at 400°C and rinsed with acetone to remove residual carbon before use.
A three-step protocol was followed in the PLFA extraction (1) total lipid extraction, (2) separation of total lipid into neutral, glycol- and phospholipids and (3) methylation of the fatty acids of the phospholipid fraction (Bossio et al., 1998, Spyrou et al., 2009, Papadopoulou et al., 2011). The extraction of PLFAs followed the modified Blight and Dyer (1959) method. In the extraction of total lipids, c. 6 g of freeze-dried soil sample was treated with phosphate buffer, methanol and chloroform (0.8:2:1 vol:vol:vol) in that order. After shaking the mixture overnight on a shaker, the internal standard was added to it and then the mixture was shaken for further 5 minutes. The mixture was centrifuged and the supernatant was decanted into new set of tubes. The volume of the supernatant was measured and subsequently more chloroform and phosphate buffer were added to obtain a ratio of 1:1:0.9 (vol:vol:vol chloroform, methanol and p-buffer, respectively). After mixing and centrifuging, the lower organic phase (containing total lipids) was collected and dried under the nitrogen stream and stored for separation of phospholipids from glycolipids and neutral lipids.
The dried lipids were re-dissolved in chloroform and poured onto ready-made separation columns (Agilent silica-based HF Bond Elut LRC-SI, 500 mg, varian) in a fume chamber. Phospholipids were separated from glyco- and neutral lipids. The neutral lipids were separated by dissolving into chloroform, glycolipids with acetone and phospholipids with methanol. Methanol fraction (phospholipid fraction) was dried under nitrogen stream and stored in the freezer (-20°C) for the next step.
27
The phospholipids were then saponificated and methylated using 1ml of 1:1MeOH:Toluene and 1ml of 0.2M methanolic KOH (freshly made). Methylated esters were then analyzed using Agilent 6890 GC connected to an Agilent 5973 mass selective detector (Tavi et al., 2013). The methylated FAs were separated with a DB-5 fused silica capillary column (30 m x 0.25 mm x 0.25), using helium as a carrier gas. The samples were injected by splitless injection using the constant flow mode and using 50°C as the initial temperature for 1 minute. The temperature was increased at the rate of 30°C min-
1 until it reached 140°C and then the rate of increase of temperature was maintained at 5°C min-1 until it reached 320°C which was maintained for another 20 minutes. The total runtime for gas chromatography was 60 minutes per sample. Identification of PLFA peaks were based on retention time and comparison with the known peaks of the external standard.
Most abundantly found signature PLFA compounds from the soil samples were chosen for the analysis. PLFAs i-14:0, i-15:0, a-15:0, i-16:0, br-17:0, i-17:0 and a-17:0 were used as biomarkers for gram-positive bacteria, while 15:1, 16:1ω9, 16:1ω7c, cy-17:0 18:1ω7c, i-19:0, a-19:0 and cy-19:0 were used as biomarkers for gram-negative bacteria (Frostegård and Bååth, 1996, Taipale et al., 2009, Tavi et al., 2010). PLFA 18:2ω6c and 18:1ω9c were used as biomarkers for fungi(AMF-) (all fungi except AMF) (Frostegård and Bååth, 1996, Bååth, 2003, Tavi et al., 2010, Olsson, 1999). PLFA 16:1ω5t was used as biomarker for fungi(AMF+) (AMF only) (Ngosong et al., 2012) (Table 3).
4.4. Statistical analysis
In the linear mixed model ANOVAs, temperature, ozone concentration and nitrogen addition levels (each had two levels) were used as fixed factors, whereas term subplot(plot) was used as a random factor. Tested variables were relative proportions (% of total PLFAs, proportions calculated on the basis of concentration data) of four major taxonomic groups (Table 3). In addition, individual PLFA concentrations (ng g-1 DW of soil) and total concentrations (ng g-1 DW of soil) of four taxonomic groups (gram-negative and gram-positive bacteria, fungi and AM fungi) were also tested with LMM ANOVA. The fungi:bacteria-ratio was calculated by dividing the sum of total signature PLFA concentrations of fungi(AMF-) and fungi(AMF+) by the sum of total signature PLFA concentrations of gram-positive and gram-negative bacteria:
fungi: bacteria = [fungi(AMF−)] + [fungi(AMF+)]
[gram − positive bacteria] + [gram − negartive bacteria]
Before statistical analyses, one outlier was omitted from the tests, thus n = 3-4 per treatment.
Normality and heterogeneity of the variances of the data were checked from the residuals. All statistical tests were performed with SPSS version 21 for Windows (SPSS Inc., Chicago, IL, USA).
28
P-values ≤ 0.05 were considered as statistically significant and P-values ≤ 0.1 as marginally statistically significant trend.
Table 3. Common fatty acid signatures used in the PLFA analysis in the current study. The first letter in the name of fatty acid compounds describes the molecular structure, i = iso, a = anti, cy = cyclic. The first number tells the number of carbon atoms in the fatty acid chain, the second number after the colon tells the number of double bonds, the ‘ω’ tells the double bond location. The letter after the double bond location represents the molecular structure of the double bond, t = trans and c = cis. Me = methyl group.
Taxa Common fatty acid signatures
Gram-positive bacteria i-14:0, i-15:0, a-15:0, i-16:0, br-17:0, i-17:0, a-17:0
Gram-negative bacteria 15:1, 16:1ω9, 16:1ω7c, cy-17:0 18:1ω7c, i-19:0, a-19:0, cy-19:0
Fungi (AMF-) 18:2ω6c, 18:1ω9c
Fungi(AMF+) 16:1ω5t
29 5. RESULTS
5.1. Treatment effects on relative proportions of PLFAs and fungi:bacteria ratio
In general, gram-positive and gram-negative bacteria were the dominating groups in the soil (their combined relative proportions were over 60 % in all the treatments, (Figure 3a-b), while the proportions of fungi(AMF+) was always less than 10 % (Figure 3d). The fungi:bacteria-ratio remained less than 1 for all the treatments (Figure 4).
Nitrogen (N) addition changed the proportions of gram-negative bacteria and fungi(AMF-) significantly (Figures 3b-c, N main effect P-value ≤ 0.05, Table 4). The N effect on these two groups were opposite, as N addition reduced the proportion of gram-negative bacteria (Fig 3b) and increased that of fungi (AMF-) (Fig 3c). N addition effect on fungi(AMF-) proportion was also reflected in fungi:bacteria-ratio as it was also increased in most N addition treatments when compared to those treatments without N addition (Figure 4, N main effect P-value ≤ 0.05, Table 5). In addition, there was a statistically significant effect of warming on fungi(AMF-) as its relative proportion was increased due to warming (Figure 3c; warming main effect P-value ≤ 0.05, Table 4). Furthermore, a statistically significant ozone x warming effect on fungi(AMF-) showed that warming in combination with O3 cancels a slight decrease caused by ozone alone to the relative proportion of fungi(AMF-) regardless of N level (Figure 3c, warming x ozone interactive effect P-value ≤ 0.05, Table 4). Increase in fungi(AMF-) due to warming were also reflected in fungi:bacteria-ratio as all warming treatments (T, TO, TN and TNO) had higher ratios than their non-warmed counterparts (C, O, N, and NO);
(Figure 4. Table 5). There were no clear warming or ozone effects on the relative proportions of bacteria (gram-positive or gram-negative) and fungi(AMF+) (Figure 3a-b, d Table 4).
30
a) b)
c) d)
Figure 3. Relative proportions of (a) gram-positive bacteria, (b) gram-negative bacteria, (c) fungi(AMF- ), and (d) fungi(AMF+) in the rhizosphere soil of Scots pine seedlings. Error bars show standard error.
Treatment abbreviations: C = control; T = elevated temperature; O = elevated ozone; N = nitrogen addition; TO = elevated temperature and ozone in combination; TN = elevated temperature and nitrogen addition in combination; NO = nitrogen addition and elevated ozone in combination; TNO = elevated temperature, elevated ozone and nitrogen addition in combination, n = 3-4 per treatment.
0 10 20 30 40 50
C T O TO N TN NO TNO
Relative proportion (%)
Treatment
Gram-positive
0 10 20 30 40 50
C T O TO N TN NO TNO
Relative proportion (%)
Treatment
Gram-negative
0 10 20 30 40 50
C T O TO N TN NO TNO
Relative proportion (%)
Treatment
Fungi(AMF-)
0 5 10 15 20 25
C T O TO N TN NO TNO
Relative proportion (%)
Treatment
Fungi(AMF+)
31
Table 4. Ozone, warming and nitrogen main and interaction effects (LMM-ANOVA) on relative proportions of gram-positive bacteria, gram-negative bacteria, fungi(AMF-) and fungi(AMF+) PLFAs found from rhizosphere soil beneath Scots pines. The bold values in the table represent statistically significant (P ≤ 0.05) P-values, n = 3-4 per treatment.
Microbial group Relative proportion (%)
F P-value
Gram-positive
Ozone 1.745 0.200
Warming 0.413 0.527
Nitrogen 0.344 0.563
Ozone x Warming 0.575 0.456
Ozone x Nitrogen 1.378 0.252
Warming x Nitrogen 1.238 0.277
Ozone x Warming x Nitrogen 0.233 0.634
Gram-negative
Ozone 0.005 0.946
Warming 0.576 0.456
Nitrogen 10.286 0.004
Ozone x Warming 2.415 0.134
Ozone x Nitrogen 0.634 0.434
Warming x Nitrogen 0.445 0.512
Ozone x Warming x Nitrogen 0.643 0.431
Fungi (AMF-)
Ozone 1.909 0.180
Warming 4.400 0.047
Nitrogen 11.271 0.003
Ozone x Warming 5.110 0.034
Ozone x Nitrogen 0.025 0.876
Warming x Nitrogen 0.048 0.829
Ozone x Warming x Nitrogen 0.890 0.355
Fungi(AMF+)
Ozone 0.091 0.765
Warming 0.988 0.331
Nitrogen 0.369 0.549
Ozone x Warming 1.593 0.220
Ozone x Nitrogen 0.003 0.958
Warming x Nitrogen 0.003 0.954
Ozone x Warming x Nitrogen 1.850 0.187
32
Figure 4. Fungi:bacteria-ratio calculated from the relative abundance of total fungal and total bacterial PLFAs in rhizosphere soil of Scots pine seedlings. Error bars show standard error.
Treatment abbreviations: C = control; T = elevated temperature; O = elevated ozone; N = nitrogen addition; TO = elevated temperature and ozone in combination; TN = elevated temperature and nitrogen addition in combination; NO = nitrogen addition and elevated ozone in combination; TNO
= elevated temperature, elevated ozone and nitrogen addition in combination, n = 3-4 per treatment.
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
C T O TO N TN NO TNO
fungi/bacteria
Treatments
