Short-term utilization of carbon by the soil microbial community under future climatic conditions in a temperate heathland

Soil Biology & Biochemistry 68 (2014) 9e19

Contents lists available at ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Short-term utilization of carbon by the soil microbial community under future climatic conditions in a temperate heathland Sabine Reinsch a, *, Anders Michelsen b, Zsuzsa Sárossy a, Helge Egsgaard a, Inger Kappel Schmidt c, Iver Jakobsen a, Per Ambus a a b c

Department of Chemical and Biochemical Engineering, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark Department of Biology, University of Copenhagen, Universitetsparken 15, 2100 Copenhagen Ø, Denmark Department of Geosciences and Natural Resource Management, University of Copenhagen, Rolighedsvej 23, 1958 Frederiksberg, Denmark

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 March 2013 Received in revised form 25 August 2013 Accepted 6 September 2013 Available online 20 September 2013

An in-situ 13C pulse-labeling experiment was carried out in a temperate heath/grassland to study the impacts of elevated CO2 concentration (510 ppm), prolonged summer droughts (annual exclusion of 7.6
0.8%) and increased temperature (w1  C) on belowground carbon (C) utilization. Recently assimilated C (13C from the pulse-label) was traced into roots, soil and microbial biomass 1, 2 and 8 days after pulse-labeling. The importance of the microbial community in C utilization was investigated using 13 C enrichment patterns in different microbial functional groups on the basis of phospholipid fatty acid (PLFA) biomarker profiles. Climate treatments did not affect microbial abundance in soil or rhizosphere fractions in terms of total PLFA-C concentration. Elevated CO2 significantly reduced the abundance of gram-negative bacteria (17:0cy), but did not affect the abundance of decomposers (fungi and actinomycetes) in rhizosphere fractions. Drought favored the bacterial community in rhizosphere fractions whereas increased temperature reduced the abundance of gram-negative bacteria (19:0cy) and changed the actinomycetes community (10Me16:0, 10Me18:0). Fastest and highest utilization of recently assimilated C was observed in rhizosphere associated gram-negative bacteria followed by gram-positive bacteria. Utilization of recently assimilated C by rhizosphere associated actinomycetes and fungi was relatively low, but much more pronounced in the soil. The utilization of recently assimilated C by the microbial community was faster under elevated CO2 conditions compared to ambient. We conclude that changing climatic conditions will affect C utilization by the soil microbial community but might not drastically change the terrestrial C balance. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Elevated CO2 Drought Increased temperature Phospholipid fatty acids 13 CO2 Pulse-labeling CLIMAITE

1. Introduction Atmospheric carbon dioxide (CO2) concentration, temperature and water availability are critical determinants of the terrestrial carbon (C) turnover. Increasing atmospheric CO2 concentration exerts a positive feedback on CO2 concentration and global warming (IPCC 2007). Warming often increases soil CO2 efflux by e.g. extending plant growth periods (Luo, 2007), stimulating microbial activity and can lead to reduced soil C residence times (Amundson, 2001; Heimann and Reichstein, 2008). Temperature increase can turn low productive environments into physiologically active systems (Miller and Smith, 2012), but can also negatively

* Corresponding author. Tel.: þ45 21 32 53 80. E-mail address: [email protected] (S. Reinsch). 0038-0717/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.soilbio.2013.09.014

affect water availability leading to reduced gross primary production and soil C loss (Ciais et al., 2005). Rising CO2 concentration has been observed to have pronounced effects on terrestrial C turnover. Plant C uptake can be increased under elevated CO2 concentration (Ainsworth and Long, 2005; Albert et al., 2011) leading to increased aboveground and belowground biomass and can result in litter increased in lignin content and/or higher C:N ratios (Ball, 1997; Henry et al., 2005). Increased root growth can be stimulated by elevated CO2 concentration (Arndal et al., 2013) leading to increased rhizodeposition into the soil matrix (Jones et al., 2009). These plant mediated changes in the rhizosphere may change the microbial community towards a decomposer based community: increased lignin content in litter demands lignin degrading enzymes produced by microorganisms (Couteaux et al., 1995) and increased root growth and rhizodeposition potentially increases plant-microbe nutrient competition and induces soil organic matter priming mediated by

10

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

decomposers (Paterson et al., 2008). However, effects of elevated CO2 concentration on the soil microbial community differ across ecosystems (Dunbar et al., 2012): increased CO2 concentration led to a fungal dominated community compared to ambient soils in the Mojave desert (Jin and Evans, 2010). Furthermore, the soil microbial community in a temperate grassland exposed to elevated CO2 concentration showed increased arbuscular mycorrhizal fungi abundance but an overall decreased fungal activity (Denef et al., 2007). Interestingly, elevated CO2 stimulated the microbial community in a temperate grassland (Sowerby et al., 2000; Drissner et al., 2007) but the soil C stock remained unchanged (Theis et al., 2007). Extensive drought periods can reduce overall ecosystem activity, resulting in reduced rhizodeposition and lower microbial activity due to water limitation (Jensen et al., 2003). Low soil water content is a stress factor for soil organisms. However, fungi might be better adapted to limited water availability because their hyphal network facilitates water transport (Augé, 2001). In contrast, warming, if not imposing water limiting conditions, is expected to enhance bacterial abundance, but less fungal appearance (Frey et al., 2008). In general, longer-term effects of increased temperatures and water availability on the composition and activity of microbial communities are sparsely reported and often inconsistent. The fungal-to-bacterial ratio increased under warmed conditions in a tallgrass prairie (Zhang et al., 2005), and 12 years warming of a forest soil reduced fungal abundance, but stimulated gram-positive bacteria (Frey et al., 2008). In contrast, warming did not affect the microbial community composition in a temperate mountain forest (Schindlbacher et al., 2011). The combination of increased temperature with elevated CO2 concentration and two levels of precipitation in an American old-field grassland revealed the potential of anticipated future temperatures to increase the abundance of gram-positive bacteria, whereas the abundances of gram-negative bacteria, arbuscular mycorrhizal and saprophytic fungi were decreased (Gray et al., 2011). In the same study, precipitation affected the soil microbial community only in combination with a temperature treatment. One year later, however, precipitation was the main predictor for changes in the microbial community (Castro et al., 2010) illustrating the need for detailed investigations of climate change effects on the soil microbial community over time. Phospholipid fatty acid (PLFA) extraction has frequently been used to assess the microbial community composition in soils and has been proven to be an appropriate method to investigate environmental effects on the soil microbial community (Frostegård et al., 1993, 2011). PLFAs combined with stable C isotope analysis can additionally be used to investigate the metabolic activity of microbial functional groups (Treonis et al., 2004; Denef et al., 2007). Sequential measurements of microbial 13C incorporation after a 13 CO2 pulse facilitate the tracing of recently assimilated C through the microbial community to evaluate not only the abundance, but also the importance of each microbial functional group in terrestrial short-term C turnover (Jin and Evans, 2010). The microbial activity is important because functional redundancy in microbial communities complicates the investigation of the importance of microbial groups on the basis of community diversity and abundance only (Nannipieri et al., 2003). In the present study, in-situ 13CO2 pulse-labeling was performed in a temperate heath/grassland to investigate climate impacts on the utilization of recently assimilated C by the microbial community. Soil microbial functional groups were identified by phospholipid fatty acid (PLFA) abundances and isotopic PLFA characteristics were related to microbial activity associated with turnover of recently assimilated C. The experiment comprised the climate treatments: ambient and elevated CO2 concentrations, summer

Fig. 1. Soil and environmental conditions during the time of the experiment in May 2011. Arrows indicate the pulse-labeling events. (a) averaged daily photosynthetic active radiation (PAR, bars) and soil temperature at five cm depth in plots with ambient temperature (open symbols) and elevated temperature (solid symbols); (b) daily precipitation (bars) and volumetric soil water contents (0e20 cm depth) in plots without drought treatment (open symbols) and drought-treated plots (solid symbols).

drought and increased temperature. The full factorial treatment reflects the anticipated climatic conditions for Denmark in 2075. We hypothesize that (i) elevated CO2 favors a fungal-based community, (ii) increased temperature leads to a bacterial-based community, (iii) prolonged drought increases the fungal abundance, (iv) C turnover is faster under elevated than ambient CO2 concentration and (v) C utilization by the microbial community is similar between ambient and future climatic conditions due to the balance of positively acting (elevated CO2, temperature) and negatively acting (drought) climate factors. 2. Methods 2.1. Study site The study site was an unmanaged dry heath/grassland in North Zealand, Denmark (55 530 N, 11580 E) situated in a hilly area characterized by sandy, nutrient-poor glacial deposits composed of w70% sand, w20% coarse sand and minor proportions of silt and clay. The climate is temperate with a mean annual temperature of 8  C and a mean annual precipitation of w610 mm. The plant community was co-dominated by the grass Deschampsia flexuosa (w70%) and the dwarf shrub Calluna vulgaris (w30%), and minor

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

occurrences of other grasses, herbs, mosses and lichens (Kongstad et al., 2012). Total aboveground green biomass varied on a seasonal scale between w300 and 730 g m2 measured in campaigns between 2004 and 2008 (Kongstad et al., 2012). Root biomass was w600 g m2 (Arndal et al., 2013) with about 80% of roots located in 0e10 cm soil depth. The experimental setup was a full-factorial split-plot design that consisted of an un-treated control (A), elevated CO2 concentration at 510 ppm during daytime hours by free-air CO2 enrichment (CO2), prolonged spring/summer droughts by horizontally moving curtains (D), increased temperature (1  C) realized with passive night time warming by reflective curtains (T) and the factorial combinations TD, DCO2, TCO2, TDCO2. Climate treatments were initiated in 2005. Treatments were applied in 12 octagons (6.8 m in diameter) arranged pair-wise in 6 blocks resulting in 6 replicates per treatment. Within each block, 1 octagon was exposed to elevated CO2, and each octagon was divided into four plots, giving a total of 48 experimental plots. Soil moisture (vol %), soil temperature ( C) and photosynthetic active radiation (PAR, mmol m2 s1) were continuously measured at the field site. For further detailed information about the experimental site, see Mikkelsen et al. (2008) and Selsted et al. (2012). Sub-plots for the 13CO2 labeling were confined by galvanized steel collars (0.8  0.4  0.1 m3) that were installed two weeks prior to labeling to minimize effects of disturbance. The experiment was conducted in two campaigns in May 2011 after 6 years of treatment, each of which included 3 experimental blocks. During the first campaign on May 16th (Octagons 1e6) weather conditions were dominated by gray skies and light showers whereas during the second campaign on May 19th (Octagons 7e12) sunny conditions prevailed (Fig. 1). On both days, the pulse-labeling took place from 12:00 to 16:00 h. Only plots containing more than 80% D. flexuosa were included (41 plots); each treatment was replicated 4e6 times. The drought treatment in 2011 was initiated two weeks prior to the pulse-labeling (May 2nd), and 11.2 mm and 29.5 mm of precipitation had been excluded before the first and second labeling campaigns, respectively. 2.2. In-situ

13

CO2 pulse-labeling

The labeling setup has been described in detail in Reinsch and Ambus (2013). In brief, on the days of labeling, transparent flowthrough Plexiglas chambers (0.8  0.4  0.3 m3) were mounted gas tight on top of soil frames by means of a water filled channel. Fans inside the chambers assured air mixing and 2 blocks of “blue ice” per chamber reduced heating of the chamber air. Incoming air (50 atom% 13CO2, Cambridge Isotope Laboratories, Inc., Saint Aubin, France) was provided from an air reservoir by an electric diaphragm pump (Model Thomas 107CCD20-164, vacuum pump, Gardner Denver Sweden AB, Bandhagen, Sweden) through Polyurethane tubing (TU0604, 4 mm, Polyurethane Tubing, SMC Pneumatic A/S, Horsens, Denmark) at a flow rate of 2.1 L chamber1 min1. The provided CO2 accounted for the maximum measured CO2 consumption of D. flexuosa (Selsted et al., 2012). Each octagon was equipped with one air reservoir (gas tight vinyl balloons, w3 m in diameter, Balloons Etc/Balloons Direct, Springfield, Virginia, USA) providing air to four chambers at 390 ppm CO2 for ambient CO2 treatments and at 510 ppm CO2 for elevated CO2 treatments. CO2 concentrations and 13C concentrations were stable in ambient and elevated CO2 treatments over the course of the experiment, only the 13C concentration in labeling chambers in elevated CO2 treatments changed slightly due to biotic processes over the course of the experiment (Reinsch and Ambus, 2013). The pulse-labeling started immediately when the chambers were mounted on the soil frames and stopped when the chambers

11

were removed after 4 h of exposure. All chambers were opened temporarily after 2 h to remove condensed water on the inside chamber walls and to replace the cooling packs. Reservoir and chamber air were sampled for all single factor drought plots at onehour intervals during the labeling period. Drought plots were chosen because it was the only single-factor climate treatment where the vegetation criteria matched in all replicates (n ¼ 12). For gas sampling, 20 mL samples were taken with a syringe (3003310, 20 mL Omnifix 3 component syringe, MEDIQ Danmark A/S, Brøndby, Denmark) from the air reservoirs and chamber outlets and flushed through pre-evacuated 5.9 mL Exetainer vials (819W, ExetainersÒ, Labco Ltd, High Wycombe, UK). Vials were stored at room temperature until analyses. 2.3. Soil sampling Pre-labeling soil samples were taken adjacent to the soil frames on May 13th. Post-labeling soil samples were taken 1, 2 and 8 days after the labeling inside the collars. Soil samples consisted of two bulked soil cores (5 cm diameter, 10 cm deep). Soil samples were transported to the laboratory for further processing the same day. Samples were sieved (2 mm diameter) for a maximum of 5 min. Unwashed roots were immediately frozen (20  C) for subsequent freeze drying and further referred to as rhizosphere fraction (roots þ rhizosphere soil). Soil subsamples were processed either freshly (chloroform fumigation extraction), air dried (C/N analysis, SWC) or frozen for subsequent freeze drying (PLFA extraction). Remaining soil was stored at 20  C. 2.4. Carbon and nitrogen analyses Carbon and nitrogen contents and isotopic compositions of soil and root samples were measured on an Isoprime isotope ratio mass spectrometer (Isoprime Ltd, Cheadle, UK) coupled to an Eurovector CN elemental analyzer (Eurovector CN EA, Milan, Italy) using continuous flow. Samples were prepared by weighing 7e20 mg air dried soil and 2e5 mg air dried (unwashed) ground roots into tin capsules. Microbial biomass C (MBC) was assessed using chloroform fumigation extraction. Paired fresh 10 g soil samples were immediately extracted (1:5 w:vol) with MilliQ water for 1 h (Andresen et al., 2009, 2010a), or extracted upon fumigation with chloroform for 24 h (Vance et al., 1987). Total organic C in the water extracts was determined (TOC-VCPH, Shimadzu, Holm & Halby, Denmark). MBC was calculated from the difference in C between fumigated and non-fumigated soils using a kEC factor of 0.45 (Joergensen, 1996). Non-fumigated (2e8 mL) and fumigated (0.5e 5 mL) soil extracts were freeze dried on quartz filters (Quartz microfiber filter QMA Whatman) that were combusted for 13C/12C analysis on the EA-IRMS (Isoprime). Gravimetric soil water content was determined from the weight loss of a 10 g soil sample after oven drying (105  C, 24 h). 2.5. Phospholipid fatty acid extraction and analyses Phospholipid fatty acids (PLFAs) were extracted from about 3 g freeze dried soil and 7e11 mg unwashed, freeze dried roots (rhizosphere fraction). Rhizosphere extractions were carried out for all sampling days, and soil extractions were carried out for prelabeling samples and samples taken 2 days after the 13CO2 pulse. A one-phase mixture (CHCl3:MeOH:citrat buffer) according to a modified Bligh and Dyer protocol was used (Bligh and Dyer, 1959; White et al., 1979; Frostegård et al., 1991). Internal sample standards C17:0 and C19:0 were added before methanolysis as Glyceryltriheptadecanoate and Glyceryltrinonadecanoate (T2151 and

12

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

T4632, Sigma Aldrich, Denmark). FAMEs were analyzed for 13C enrichment via gas-chromatography-combustion-isotope ratio mass spectrometry (GC-c-IRMS). The GC (Agilent, HP6890, Denmark) was equipped with a 60 m fused silica column (Varian FactorFour WCOT, 0.25 mm id.  0.25 mm thickness) using He as carrier gas (1 mL min1). The GC was coupled via a GC combustion interface (Thermo Scientific, Bremen, Germany) in continuous flow mode with a Finnigan DeltaPLUS isotope ratio mass spectrometer (Thermo Scientific). For further details about the injection cycle see Reinsch et al. (2013). Only peaks with a peak area (PA) above a ratio of 0.02 for PA/PA-C16:0 were considered in further calculations. Independent analysis of washed roots resulted in undetectable FAME concentrations (not shown). FAMEs were analyzed for identification by GCeMS (Hewlett Packard HP 6890 GC) interfaced to a Mass Selective Detector (HP5973, Agilent, Denmark). Samples (1 mL) were injected in split mode (1:20) using an autosampler (HP 7683, Agilent, Denmark). The source and rod temperatures were 230  C and 150  C, respectively. The products were separated on a 30 m fused silica column (WCOT, 0.32 mm id.  0.25 mm thickness, Analytical, Denmark) using He as carrier gas (1.2 mL min1). Products were identified using the NIST search engine version 2.0 f. (Agilent, Denmark). The identification was based on the mass spectrum of the compounds and the suggestion of the NIST library search. The identification was also supported by the retention times of two standard mixtures, a Supelco FAME mix and a bacterial acid methyl ester mix (47080-U: BAME mix, 47885-U: Supelco 37 component FAME mix, SigmaeAldrich). The d13C values of the individual fatty acids were corrected for the C added during methanolysis (Denef et al., 2007), where the measured d13C value for the methanol used for methanolysis was 37.7
3.2&. PLFA concentrations (mmol PLFA-C) were used to calculate the mole fraction (%) of each individual PLFA. The amount of each PLFA in the extracts (mg PLFA-C g1 rhizosphere fraction or soil) was calculated relative to the internal standards C17:0 and C19:0 of known concentrations. The soil microbial community composition was investigated on the basis of PLFAs specific for different microbial groups (Vestal and White, 1989; Frostegård and Bååth, 1996; Zak et al., 1996; Zelles, 1999; Jin and Evans, 2010). Specific PLFAs characteristic for gramnegative bacteria were the cyclic 17:0cy and 19:0cy as well as 16:1u7c and 18:1u7c, for gram-positive bacteria the PLFAs 15:0a, 15:0i, 16:0i, 17:0a, 17:0i were characteristic. Actinomycetes (belonging to gram-positive bacteria) were specifically distinguished on the basis of methylated PLFAs (10Me16:0, 10Me17:0, 10Me18:0). Fungal abundance was based on the 18:1u9c, 18:2u6,9 and 18:3u3 PLFA biomarkers. General PLFAs were 14:0, 15:0, 16:0 and 18:0. 2.6. Isotopic calculations and data analyses The carbon 13C/12C isotopic ratios are expressed by the d13C notation vs. Vienna Pee Dee Belemnite (VPDB) reference material calculated as d13C (&) ¼ (Rsample/Rstandard  1) * 1000, where R ¼ 13C/12C. As working standard we used pure CO2 that had been calibrated against certified IAEA reference material (IAEAeCHe6, Sucrose). The 13C excess in different C pools was calculated as difference in total atom% 13C of pulse-labeled samples (post-labeling) and non-labeled samples (pre-labeling), and expressed as the atom % 13C excess (APE). Due to low numerical values, the reported APE is multiplied by 103 for clarification. The occurrence of recently assimilated C (13C from the pulse) in specific biomarkers was calculated accordingly: firstly, the isotopic enrichment of specific PLFAs was related to the 13C enrichment in the CO2 plants were exposed to during the pulse-labeling by using a two end-member mixing model:

PLFA  Cnew% ð%Þ ¼




d13 CPLFA  d13 CcontrolPLFA 13

 d CcontrolPLFA

.




d13 Cchamber

where PLFA  Cnew% expresses the proportion of recently assimilated 13 CeCO2 that is recovered in individual PLFAs in percent, d13CPLFA is the isotopic value of each PLFA post-labeling, d13CcontrolPLFA is the isotopic value of PLFAs pre-labeling and d13Cchamber is the isotopic value of CO2 measured in the chambers during labeling. The chamber 13C concentration ranged from 21 to 33 atom% (Reinsch and Ambus, 2013). Secondly, the amount of recently assimilated 13C into individual PLFAs per gram soil (PLFA  Cnew) was calculated by multiplying PLFA  Cnew% with the amount of each PLFA in the sample. Finally, the amounts of recently assimilated 13C in each functional group in each sample were converted to a percentage scale (PLFA  Cnew/group):

PLFA  Cnew=group ¼

X

PLFA  Cgroup *100=

X

PLFA  Cnew all

P where PLFA  Cgroup is the sum of PLFA  Cnew for each microbial P functional group and PLFA  Cnew all is the sum of all PLFA  Cnew in each sample. Calculations were carried out for all samples and presented as mean percentages of recently assimilated C that was utilized by each microbial functional group
standard error. 2.7. Statistical analyses Statistical analysis was carried out in R version 12.2.1 (http:// www.r-project.org). When necessary, data were tested for normality (ShapiroeWilk test of normality) and were always tested for homogeneity of variances (Levene test of equality of variances, “car” package) prior to analysis. Differences within climate treatments over time were analyzed using an ANOVA with time as predictor. TukeyHSD test was used to extract differences between times. If data showed lack of variance homogeneity, the Kruskale Wallis test was used instead. Differences between climate treatments were analyzed with a Fit Mixed-Effect Model (lmer) in the “LMERConvenienceFunctions” package using the full factorial statement “CO2  D  T”. The random statement “(1jBlock) þ (1jBlock:CO2) þ (1jBlock:CO2:T) þ (1jBlock:CO2:D)” was used in all analyses and was extended by the factor (1jTime) when datasets were analyzed across times. P-value outputs were generated using the MixMod-package that uses the SAS-algorithm. Principle component analysis (PCA) was performed on PLFA and PLFA-Cnew data from rhizosphere and soil using a covariance matrix in the package “prcomp”. Statistical analysis was performed on PCA scores (climate treatments) using the Fit Mixed-Effect Model described above. Scores for PC1 and PC2 were tested individually. A pronounced spatial variability in vegetation cover across the experimental site (Kongstad et al., 2012) was expected to foster datasets characterized by high variability. Moreover, the change in environmental conditions between the two pulse-labeling campaigns, and successively increased soil water contents (w1.5 vol%) in the non-drought treatments (Fig.1) may have resulted in even higher spatial variability. Due to this expected high influence of variable field conditions, we reported p-values not only at the generally accepted level of significance, p  0.05, but also for trends in the data, p  0.1. 3. Results 3.1. Pulse-labeling conditions Average photosynthetic active radiation (PAR) varied during the experimental period and averaged 217
42 mmol m2 s1 and

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

490
80 mmol m2 s1 on the first and second labeling days, respectively (Fig. 1a). Soil temperatures in 5 cm depth were little influenced by PAR but raised continuously from 9.6
0.1  C to 10.7
0.1  C in plots without temperature treatment (A, CO2, D, DCO2) and from 10.3
0.1  C to 11.0
0.1  C in plots with elevated temperature. Soil water content in drought plots decreased continuously during the experimental period from 11.7
0.7 vol% to 9.1
0.8 vol % (Fig. 1b). In contrast, non-drought plots received precipitation during the experimental period and soil water contents varied. Due to a rain event between the two labeling campaigns, the soil water content in non-drought plots increased from 12.9
0.6 vol% to 15.3
0.6 vol%. Total plant biomass was significantly reduced by drought (data not shown).

higher 13C isotopic enrichment throughout the 8-day period. At day 8, root APE was significantly influenced by the interactions of CO2, D and T in the simulation of future climatic conditions (Table 2i). This was also true for the two-factor interactions TCO2 and TD. A transfer of assimilated 13C into the soil matrix was barely detectable as displayed by the very low APE values in soil because of high background of C (Table 2, soil). The observed peak of 13C enrichment in the soil coincided with the peak in root 13C enrichment. Average APE in soil MBC often peaked 2 days after labeling, coincident with the 13C peak enrichments in roots and soil (Table 2, MBC). The 13C enrichment of MBC was generally not affected by climate treatments, which is in accordance with the observation of unchanged MBC and unchanged total PLFA-C concentrations under different climatic conditions. Treatments with elevated temperature (except the combined TDCO2 treatment) tended to have lower 13C enrichments in MBC than the nontemperature treatments.

3.2. Soil microbial community structure Soil C and N concentrations were similar for all treatments and were on average 2.4
0.6% and 0.15
0.07%, respectively (Table 1). Microbial biomass C (MBC) assessed with the chloroform fumigation extraction method was similar among treatments but TCO2 had a significantly positive effect on MBC. In contrast to MBC, microbial biomass assessed by total PLFA-C concentration extracted from rhizosphere fractions showed the trend towards reduced concentrations under elevated CO2. The PLFA-C concentration in soil was negatively affected by the TD treatment. Carbon concentrations were high in the gram-negative specific PLFAs 17:0cy and 16:1u7c extracted from rhizosphere fractions (Fig. 2). The actinomycetes specific PLFA 10Me17:0 and the fungi specific PLFA 18:3u3 were highly abundant under all climatic conditions. Elevated CO2 tended to reduce PLFA-C concentrations (Fig. 2a) and had a significantly negative effect on the occurrence of the general biomarker 16:0, the gram-negative biomarker 17:0cy, the gram-positive specific PLFA 16:0i and the fungal biomarker 18:2u6,9. In contrast, prolonged summer droughts tended to increase biomarker abundances and had a significantly positive effect on the actinomycetes specific PLFA 10Me18:0 (Fig. 2b). Elevated temperature decreased the abundance of the gram-negative biomarker 19:0cy negatively and the actinomycetes biomarker 10Me18:0, but increased the actinomycetes biomarker 10Me16:0 (Fig. 2c). PLFA-C concentrations in soil samples were not affected by climate treatments (Table S1). 3.3. Concentrations of

13

13

3.4. Recently assimilated C in phospholipid fatty acids The utilization of recently assimilated C (%) describes the relative distribution of recently photo-assimilated C among the different microbial groups, e.g. if 100% of all PLFA related 13C was observed in gram-negative PLFA biomarkers, then it is assumed that all recently photo-assimilated C that ended up in the microbial community was exclusively utilized by the group of gram-negative bacteria. Thus, recently assimilated C observed in each functional group expresses the importance of each microbial functional group in the short-term C turnover. In the rhizosphere fraction, overall utilization of recently assimilated C by gram-negative bacteria was high (Table 2). The sum of 13C observed in the general biomarker PLFAs was on average 20%e40%. The temporal dynamics in C-transfer patterns are indicated by the shading in Table 2, and show that gram-negative bacteria were generally the first group to utilize recently assimilated C succeeded by an uptake of root exudates by gram-positive bacteria. The uptake of 13C in actinomycetes was delayed compared to gram-negative and gram-positive bacteria. The utilization of recently assimilated C by actinomycetes in the rhizosphere was less than 1% one day after labeling (with exception of TCO2) but increased over time. The 13C uptake by actinomycetes and fungi related to the rhizosphere was generally lower compared to their activity associated to recent C uptake in soil at day 2, except for the TCO2 treatment. Rhizosphere associated fungi often took up a high proportion of 13C in the rhizosphere at day 1 (average 10%), whereas the utilization of recently assimilated C by fungi was lower at day 2 (average 4%), combined with a high proportion of 13 C in the fungal group in soil at day 2 (20%). The allocation of recently assimilated C into rhizosphere associated gram-negative bacteria was significantly decreased by the

C in belowground compartments

Root atom% 13C excess (APE) peaked 2 days after the labeling for all treatments, except the single factor CO2 treatment that showed the highest enrichment after 8 days (Table 2, roots). One day after labeling, the 13C excess in roots increased significantly by the CO2 only treatment and interactive effects were found for TCO2 and TD (Table 2f). The CO2 treatment maintained a significantly

Table 1 Soil and microbial properties for each climate treatment: soil moisture, soil C and N concentrations (% of dry matter), microbial biomass C (MBC, mg C g1 dry soil), rhizosphere and soil PLFA-C concentrations (mg C g1 rhizosphere fraction or soil). Effects indicate results from the mixed effect model statistical analysis. Values are means
SE (n ¼ 4e6). Treatment

A

Soil moisture (vol%) Soil C (%) Soil N (%) MBC (mg C g1) Rhizosphere PLFA-Ca (mg C g1) Soil PLFA-Ca (mg C g1)

15.4 2.3 0.14 661 0.5 10.8

CO2







2.3 0.4 0.01 51 0.1 1.7

14.4 2.1 0.14 611 0.4 9.0

D







1.1 0.2 0.01 64 0.1 2.2

11.6 2.6 0.15 570 0.7 11.9

DCO2







1.9 0.2 0.01 62 0.1 1.3

12.7 2.6 0.15 551 0.5 11.8









T 1.8 0.1 0.01 93 0.1 1.9

13.5 2.2 0.14 509 0.5 12.0

TCO2







1.0 0.1 0.003 75 0.1 1.5

13.6 2.3 0.15 753 0.3 10.2









TD 1.0 0.2 0.01 103 0.1 1.6

*p < 0.05; xp < 0.1. a PLFA concentrations as sum of all PLFAs (general, gram-negative, gram-positive, actinomycetes and fungi biomarkers).

8.4 2.2 0.14 465 0.5 9.3

TDCO2







2.1 0.1 0.07 68 0.1 1.7

9.8 2.5 0.15 555 0.3 8.7









Effects 0.9 0.3 0.01 70 0.1 1.7

CO2  T* COx2 D  T*

14

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

Fig. 2. Average concentrations of C in individual PLFAs extracted from rhizosphere fractions (ng PLFA-C mg1) compared across the main treatments. (a) ambient CO2 concentration vs. elevated CO2 concentration, (b) ambient (non-drought) vs. drought treated plots and (c) ambient (non-temperature) vs. increased temperature plots. xindicates trends between ambient and the climate treatment with p  0.1. (n ¼ 16e24).

CO2 treatment at day 2 (p < 0.05). The DCO2 combined treatment (also in TDCO2) displayed a significantly higher utilization of recently assimilated C in rhizosphere associated gram-negative bacteria at day 8 (31
9%, p < 0.05) compared to the single factor treatments CO2 and D (average 17%). The combined treatment TD showed a significantly reduced utilization of recently assimilated C in rhizosphere associated gram-positive bacteria at day 1 (13
5%, p < 0.05) compared to the single factor treatments D and T (average 17%). The utilization of recently assimilated C by actinomycetes was not affected by climate treatments, and no recently assimilated 13C was detected in rhizosphere associated actinomycetes at various days in the A, T, TCO2, TD and TDCO2 treatments. Utilization of recently assimilated C by rhizosphere associated fungi was negatively affected by elevated CO2 exposure (CO2, DCO2, TDCO2) and showed on average less than 5% C allocation. The combination of T and CO2 increased the rate of utilization of recently assimilated C in rhizosphere associated fungi from on average 9% and 1% in the sole treatments to 12
8% in the combined treatment at day 1 (p < 0.05). In contrast, the combination of T and D reduced the utilization of recently assimilated C by fungi at day 1 from an average of 18% in the sole treatments to 3
2% (p < 0.05).

The utilization of recently assimilated C by fungi under elevated CO2 was approximately twice as high in those associated with soil samples compared with those associated to the rhizosphere fraction. An example: recently assimilated C as displayed in fungal biomarkers in the rhizosphere fraction in the CO2 and T treatments at day 2 was low (average 5%) and a corresponding high 13C allocation in soil was observed the same day (average 26%). The full factorial combination TDCO2 showed a different fungal allocation pattern of recently assimilated C than any other treatment since there was no allocation of CO2-derived C into fungal PLFAs at day 1. The utilization of recently assimilated C by gram-negative and gram-positive bacteria in soil was little affected by climate treatments, or interactions. One exception was the treatment with increased temperature only (T) that revealed a significant increase in 13C allocation into gram-negative bacteria PLFAs (19
7%) compared to the other single factor treatments (average 14%). In addition, the utilization of recently assimilated C by gram-positive bacteria was significantly reduced from on average 19% to only 4
2% (p < 0.05) in the two factor treatment with drought and elevated CO2 (DCO2). Principle component analysis for the relative distribution of recently assimilated C in individual PLFAs showed a significant

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

15

Table 2 Effects of climate treatments (a-h) on belowground C pools. Treatment effects (mixed effect model statistical analysis) are presented in (i). Atom% 13C excess (mean values APE  103) for roots, soil and microbial biomass C (MBC) measured at day1, day2 and day8 (D1, D2, D8). Utilization of recently assimilated C (Utilized C) describes the distribution of recently photo-assimilated C in different microbial functional groups (general, gram-negative, gram-positive, actinomycetes, fungi) extracted from rhizosphere fractions at day1, day2 and day8 (R1, R2, R8) and from soil samples at day2 (S2). Letters indicate differences between days within climate treatments, where letter ‘a’ indicates no difference between pre- and post-labeling. Increased C utilization is illustrated by increasing shading intensity based on six group categories (0, 1e10, 11e20, 21e30, 31e40 and 41e50). Standard errors are avoided for clearness, but the full dataset is available as supplementary tables S2 and S3.

APE

Roots Soil MBC

Utilized C (%)

(e) T

General Gram-negative Gram-positive Actinomycetes Fungi

APE

Roots Soil MBC

Utilized C (%)

(g) TD

General Gram-negative Gram-positive Actinomycetes Fungi

D1 5b 0 8b R1 43 31 15 1 9

D2 4b 0 7 bc R2 33 41 11 7 7

D8 5b 0 4c R8 37 31 13 0 19

D1 4 0 12 b R1 24 44 13 0 3

D2 8 0 9b R2 35 29 10 6 3

D8 3 0 5 ab R8 33 24 16 0 10

(i) Effects

D1

D2

Roots Soil MBC

CO2*, TCO2*, TD*

CO2* CO2*, T* T*, TD* R2

TCO2§, TDCO2§ R1

General Gram-negative Gram-positive TD§ Actinomycetes TD§, Fungi CO2§, TCO2§, TD§ * p<0.05; § p<0.1

CO2§

APE Utilized C (%)

D8 2a 0 4c R8 37 22 17 ab 5 19

General Gram-negative Gram-positive Actinomycetes Fungi

(d) DCO2 Roots Soil MBC APE

D2 5b 0.2 10 b R2 32 42 9b 8 8

S2 19 17 10 11 23

Roots Soil MBC

S2 26 14 28 19 12

Utilized C (%)

D1 0.1 a 0 7 bc R1 26 24 19 a 5 27

(b) CO2

General Gram-negative Gram-positive Actinomycetes Fungi

(f) TCO2 Roots Soil MBC APE

General Gram-negative Gram-positive Actinomycetes Fungi

D8 4b 0.2 4 ab R8 20 26 15 12 10 ab

S2 18 19 14 17 31

Utilized C (%)

APE

Roots Soil MBC

Utilized C (%)

(c) D

D2 5b 0.4 19 c R2 31 27 20 4 0b

General Gram-negative Gram-positive Actinomycetes Fungi

(h) TDCO2 Roots Soil MBC APE

General Gram-negative Gram-positive Actinomycetes Fungi

D1 3 ab 0 10 b R1 29 19 10 0 21 a

S2 19 20 17 14 11

Utilized C (%)

Utilized C (%)

APE

(a) Ambient Roots Soil MBC

General Gram-negative Gram-positive Actinomycetes Fungi

D1 7 ab 0.2 12 ab R1 34 22 11 0.03 1

D2 5 ab 0.4 17 b R2 21 14 23 5 3

D8 9b 0.3 5 ab R8 20 12 14 13 9

S2 7 11 10 17 21

D1 2 ab 0.2 15 b R1 17 37 20 5 4 ab

D2 4b 0.1 14 b R2 43 11 26 3 0a

D8 3.2 b 0.2 2b R8 21 31 12 10 9b

S2 23 20 4 21 13

D1 5 ab 0.03 6b R1 28 15 12 16 12

D2 8b 0.1 7b R2 31 29 16 0 6

D8 5 ab 0 6b R8 23 15 27 0 14

S2 12 22 19 7 23

D1 3b 0 13 ab R1 37 31 12 3 0

D2 3b 0 15 b R2 41 26 12 1 4

D8 3 ab 0 7 ab R8 19 36 10 0 15

S2 14 31 8 6 24

D8 CO2§, D*, TCO2*, TD§, TDCO2* CO2§, T*, TCO2§ R8 CO2§ DCO2§

S2 T§ DCO2§

T§

separation of the single factor treatments CO2 and T on PC1 in the rhizosphere fraction at day 2 (Fig. 3, loadings Table S4). The ambient and TDCO2 treatments did not separate significantly on either of the PCA axes. PLFA concentrations in the rhizosphere and the soil are not reported because statistical analyses showed that treatment effects on PLFA distributions were lacking.

4. Discussion 4.1. Climate effects on the utilization of newly photo-assimilated C by different microbial groups The almost immediate recovery of recently assimilated C in belowground compartments (Table 2) reflects the close linkage

16

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

Fig. 3. Principle component analysis (PCA) for recently assimilated C recovered in phospholipid fatty acids (PLFAs) as affected by climate treatments in the rhizosphere at day 2. Statistical analysis on PCA scores was performed to test for treatment effects. PCA loadings are presented in Table S8. A: ambient (non-treated control), CO2: elevated CO2 concentration (þ120 ppm), D: drought treatment, T: temperature treatment and TD, TCO2, DCO2 and TDCO2 are combined treatments.

between aboveground and belowground C allocation (Wu et al., 2009; Brüggemann et al., 2011). Belowground utilization of recently assimilated C showed the general trend of peak enrichments 2 days after labeling (Table 2). This pattern, however, might depend on the speed of C re-allocation under different climatic conditions (see discussion below) and the specific activity of various processes in the short-term C-cycle, such as plant photosynthesis, root exudation and the C use efficiency in the diverse groups of microorganisms. Observed peak enrichments are consistent with other studies showing 13C peaks 1 or 2 days after pulse-labeling (Lu et al., 2004; Leake et al., 2006; Jin and Evans, 2010). In general, the patterns of 13C excess abundances were often similar across climate treatments and suggest rather small effects of climatic conditions on the general short-term C utilization. The microbial community was dominated by gram-negative bacteria (Fig. 2) and fungi (arbuscular mycorrhizal fungi, saprophytic fungi) were underrepresented compared to other grassland studies (Butler et al., 2003; Denef et al., 2007). This underrepresentation of fungi (Fig. 2) could have been due to an antagonistic effect of the highly abundant actinomycetes in soil samples (Jayasinghe and Parkinson, 2008). Furthermore, mycorrhizal fungi biomarkers were not detected with the performed PLFA analysis. This was unexpected as other studies at the field site observed substantial mycorrhizal abundance (Arndal et al., 2013). It might be that the mycorrhiza specific PLFA 16:1u5 was masked by a close adjacent fatty acid and could therefore not be detected. Furthermore, the neutral lipid fatty acid (NLFA) 16:1u5 has been shown to be a better and more reliable biomarker for arbuscular mycorrhizal fungi (Olsson, 1999), but NLFAs were not identified in this study. Elevated CO2 tended to reduce bacterial PLFA-C concentrations (apart from actinomycetes), but did not affect fungal PLFA-C concentrations in the rhizosphere fraction (Fig. 2). The reduced bacterial abundance, combined with a more stable fungal community

is in accordance with hypothesis (i) suggesting that elevated CO2 favors the fungal community. This observation also confirms other studies showing that increased atmospheric CO2 concentrations shifted the microbial community towards a fungal dominance after several years of exposure to elevated CO2 (Carney et al., 2007; Jin and Evans, 2010). Increased temperature in our study had a similar effect as CO2 fumigation on the microbial community, but the effect was smaller. This does not agree with our hypothesis (ii) and is in conflict with results from a 12 year warming experiment in the Harvard Forest showing a shift in the microbial community composition with reduced fungal abundance and increased abundance of grampositive bacteria (Frey et al., 2008). However, in a warming experiment in a tallgrass prairie, the soil fungal community was favored under warming conditions, probably due to an indirect positive effect of warming on plant growth (Zhang et al., 2005). Moreover, Yuste et al. (2011) observed a drought-resistant fungal community in response to warming in a shrubland and holm-oak forest and thus, the observed effect of warming is similar to our results. Extended summer droughts in our study tended to favor the bacterial community by increasing the abundance of the gramnegative bacteria (17:0cy, 19:0cy, Fig. 2b) and decreasing the fungal abundance compared to non-drought plots. A high production of cyclic PLFAs, specific for gram-negative bacteria, can be triggered by drought conditions and facilitate stronger membrane lipids to better tolerate drought periods (Bossio and Scow, 1998). Our results suggest a rejection of hypothesis (iii) that a fungal based community is established under drought conditions to meet low nutrient availability. Hawkes et al. (2011) reported fungal community responses to different precipitation treatments and showed that fungal communities are more stable under drought conditions. However, the free-living gram-negative species Azotobacter sp. is very important in nitrogen cycling due to its ability to fix atmospheric nitrogen (Strandberg and Wilson, 1968) and the gramnegative Nitrosomonas sp. plays a key role in the nitrogen cycle (Hofman and Lees, 1952). Therefore, the maintenance of gramnegative bacteria community under drought conditions can also fulfill nutrient demands. 4.2. Gram-negative bacteria as indicator of short-term carbon turnover Myccorhizal fungi are important organisms for the short-term C cycle (Jakobsen and Rosendahl, 1990; Johnson et al., 2002) due to their strong dependence on plant photosynthetic activity (Moyano et al., 2007). Additionally, gram-negative bacteria have also been shown to be highly responsive to plant rhizodeposition (Paterson et al., 2007; Jin and Evans, 2010). In the present study, gram-negative bacteria was the first group that was detected to utilize recently assimilated C in the studied heathland (Table 2), which is in accordance with previous observations of grassland vegetation (Butler et al., 2003; Treonis et al., 2004). Furthermore, the gram-negative bacteria showed the most consistent patterns of utilization of recently assimilated C across climate treatments disregarding the different temporal lags in short-term C transfer due to climatic conditions. Previous work at the study site has shown that C turnover is faster at elevated CO2 concentration than under ambient or drought conditions (Selsted et al., 2012). And, prolonged drought periods at the study had a negative effect on plant photosynthesis (Albert et al., 2011), plant biomass (Kongstad et al., 2012) and the microbial community biomass size (Andresen et al., 2010b). Therefore, we assume that short-term C cycling under the drought conditions prevailing during the experimental course was slower in the single factor drought treatment (D) than in the other single factor treatments, whereas

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

17

the fastest short-term C turnover occurred in plots exposed to elevated CO2 only. Combining this perception with current observations, we suggest that the short-term C utilization patterns displayed by the gram-negative group can be used to differentiate and describe the effects of climatic manipulations on the rate of shortterm C cycling. Three out of four CO2 treatments with elevated CO2 (CO2, DCO2, TDCO2) showed a high utilization of recently assimilated C by gram-negative bacteria 1 day after the pulse-label suggesting an overall positive effect of elevated CO2 on short-term C transport and utilization by the microbial community. In contrast, three out of four treatments at ambient CO2 concentration (A, D, T) showed a one-day lag prior to the 13C peak in gram-negative bacteria, and this trend supports our hypothesis (iv) of a faster turnover of recently assimilated C under elevated CO2 concentrations, which is also in agreement with other studies (Denef et al., 2007; Drake et al., 2011; Selsted et al., 2012). Drought and elevated temperature treatments did not show consistent patterns of shortterm C utilization by gram-negative bacteria in our study, suggesting that the two environmental variables are not the main determinants of short-term C turnover rates in this heathland. Nevertheless, warming and drought are known to affect plant photosynthesis, and thus C uptake indirectly and can have reducing and accelerating effects on the short-term C turnover depending on plant acclimatization (Ciais et al., 2005; Luo, 2007; Albert et al., 2011).

Fungi are known to transport C through hyphae and provide easily available C to rhizosphere bacteria (Olsson and Johnson, 2005). The re-allocation of recently assimilated C from roots into the soil via hyphae was similar under ambient and future climatic conditions, but the utilization of re-distributed recent C by the bacterial groups changed. In the future treatment, most of the recently assimilated C was allocated to gram-negative bacteria. In contrast, under ambient conditions recently assimilated C was evenly distributed between the four functional groups encountered in this study. The increased bacterial involvement in short-term C cycling (gram-negative and positive) observed in the future treatment also points towards a lower importance of the decomposer community (especially actinomycetes) in terms of short-term C turnover under future climate. Our results suggest that changing climatic conditions affect the soil microbial community and the distribution of recently assimilated C within and between microbial functional groups (Fig. 3). The same conclusion was drawn on the basis of climate change experiments with single factor manipulations (Carney et al., 2007; Frey et al., 2008; Jin and Evans, 2010; Kim et al., 2012) as well as from a climate change experiment manipulating multiple factors in combination, e.g. CO2 concentration, precipitation and temperature in an old-field ecosystem (Gray et al., 2011).

4.3. Consequences of future climatic conditions on short-term carbon cycling

The turnover of recently assimilated C is influenced by climate change factors (Hungate et al., 1997; Ciais et al., 2005; Heimann and Reichstein, 2008; Selsted et al., 2012). In our study, short-term C turnover was stimulated by elevated CO2 resulting in different C utilization patterns of the microbial community over time indicating a potential change in the dynamics of the C cycle under future climatic conditions. Generally, responses of the microbial community to climatic changes are highly variable within and among ecosystems, and also depend on plant community composition (De Deyn et al., 2008). Expressing isotopic values as the utilization of recently assimilated C by the microbial community facilitates the evaluation of the importance of each component of the microbial community in short-term C cycling. Furthermore, timing (Butler et al., 2003) and scale of climatic changes affect soil microbial responses rather than the occurrence of climate change itself (Hawkes et al., 2011; Sheik et al., 2011). This study presented a semi-quantitative assessment on the short-term C turnover in a heathland ecosystem and the role of microbial functional groups in short-term C cycling. Future work is needed to elucidate a more thorough C balance of recently assimilated C to assess whether the observed changes in C utilization by microbial functional groups are quantitatively important on an ecosystem scale.

The applied climate treatments were chosen to simulate climatic conditions for Denmark in 2075 (Mikkelsen et al., 2008) as realized in the full factorial combination TDCO2 (future). We observed that the utilization of recent photosynthetic C in different belowground compartments, particularly microbes, was little affected by future climatic conditions (Table 2a,h), in accordance with our hypothesis (v). Even though the overall patterns in 13C abundance did not differ under ambient and future climatic scenarios, the turnover of recently assimilated C did vary as revealed by different C utilization patterns in the microbial community, in disagreement with our hypothesis (v). The turnover of recently assimilated C as evaluated on the basis of the activity associated with the occurrence of recent C in rhizosphere associated gram-negative bacteria (as described in Section 4.2) was faster under future climatic conditions than under ambient conditions. Generally, under future climatic conditions the gram-negative bacteria persistently utilized 30% of the recently assimilated C and this emphasizes the quantitative importance of this functional group (Table 2h). Assimilation of recent C by the gram-positive bacteria was similar under ambient and future climatic conditions. In contrast, the utilization of recently assimilated C by rhizosphere associated actinomycetes was always lower under future climatic conditions (average 2%) compared to any other single- or two-factor climate treatment. It can be argued that the sampling frequency was too low to pick up 13C enrichment peaks during the fast short-term 13C utilization. Still, this is rather unlikely because 13C enrichment peaks in roots and MBC show the same patterns for all other treatments and therefore our observation rather suggests that the actinomycetes community was suppressed under future climatic conditions. Actinomycetes, like fungi, are decomposers of recalcitrant soil organic matter (Lacey, 1997), and a decreased importance of actinomycetes in the turnover of recently assimilated C in a future soil environment hints towards a higher utilization of older soil organic carbon which may have implications for mineralization of recalcitrant C compounds (Carney et al., 2007; Paterson et al., 2009; Garcia-Pausas and Paterson, 2011).

4.4. Considerations for future studies

Acknowledgments The experiment was carried out within the CLIMAITE project that is financially supported by the Villum Kann Rasmussen Foundation, Air Liquide DenmarkA/S and Dong Energy. Additional support was achieved from the INCREASE network funded by the EC FP7-Infrastructure-2008-1 grant agreement 227628. In particular, we thank Preben Jørgensen, Nina W. Thomson, Svend Danbæk and Poul T. Sørensen for keeping the experiment running constantly. The authors would also like to thank helpers of the pulse-labeling with special thanks to Nina B. Mikkelsen, Aslak K. Hansen and Matthias J. Justesen for their great effort during the whole experimental time, all involved Climaite Ph.D. students and technicians. Special thanks go to Marie P. Merrild for her contribution to the fatty acid protocol, Anja C. Nielsen for her help with

18

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19

the fatty acid analysis and Mette H. Flodgaard for her flexible and fast help with sample analyses. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2013.09.014. References Ainsworth, E.A., Long, S.P., 2005. What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytol. 165, 351e372. Albert, K.R., Mikkelsen, T.N., Michelsen, A., Ro-Poulsen, H., Van der Linden, L., 2011. Interactive effects of drought, elevated CO2 and warming on photosynthetic capacity and photosystem performance in temperate heath plants. J. Plant Physiol. 168, 1550e1561. Amundson, R., 2001. The carbon budget in soils. Ann. Rev. Earth Planet. Sci. 29, 535e562. Andresen, L.C., Michelsen, A., Ambus, P., Beier, C., 2010a. Belowground heathland responses after 2 years of combined warming, elevated CO2 and summer drought. Biogeochemistry 101, 27e42. Andresen, L.C., Michelsen, A., Jonasson, S., Beier, C., Ambus, P., 2009. Glycine uptake in heath plants and soil microbes responds to elevated temperature, CO2 and drought. Acta Oecol. 35, 786e796. Andresen, L.C., Michelsen, A., Jonasson, S., Schmidt, I.K., Mikkelsen, T.N., Ambus, P., Beier, C., 2010b. Plant nutrient mobilization in temperate heathland responds to elevated CO2, temperature and drought. Plant Soil 328, 381e396. Arndal, M.F., Merrild, M.P., Michelsen, A., Schmidt, I.K., Mikkelsen, T.N., Beier, C., 2013. Net root growth and nutrient acquisition in response to predicted climate change in two contrasting heathland species. Plant Soil 369, 615e629. Augé, R.M., 2001. Water relations, drought and vesicular-arbuscular mycorrhizal symbiosis. Mycorrhiza 11, 3e42. Ball, A.S., 1997. Microbial decomposition at elevated CO2 levels: effect of litter quality. Glob. Change Biol. 3, 379e386. Bligh, E.G.D., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911e917. Bossio, D.A., Scow, K.M., 1998. Impacts of carbon and flooding on soil microbial communities: phospholipid fatty acid profiles and substrate utilization patterns. Microb. Ecol. 35, 265e278. Brüggemann, N., Gessler, A., Kayler, Z., Keel, S.G., Badeck, F., Barthel, M., Boeckx, P., Buchmann, N., Brugnoli, E., Esperschütz, J., Gavrichkova, O., Ghashghaie, J., Gomez-Casanovas, N., Keitel, C., Knohl, A., Kuptz, D., Palacio, S., Salmon, Y., Uchida, Y., Bahn, M., 2011. Carbon allocation and carbon isotope fluxes in the plant-soil-atmosphere continuum: a review. Biogeosciences 8, 3457e3489. Butler, J.L., Williams, M.A., Bottomley, P.J., Myrold, D.D., 2003. Microbial community dynamics associated with rhizosphere carbon flow. Appl. Environ. Microbiol.Environ. Microbiol.Appl. Environ. Microbiol. 69, 6793e6800. Carney, K.M., Hungate, B.A., Drake, B.G., Megonigal, J.P., 2007. Altered soil microbial community at elevated CO2 leads to loss of soil carbon. PNAS 104, 4990e4995. Castro, H.F., Classen, A.T., Austin, E.E., Norby, R.J., Schadt, C.W., 2010. Soil microbial community responses to multiple experimental climate change drivers. Appl. Environ. Microbiol. 76, 999e1007. Ciais, P., Reichstein, M., Viovy, N., Granier, A., Ogée, J., Allard, V., Aubinet, M., Buchmann, N., Bernhofer, C., Carrara, A., Chevallier, F., De Noblet, N., Friend, a D., Friedlingstein, P., Grünwald, T., Heinesch, B., Keronen, P., Knohl, A., Krinner, G., Loustau, D., Manca, G., Matteucci, G., Miglietta, F., Ourcival, J.M., Papale, D., Pilegaard, K., Rambal, S., Seufert, G., Soussana, J.F., Sanz, M.J., Schulze, E.D., Vesala, T., Valentini, R., 2005. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437, 529e533. Couteaux, M.-M., Bottner, P., Berg, B., 1995. Litter decomposition, climate and litter quality. Trends Ecol. Evol. 10, 63e66. De Deyn, G.B., Cornelissen, J.H.C., Bardgett, R.D., 2008. Plant functional traits and soil carbon sequestration in contrasting biomes. Ecol. Lett. 11, 516e531. Denef, K., Bubenheim, H., Lenhart, K., Vermeulen, J., Van Cleemput, O., Boeckx, P., 2007. Community shifts and carbon translocation within metabolically-active rhizosphere microorganisms in grasslands under elevated CO2. Biogeosciences 4, 769e779. Drake, J.E., Gallet-Budynek, A., Hofmockel, K.S., Bernhardt, E.S., Billings, S. a, Jackson, R.B., Johnsen, K.S., Lichter, J., McCarthy, H.R., McCormack, M.L., Moore, D.J.P., Oren, R., Palmroth, S., Phillips, R.P., Pippen, J.S., Pritchard, S.G., Treseder, K.K., Schlesinger, W.H., Delucia, E.H., Finzi, A.C., 2011. Increases in the flux of carbon belowground stimulate nitrogen uptake and sustain the longterm enhancement of forest productivity under elevated CO2. Ecol. Lett. 14, 349e357. Drissner, D., Blum, H., Tscherko, D., Kandeler, E., 2007. Nine years of enriched CO2 changes the function and structural diversity of soil microorganisms in a grassland. Eur. J. Soil Sci. 58, 260e269. Dunbar, J., Eichorst, S. a, Gallegos-Graves, L.V., Silva, S., Xie, G., Hengartner, N.W., Evans, R.D., Hungate, B. a, Jackson, R.B., Megonigal, J.P., Schadt, C.W., Vilgalys, R., Zak, D.R., Kuske, C.R., 2012. Common bacterial responses in six ecosystems

exposed to 10 years of elevated atmospheric carbon dioxide. Environ. Microbiol. 14, 1145e1158. Frey, S.D., Drijber, R., Smith, H., Melillo, J., 2008. Microbial biomass, functional capacity, and community structure after 12 years of soil warming. Soil Biol. Biochem. 40, 2904e2907. Frostegård, A., Bååth, E., 1996. The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Biol. Fertil. Soils 22, 59e65. Frostegård, Å., Tunlid, A., Bååth, E., 1991. Microbial biomass measured as total lipid phosphate in soils of different organic content. J. Microbiol. Methods 14, 151e 163. Frostegård, Å., Tunlid, A., Bååth, E., 1993. Phospholipid fatty acid composition, biomass, and activity of microbial communities from two soil types experimentally exposed to different heavy metals. Appl. Environ. Microbiol. 59, 3605e3617. Frostegård, Å., Tunlid, A., Bååth, E., 2011. Use and misuse of PLFA measurements in soils. Soil Biol. Biochem. 43, 1621e1625. Garcia-Pausas, J., Paterson, E., 2011. Microbial community abundance and structure are determinants of soil organic matter mineralisation in the presence of labile carbon. Soil Biol. Biochem. 43, 1705e1713. Gray, S.B., Classen, A.T., Kardol, P., Yermakov, Z., M, M.R., 2011. Multiple climate change factors interact to alter soil microbial community structure in an oldfield ecosystem. Soil Sci. Soc. Am. J. 75, 2217e2226. Hawkes, C.V., Kivlin, S.N., Rocca, J.D., Huguet, V., Thomsen, M. a., Suttle, K.B., 2011. Fungal community responses to precipitation. Glob. Change Biol. 17, 1637e1645. Heimann, M., Reichstein, M., 2008. Terrestrial ecosystem carbon dynamics and climate feedbacks. Nature 451, 289e292. Henry, H.A.L., Cleland, E.E., Field, C.B., Vitousek, P.M., 2005. Interactive effects of elevated CO2, N deposition and climate change on plant litter quality in a California annual grassland. Oecologia 142, 465e473. Hofman, B., Lees, H., 1952. The biochemistry of the nitrifying organisms. Biochem. J. 54, 579e583. Hungate, B.A., Holland, E.A., Jackson, R.B., Chapin III, F.S., Mooney, H.A., Field, C.B., 1997. The fate of carbon in grasslands under carbon dioxide enrichment. Nature 388, 576e579. Jakobsen, I., Rosendahl, L., 1990. Carbon flow into soil and external hyphae from roots of mycorrhizal cucumber plants. New Phytol. 115, 77e83. Jayasinghe, B.A.T.D., Parkinson, D., 2008. Actinomycetes as antagonists of litter decomposer fungi. Appl. Soil Ecol. 38, 109e118. Jensen, K.D., Beier, C., Michelsen, A., Emmett, B.A., 2003. Effects of experimental drought on microbial processes in two temperate heathlands at contrasting water conditions. Appl. Soil Ecol. 24, 165e176. Jin, V.L., Evans, R.D., 2010. Microbial 13C utilization patterns via stable isotope probing of phospholipid biomarkers in Mojave Desert soils exposed to ambient and elevated atmospheric CO2. Glob. Change Biol. 16, 2334e2344. Joergensen, R.G., 1996. The fumigation-extraction method to estimate soil microbial biomass: calibration of the kEC value. Soil Biol. Biochem. 28, 25e31. Johnson, D., Leake, J.R., Ostle, N., Ineson, P., Read, D.J., 2002. In situ 13CO2 pulselabelling of upland grassland demonstrates a rapid pathway of carbon flux from arbuscular mycorrhizal mycelia to the soil. New Phytol. 153, 327e334. Jones, D.L., Nguyen, C., Finlay, R.D., 2009. Carbon flow in the rhizosphere: carbon trading at the soileroot interface. Plant Soil 321, 5e33. Kim, S.-Y., Freeman, C., Fenner, N., Kang, H., 2012. Functional and structural responses of bacterial and methanogen communities to 3-year warming incubation in different depths of peat mire. Appl. Soil Ecol. 57, 23e30. Kongstad, J., Schmidt, I.K., Riis-Nielsen, T., Arndal, M.F., Mikkelsen, T.N., Beier, C., 2012. High resilience in heathland plants to changes in temperature, drought, and CO2 in combination: results from the CLIMAITE experiment. Ecosystems 15, 269e283. Lacey, J., 1997. Actinomycetes in composts. Ann. Agric. Environ. Med. 4, 113e121. Leake, J.R., Ostle, N.J., Rangel-Castro, J.I., Johnson, D., 2006. Carbon fluxes from plants through soil organisms determined by field 13CO2 pulse-labelling in an upland grassland. Appl. Soil Ecol. 33, 152e175. Lu, Y., Murase, J., Watanabe, A., Sugimoto, A., Kimura, M., 2004. Linking microbial community dynamics to rhizosphere carbon flow in a wetland rice soil. FEMS Microbiol. Ecol. 48, 179e186. Luo, Y., 2007. Terrestrial carbonecycle feedback to climate warming. Ann. Rev. Ecol. Evol. Syst. 38, 683e712. Mikkelsen, T.N., Beier, C., Jonasson, S., Holmstrup, M., Schmidt, I.K., Ambus, P., Pilegaard, K., Michelsen, A., Albert, K., Andresen, L.C., Arndal, M.F., Bruun, N., Christensen, S., Danbæk, S., Gundersen, P., Jørgensen, P., Linden, L.G., Kongstad, J., Maraldo, K., Priemé, A., Riis-Nielsen, T., Ro-Poulsen, H., Stevnbak, K., Selsted, M.B., Sørensen, P., Larsen, K.S., Carter, M.S., Ibrom, A., Martinussen, T., Miglietta, F., Sverdrup, H., 2008. Experimental design of multifactor climate change experiments with elevated CO2, warming and drought: the CLIMAITE project. Funct. Ecol. 22, 185e195. Miller, P.A., Smith, B., 2012. Modelling tundra vegetation response to recent arctic warming. Ambio, 281e291. Moyano, F.E., Kutsch, W.L., Schulze, E.-D., 2007. Response of mycorrhizal, rhizosphere and soil basal respiration to temperature and photosynthesis in a barley field. Soil Biol. Biochem. 39, 843e853. Nannipieri, P., Ascher, J., Ceccherini, M.T., Landi, L., Pietramellara, G., Renella, G., 2003. Microbial diversity and soil functions. Eur. J. Soil Sci., 655e670. Olsson, P.A., 1999. Signature fatty acids provide tools for determination of the distribution and interactions of mycorrhizal fungi in soil. FEMS Microbiol. Ecol. 29, 303e310.

S. Reinsch et al. / Soil Biology & Biochemistry 68 (2014) 9e19 Olsson, P.A., Johnson, N.C., 2005. Tracking carbon from the atmosphere to the rhizosphere. Ecol. Lett. 8, 1264e1270. Paterson, E., Gebbing, T., Abel, C., Sim, A., Telfer, G., 2007. Rhizodeposition shapes rhizosphere microbial community structure in organic soil. New Phytol. 173, 600e610. Paterson, E., Midwood, A.J., Millard, P., 2009. Through the eye of the needle: a review of isotope approaches to quantify microbial processes mediating soil carbon balance. New Phytol. 184, 19e33. Paterson, E., Thornton, B., Midwood, A.J., Osborne, S.M., Sim, A., Millard, P., 2008. Atmospheric CO2 enrichment and nutrient additions to planted soil increase mineralisation of soil organic matter, but do not alter microbial utilisation of plant- and soil C-sources. Soil Biol. Biochem. 40, 2434e2440. Reinsch, S., Ambus, P., 2013. In situ 13CO2 pulse-labeling in a temperate heathland e development of a mobile multi-plot field setup. Rapid Commun. Mass Spectr. 27, 1417e1428. Reinsch, S., Ambus, P., Thornton, B., Paterson, E., 2013. Impact of future climatic conditions on the potential for soil organic matter priming. Soil Biol. Biochem. 65, 133e140. Schindlbacher, A., Rodler, A., Kuffner, M., Kitzler, B., Sessitsch, A., ZechmeisterBoltenstern, S., 2011. Experimental warming effects on the microbial community of a temperate mountain forest soil. Soil Biol. Biochem. 43, 1417e1425. Selsted, M.B., Linden, L., Ibrom, A., Michelsen, A., Larsen, K.S., Pedersen, J.K., Mikkelsen, T.N., Pilegaard, K., Beier, C., Ambus, P., 2012. Soil respiration is stimulated by elevated CO2 and reduced by summer drought: three years of measurements in a multifactor ecosystem manipulation experiment in a temperate heathland (CLIMAITE). Glob. Change Biol. 18, 1216e1230. Sheik, C.S., Beasley, W.H., Elshahed, M.S., Zhou, X., Luo, Y., Krumholz, L.R., 2011. Effect of warming and drought on grassland microbial communities. ISME J. 5, 1692e1700. Sowerby, A., Blum, H., Gray, T.R.G., Ball, A.S., 2000. The decomposition of Lolium perenne in soils exposed to elevated CO2: comparisons of mass loss of litter with soil respiration and soil microbial biomass. Soil Biol. Biochem. 32, 1359e1366.

19

Strandberg, G., Wilson, P., 1968. Formation of the nitrogen-fixing enzyme system in Azotobacter vinelandii. Can. J. Microbiol. 14, 25e31. Theis, D.E., Jaeggi, M., Aeschlimann, D., Blum, H., Frossard, E., Siegwolf, R.T.W., 2007. Dynamics of soil organic matter turnover and soil respired CO2 in a temperate grassland labelled with 13C. Eur. J. Soil Sci. 58, 1364e1372. Treonis, A.M., Ostle, N.J., Stott, A.W., Primrose, R., Grayston, S.J., Ineson, P., 2004. Identification of groups of metabolically-active rhizosphere microorganisms by stable isotope probing of PLFAs. Soil Biol. Biochem. 36, 533e537. Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biol. Biochem. 19, 703e707. Vestal, J.R., White, D.C., 1989. Lipid analysis microbial ecology. BioScience 39, 535e 541. White, D.C., Davis, W.M., Nickels, J.S., King, J.D., Bobbie, R.J., 1979. Determination of the sedimentary microbial biomass by extractible lipid phosphate. Oecologia 40, 51e62. Wu, Y., Tan, H., Deng, Y., Wu, J., Xu, X., Wang, Y., Tang, Y., Higashi, T., Cui, X., 2009. Partitioning pattern of carbon flux in a Kobresia grassland on the QinghaiTibetan Plateau revealed by field 13C pulse-labeling. Glob. Change Biol. 16, 2322e2333. Yuste, J., Penuelas, J., Estiarte, M., Garcia-Mas, J., Mattana, S., Ogaya, R., Pujol, M., Sardans, J., 2011. Drought-resistant fungi control soil organic matter decomposition and its response to temperature. Glob. Change Biol. 17, 1475e 1486. Zak, D.R., Ringelberg, D.B., Pregitzer, K.S., Randlett, D.L., White, D.C., Curtis, P.S., 1996. Soil microbial communities beneath Populus grandidentata grown under elevated atmospheric CO2. Ecol. Appl. 6, 257e262. Zelles, L., 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities in soil: a review. Biol. Fertil. Soils 29, 111e129. Zhang, W., Parker, K.M., Luo, Y., Wan, S., Wallace, L.L., Hu, S., 2005. Soil microbial responses to experimental warming and clipping in a tallgrass prairie. Glob. Change Biol. 11, 266e277.