CIAT Research Online - Accepted Manuscript Native arbuscular mycorrhizal fungi increase the abundance of ammonia-oxidizing bacteria, but suppress nitrous oxide emissions shortly after urea application The International Center for Tropical Agriculture (CIAT) believes that open access contributes to its mission of reducing hunger and poverty, and improving human nutrition in the tropics through research aimed at increasing the eco-efficiency of agriculture. CIAT is committed to creating and sharing knowledge and information openly and globally. We do this through collaborative research as well as through the open sharing of our data, tools, and publications. Citation: Teutscherova, Nikola, Vazquez, Eduardo, Arango, Jacobo, Arevalo, Ashly, Benito, Marta, & Pulleman, Mirjam (2018). Native arbuscular mycorrhizal fungi increase the abundance of ammonia-oxidizing bacteria, but suppress nitrous oxide emissions shortly after urea application. Geoderma, 1-9 p. Publisher’s DOI: http://doi.org/10.1016/j.geoderma.2018.09.023 Access through CIAT Research Online: http://hdl.handle.net/10568/97545 Terms: © 2018. CIAT has provided you with this accepted manuscript in line with CIAT’s open access policy and in accordance with the Publisher’s policy on self-archiving. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. You may re-use or share this manuscript as long as you acknowledge the authors by citing the version of the record listed above. You may not change this manuscript in any way or use it commercially. For more information, please contact CIAT Library at CIAT-Library@cgiar.org. 1 Native arbuscular mycorrhizal fungi increase the abundance of 1 ammonia-oxidizing bacteria, but suppress nitrous oxide emissions 2 shortly after urea application 3 4 Nikola Teutscherova 1,2,3*, Eduardo Vazquez 2,3, Jacobo Arango 3, Ashly Arevalo 3, Marta 5 Benito 2, Mirjam Pulleman 3,4 6 7 1 Faculty of Tropical AgriSciences, Czech University of Life Sciences Prague, Czech Republic 8 2 Departament of Agricultural Production, Technical University of Madrid, Madrid, Spain 9 3 International Centre for Tropical Agriculture (CIAT), Palmira, Colombia 10 4 Department of Soil Quality, Soil Biology Group, Wageningen University and Research, 11 Wageningen, The Netherlands 12 13 Abstract 14 The potential of the symbiosis between plants and arbuscular mycorrhizal fungi (AMF) to 15 reduce emissions of the greenhouse gas N2O has gained scientific attention in the last years. 16 Given the high nitrogen (N) requirements of AMF and their role in plant N uptake, they may 17 reduce the availability of mineral N that could be subject to N2O emissions and leaching losses. 18 We investigated the impact of AMF on the growth of tropical grass Brachiaria decumbens 19 Stapf. and on N2O released after fertilization with urea in a mesocosm study. To evaluate the 20 role of nitrification in N2O emissions, we used nitrification inhibitor dicyandiamide (DCD). 21 The study included a full-factorial design (n=6) with two AMF treatments (with and without 22 AMF inoculation) and three fertilization treatments (control, urea and urea+DCD), applied after 23 92 days of growth. Plant growth, soil properties and N2O emissions were measured during the 24 2 following two weeks and the abundance of nitrifiers was quantified one and two weeks after 25 fertilization. The production of N2O increased after urea application but only without DCD, 26 indicating the importance of nitrification in N2O emissions. The emissions of N2O after urea 27 application were reduced by 46% due to the presence of AMF. Nevertheless, the abundance of 28 ammonia-oxidizing bacteria (AOB) was increased by urea and AMF, while plant growth was 29 reduced by the AMF. The increased root:shoot ratio of the biomass in AMF pots suggests 30 competition between AMF and plants. This study demonstrated that immobilization of N by 31 AMF can reduce N2O emissions after fertilization, even when plant growth is reduced. The 32 inverse relationship between (higher) AOB abundance and (lower) nitrification rates suggests 33 that changes in the activity of AOB, rather than abundance, may be indicative of the impact of 34 the AMF-Brachiaria symbiosis on N cycling in tropical grasslands. Alternatively, the 35 difference between N2O emissions from AMF and non-AMF pots may be explained by 36 increased reduction of N2O in the presence of AMF. Longer-term studies are needed to verify 37 whether the effects of AMF on N2O emissions and/or plant growth persist over time or are 38 limited to initial immobilization of N by AMF in N-limited systems. 39 40 Key words 41 arbuscular mycorrhizal fungi; nitrification; nitrous oxide; tropical grasses; urea 42 1. Introduction 43 The productivity of tropical grasslands, which are prone to degradation upon overgrazing, 44 is generally limited by nitrogen (N) availability. Therefore, synthetic fertilizers, most 45 commonly in the form of urea, are applied to meet the plants nutritional requirements. However, 46 in many soils, applied urea is rapidly decomposed and release ammonium (NH4+) which prone 47 to nitrification and linked increased risk of leaching losses. Furthermore, during nitrification 48 and heterotrophic denitrification (Hu et al., 2015), nitrous oxide (N2O) is produced which 49 3 contributes to global warming. In tropical grasslands, the nitrification-related pathway of N2O 50 production is a significant source of N2O emissions from soil (Byrnes et al., 2017). The first 51 and rate-limiting step of nitrification is the oxidation of ammonium (NH4+) to nitrite (NO2-) by 52 ammonia oxidizers (AOs), followed by oxidation of NO2- to nitrate (NO3-) by nitrite oxidizing 53 bacteria, while releasing part of nitrogen as N2O (Davidson, 1991; Firestone and Davidson, 54 1989). If not taken up by plants or soil microorganisms, NO3- is prone to leaching causing 55 eutrophication of ground and surface waters. Furthermore, NO3- is the substrate used by soil 56 denitrifiers further contributing to N2O emissions. Uptake of NH4+ is energetically beneficial 57 when compared to NO3- (Salsac et al., 1987) and is therefore often the preferred form of N for 58 soil microorganisms and several plant species, as observed for example in case of Brachiaria 59 humidicola (Rendle) Schweick (Rao et al., 1996). Immobilization of NH4+ by plants and other 60 soil biota decreases the substrate availability for AOs and thus suppresses N2O emissions from 61 nitrification, and, subsequently, decreases the availability of substrate for denitrifiers mediating 62 N2O emissions from denitrification. The interactions between plants and soil microbes can thus 63 exert a strong control on N2O emissions from soils. 64 Arbuscular mycorrhizal fungi (AMF), which form symbiotic associations with two-thirds 65 of all land plants and are widespread in most terrestrial ecosystems (Smith and Read, 2008), 66 can improve host plant nutrition by enhanced uptake of several soil nutrients, such as 67 phosphorus (P), N and zinc (Zn), via extensive hyphal networks in soil and transfer of nutrients 68 from fungus to plant root in exchange for assimilated carbon (C) (Fellbaum et al., 2012; Smith 69 et al., 2008). By increasing the N uptake of plant, either by N transfer or by reduced limitation 70 of other nutrients and linked increased N uptake by plant itself, AMF symbiosis can potentially 71 play an important role in controlling N availability for N2O production. As the growth of both 72 partners and the efficiency of the symbiosis can be improved only if no other nutrient is limiting, 73 4 the uptake of N by either plant or AMF fungi depends on many interrelated factors with soil 74 nutrient stoichiometry playing a key role (Johnson et al., 2015). 75 Although AMF have been observed to absorb N in various forms including NH4+ (Tanaka 76 and Yano, 2005), NO3- (Cavagnaro et al., 2012) and organic N (Whiteside et al., 2012), their 77 preferred N source seems to be NH4+ (Govindarajulu et al., 2005; Read and Perez-Moreno, 78 2003). It has been suggested that AMF can have an adverse effect on AOs as the latter are 79 considered to be weak competitors for NH4+ (Bollmann et al., 2002). Nevertheless, inconsistent 80 results have been found with respect to the potential competition between AMF and 81 microorganism involved in N-cycling (Amora-Lazcano et al., 1998; Cavagnaro et al., 2007; 82 Veresoglou et al., 2011; Chen et al., 2013; Storer et al., 2017) and the conditions explaining 83 such seemingly conflicting observations remain unclear. 84 While Storer et al. (2017) observed a reduction in N2O emissions after the application of 85 organic material due to the presence of AMF hyphae and hypothesized that AMF outcompeted 86 slow-growing AOs, the vast majority of studies addressing the potential of mycorrhizal 87 symbiosis to reduce N2O emissions from soil focused on the emissions under high water-filled 88 pore space (WFPS) conditions applying NO3- as a substrate for denitrifiers (Bender et al., 2014; 89 Lazcano et al., 2014). Under such conditions of high soil moisture content, mycorrhizal 90 symbiosis was observed to lower the N2O emissions, which the authors speculated to be related 91 to (i) increased N immobilization in microbial and/or plant biomass resulting in reduced pool 92 of available N for denitrifiers, (ii) reduction of C exudation from the roots and increased C 93 release in the hyphosphere, or (iii) changes in soil-water relations due to the improved soil 94 structure and increased water uptake by AMF plants (Bender et al., 2014; Lazcano et al., 2014). 95 Urea is the most frequently used N fertilizer in agriculture (Glibert et al., 2006) and the key 96 N input in pastures where urea is deposited as urine. To our best knowledge, all studies 97 investigating AMF effects on N2O release from grassland soil have focused on temperate 98 5 climates and the majority of studies has considered the denitrification pathway of N2O 99 production using NO3--based fertilizers at high soil water contents, which is not a realistic 100 scenario in tropical grasslands under well drained conditions, as plant water uptake and 101 evapotranspiration rates are very high. No studies have related AMF-induced changes in AOs 102 abundance with N2O released via the nitrification pathway. 103 The aim of the present study was to quantify the effect of AMF in a Brachiaria grassland 104 soil on N2O emissions under aerobic conditions after urea application. To obtain insight into 105 the mechanisms that can explain AMF-induced changes, N uptake by plant and microbial 106 biomass were quantified as well as functional amoA gene involved in NH4+ oxidation. We 107 hypothesized that the presence of AMF reduces N2O emissions after urea application, and that 108 this effect can be attributed to the (i) a negative effect of AMF on the abundance of AOs as a 109 result of competition for NH4+, and (ii) an overall reduction of mineral N in soil due to increased 110 plant N uptake and microbial N immobilization in the presence of AMF. 111 112 2. Materials and Methods 113 2.1. Experimental design 114 A pot experiment (106 days since the plant sowing until the final harvest) with 115 Brachiaria decumbens Stapf., which is one of the most common forage grasses grown the 116 tropics. Only in Brazilian savanna, the estimated surface covered by Brachiaria pastures was 117 50 million of ha (Sano et al., 2002). Many Brachiaria genotypes have the potential to be used 118 to restore degraded grasslands while improving cattle nutrition. The experiment was established 119 in the greenhouse to test the effects of the following factors and their interactions: (i) AMF 120 (with and without AMF) and (ii) fertilization (control without fertilization, urea and 121 urea+dicyandiamide (DCD)). All pots were distributed in a completely randomized block 122 design with eight repetitions per treatment. Fertilizer treatments were applied on day 92 of the 123 6 plant growth and the emissions of N2O and CO2 were determined right before fertilizer 124 treatments application and during 14 days after fertilization (between day 92 and 106 of plant 125 growth) when the N2O emissions dropped to the values close to pre-fertilization values. As 126 synthetic fertilizers are commonly applied after grazing, plant aboveground biomass was cut 127 before the application of the fertilizer treatments. Seven and 14 days after fertilization, four 128 plants were destructively sampled for biomass quantification and N content (Supplementary 129 material Fig. S1). 130 2.2. Soil sterilization and microbial inoculation 131 The study was performed in a controlled greenhouse at the International Centre for Tropical 132 Agriculture (CIAT) in Palmira, Colombia. Plastic pots (17 cm height, 18 cm diameter) were 133 filled with two kilograms of soil (Vertisol) collected from the experimental fields at CIAT. Soil 134 samples were analyzed by the Analytical Services Laborayory at CIAT and contained 152 mg 135 kg-1 of available P (P-BrayII), 1467 mg kg-1 of calcium, 469 mg kg-1 of magnesium, and 628 136 mg kg-1 of available potassium. Soil was obtained from the margins of pastoral field trial at 137 CIAT HQ in Palmira, Colombia, where Brachiaria hybrid (cv. Cayman) is grown. The bulk 138 density of the field is 1.40 g cm-3, which was also the packing density of the soil in the pot 139 experiment. Field moist soil was homogenized and sieved (<5 mm) prior sterilization with 140 autoclave (121° C, 90 min). All pots were re-inoculated with a microbial extract of the fresh 141 soil collected within the same area, which was filtered to exclude AMF spores and prepared as 142 follows: shaking (30 min) of one kilogram of soil with five liters of deionized water, followed 143 by sieving through 125 µm, 40 µm and 20 µm sieves. The extract was then filtered twice 144 through Whatman 2 filter paper. Fifty ml of final filtrate, corresponding to an extract from 5 145 grams for each kilogram of sterile soil, was added to all pots. The moisture content was adjusted 146 to 60% of water-filled pore space (WFPS) and pots were placed in the greenhouse (22-28°C) 147 for two weeks for microorganisms to utilize the substrate released by soil sterilization and to 148 7 colonize the whole pot area. Water filled pore space (WFPS) was calculated by dividing the 149 volumetric water content (calculated as the gravimetric water content * soil bulk density/water 150 density) by total soil porosity, while total soil porosity was calculated according to: soil porosity 151 = 1 – (soil bulk density/2.65) assuming a soil particle density of 2.65 g cm-3.. The soil bulk 152 density in the pots was 1.40 g cm-3. Soil moisture content was controlled gravimetrically and 153 adjusted every 1-2 days. 154 155 2.3. Seeding and inoculation with AMF 156 Two weeks after microbial wash application, seeds of Brachiaria decumbens Stapf. 157 were surface-sterilized with ethanol (50%, 30s) and bleach (2.5%, 5 min) and washed three 158 times with deionized water. Seeds were pre-germinated in sterile Petri dishes for three days and 159 then transplanted to sterilized sand. At the two-leaves stage, the plantlets were transplanted to 160 the pots and on the same day (day one of the experiment) half of the pots was inoculated with 161 commercial AMF inoculum obtained from Abonamos Micorrizas (registration number ICA 162 3556) containing inert substrate, clay, mycorrhizal roots, mycelium and spores. Same amount 163 of sterilized (autoclave, 121°C twice 1h in two consecutive days) inoculum was applied to 164 control non-mycorrhizal pots. The success of the inoculation was checked two weeks later in 165 the additional pots. However, no root colonization or AMF growth was detected. The pots with 166 seedling were re-inoculated with native AMF collected from the area of soil collection on day 167 14 of the experiment. In brief, AMF spores were extracted from one kg of fresh soil using wet-168 sieving and decanting method followed by sucrose centrifugation (Sieverding, 1991). Obtained 169 spores were washed, isolated and applied to the pots at a density of approximately 500 spores 170 pot-1. As the AMF spores originated from the same soil as the microbial inoculum, no surface 171 sterilization of spores was performed. 172 8 A total number of 72 pots was prepared: 24 AMF-inoculated pots (+M), 24 non-AMF 173 (-M) pots and 24 additional pots. The additional pots were used to confirm the soil moisture 174 content and the presence of AMF by destructive sampling throughout the experiment. Plants 175 were watered every one or two days to 60% WFPS. Pots were fertilized twice with a composite 176 fertilizer containing 6 mg N (all applied N in the form of urea), 0.25 mg Mg, 0.25 mg Ca, 5 mg 177 P, 5 mg K, 0.3 mg S and 0.1 mg Zn pot-1 (corresponding to 4.21, 0.18, 0.18, 3.51, 3.51, 0.21 178 and 0.07 kg ha-1, respectively) to prevent limitation by other soil nutrients than N. The small 179 amount of N as urea was added to stimulate the N cycling. The first fertilization was performed 180 on May 17th (day 83 of the plant growth) and the second one on May 26th (day 92 of the plant 181 growth), which was the day of the fertilizer treatment application. 182 183 2.4. Fertilization treatments 184 Two months after AMF inoculation, which corresponds to day 92 after planting, urea 185 solution (0.117 g N pot-1 corresponding to 82 kg ha-1) was applied to eight +M and eight -M 186 pots (-M/N and +M/N, respectively). The recommended amount of N applied to Brachiaria 187 decumbens pastures ranges between 50 and 150 kg N ha-1 (Alvim et al., 1990; De Morais et al., 188 2006). Similarly, to another eight +M pots and eight -M pots, the same amount of urea together 189 with DCD (10% of applied N) were added (-M/DCD and +M/DCD). The remaining 16 pots 190 (eight +M and eight -M) were watered with the same amount of water (-M/Ctr; +M/Ctr). Four 191 repetitions of each treatment were sampled seven days after fertilization (day 99 of plant 192 growth) and another four repetitions at the end of the experiment (14 days after fertilization, 193 day 106 of plant growth) for the soil analysis, plant biomass and N uptake. The application of 194 nitrification inhibitors, such as DCD in order to suppress NH4+ oxidation has been successfully 195 used in pot and field experiments (Tao et al., 2018) and can provide an important insight into 196 the N2O releasing pathways. Nitrification inhibiting substances selectively target AOs by 197 9 deactivating the ammonium monooxygenase enzyme so that NH4+ remains available for AMF, 198 other soil heterotrophs and plants. The reduced mobility of N resulting from suppressed NH4+ 199 oxidation could have implications for the relationship between AMF and the host plant towards 200 higher dependency on the symbiosis. 201 202 2.5. Measurement of greenhouse gases 203 The emissions of N2O and CO2 were measured one day before fertilization treatment 204 application and periodically (1-2 measurements per day) for two weeks since the application of 205 the fertilization treatments using portable Fourier Transform Infrared Spectroscopy (FTIR) Gas 206 Analyzer (Gasmet DX4040, USA). After this period, N2O emissions were stable and close to 207 zero. During each measurement, pots were covered with a non-transparent plastic chamber (2.5 208 l volume), sealed and the gas concentration was determined every 20 seconds for ten minutes 209 in order to obtain a linear regression of N2O (and CO2) concentration with time. 210 211 2.6. Plant growth and nitrogen uptake 212 Plant aboveground biomass was cut twice before the fertilization treatments, and again 213 right before the fertilization treatments were applied (day 92 of plant growth), and nine days 214 later (day 101 of plant growth). All biomass was dried at 70°C until constant weight and 215 weighed. Cut biomass was collected during the whole experiment and pooled with the shoot 216 biomass obtained during destructive sampling of the same pot. Seven and 14 days after 217 fertilization, four pots from each treatment were destructively sampled and plant roots were 218 manually removed from the soil, dried and weighed. The total N content in both shoot and root 219 biomass was determined using Kjeldahl digestion followed by steam distillation (Bremner and 220 Mulvaney, 1982). 221 222 10 2.7. Mycorrhizal parameters 223 Mycorrhizal root colonization was determined in all destructively sampled pots (day 224 seven and day 14 after fertilization) from 100 views as a percentage of colonized root segments 225 of total segments observed. Fresh fine roots were hand-picked from the soils, washed and 226 cleared with 10% KOH for 30 minutes in water-bath at 85°C. Cleared roots were stained during 227 heating (5 min, 80°C) with 5% ink-vinegar solution (Vierheilig et al., 1998). AMF spores were 228 extracted from the soil using wet-sieving and decanting method followed by sucrose 229 centrifugation (Sieverding, 1991) and counted under stereo-microscope. The roots used for 230 mycorrhizal colonization were weighed at the fresh state and a subsample was dried to 231 determine the moisture content. The biomass of roots used for mycorrhizal colonization 232 quantification was summed up to the total root biomass. 233 234 2.8. Soil carbon and nitrogen pools, potential urease activity and nitrification rate 235 In all destructively sampled pots, mineral N was extracted with 1M KCl solution (1:10 236 w/v) and NH4+-N and NO3--N contents were determined colorimetrically using the sodium 237 salicylate method (Forster JC, 1995) and sulphanilamide and N-(-naphthyl) ethylendiamine 238 dihydrochloride method (Miranda et al., 2001), respectively. For the potential urease activity 239 determination, a method proposed by Kandeler and Gerber 1988) and modified by Kandeler et 240 al. (1999) was used and the activity was determined as NH4+ produced during the incubation 241 with urea solution as a substrate. The rest of the soil was stored (not more than 14 days) at 4 242 °C until the analysis of microbial biomass C (MBC) and N (MBN), by fumigating 15 g of fresh 243 soil with ethanol-free chloroform followed by extraction with 0.5M K2SO4 (1:4 w/v) (Vance et 244 al., 1987). The concentration of microbial biomass N was determined by Kjeldahl digestion and 245 steam distillation (Bremner and Mulvaney, 1982) while MBC was determined colorimetrically 246 (578 nm) by quantification of Cr3+ produced by reduction of Cr6+ after microwave digestion 247 11 (Speedwave four, Berghog, Eningen, Germany) at 135°C for 30 min. Microbial biomass C and 248 N were calculated as the difference between the C and N contents in fumigated and non-249 fumigated samples, divided by 0.38 (Joergensen, 1996) and 0.54 (Brookes et al., 1985), 250 respectively. Potential nitrification rate (PNR) was determined with the modified shaken-slurry 251 method (Hart et al., 1994) using 5 g of fresh sieved soil. Duplicates of soil samples were mixed 252 with 50 ml of nitrification potential solution (1 mM potassium phosphate pH 7.2; 0.5 mM 253 ammonium sulphate) and agitated at 200 rpm in orbital shaker. One set of samples was taken 254 immediately after buffer addition while the other set was agitated for 24 hours before the 255 quantification of NO3- using the same method as in soil extracts. 256 257 2.9. DNA extraction and real-time PCR quantification 258 Soil DNA was extracted from 0.25 g of fresh soil from destructively sampled pots using 259 DNeasy PowerSoil DNA isolation kit (QUIAGEN, Hilden, Germany) according to the 260 manufacturer’s instructions. The quantity and purity of the obtained DNA was determined by 261 260/280 nm and 260/230 nm measurements using a Nanodrop spectrophotometer (DeNovix, 262 Wilmington, DE, USA). 263 Quantitative PCR (qPCR) was performed to assess the abundance of the amoA gene of both 264 ammonia oxidizing archaea (AOA) and ammonia oxidizing bacteria (AOB) (Supplementary 265 materia Table S1). The qPCR was performed in 10-μL reaction mixtures containing the 266 following components: 5 μL of iTaq™ Universal SYBR® Green Supermix (Bio-Rad, BioRad 267 Laboratories, Inc., Hercules, CA), 0.5 μM of each primer (Supplementary material Table S1) 268 and 1 μL of diluted DNA extracts. The optimal dilution of DNA extracts was tested to 269 compensate any reaction inhibition by humic acids co-extracted during DNA isolation (data not 270 shown). All qPCR assays were run on an Applied Biosystems ABI 7300 (Applied Biosystems, 271 NJ, USA) sequence detection system starting with the initial denaturation step at 95 ºC for 10 272 12 minutes, followed by amplification cycles specific for each target gene (Supplementary material 273 Table S1). A melting curve analysis was performed after each assay to ensure that only the 274 products of the desired melting temperature were generated. The standard curves for 275 quantifying gene copy numbers were determined by cloning the PCR products in a plasmid 276 using the procedures reported by (Okano et al., 2004). The population sizes of AOA and AOB 277 were estimated as the normalized copies per gram of dry soil. 278 279 2.10. Calculations and statistical analysis 280 The mycorrhizal dependence of selected parameters was calculated according to Hetrick 281 et al. (1992). The plant or soil trait for which the effect size is calculated is the percentage 282 increase of +M respect to the mean of –M treatments. 283 The repeated measurements of GHG were analyzed using SPSS 22.0 program (IBM 284 SPSS, Inc., Chicago, USA) using Linear Mixed Model. The presence of AMF, the type of 285 fertilization (control, urea or urea+DCD) and time of measurement were used as fixed factors, 286 while each pot was considered as a random factor in which time was nested as a repeated 287 measurement. Several models with different covariance structure were carried out and Linear 288 Mixed Model was selected according to the lowest Akaike’s information criterion. When a 289 significant single or interaction effect was detected (p<0.05), the LSD post-hoc tests (p<0.05) 290 were used to test the differences between fertilization treatments. The cumulative GHG 291 emissions were calculated by linear interpolation between measurements. The measured soil 292 properties, AMF parameters, plant biomass and biomass N content were analysed using a 293 Linear Mixed Model with presence of AMF, the type of fertilization (control, urea or 294 urea+DCD) and time of measurement as fixed and the block as a random factor. Normality and 295 homogeneity of the variance were tested using Shapiro-Wilk and Levene’s tests, respectively, 296 13 and when necessary, the values were log-transformed to meet the normality and homogeneity 297 criteria. 298 3. Results 299 The native AMF successfully colonized all inoculated plants with a mean root 300 colonization of 71%, while all the -M pots remained without colonization (Supplementary 301 material Table S2). Similarly, the spore density was high in all +M treatments (3600 spores 100 302 g-1 soil) and negligible in -M pots (2.8 spores 100 g-1 soil). No significant differences in 303 mycorrhizal parameters (root colonization and AMF spore density) were found between the 304 three fertilizer treatments applied (data not shown). Pots were periodically weighted during the 305 experiment in order to detect possible difference in evapotranspiration between mycorrhizal 306 and non-mycorrhizal treatments and no differences were detected. Therefore, the same amount 307 of water was applied to all pots in order to reach the same moisture content. 308 309 3.1. GHG emissions 310 The N2O emission rates during the first 14 days after fertilization were affected by 311 fertilization (F=43.57, p<0.001) and reduced by the inoculation with AMF (F=8.736, P<0.01) 312 (Fig. 1, Supplementary material Table S3). However, the differences between the +M and -M 313 treatments were significant only in case of urea application without DCD, with N2O emissions 314 being 46% lower in the +M than in the -M pots (Fig 1; Supplementary material Table S3). The 315 highest N2O emission rates and the highest differences between -M and +M treatments were 316 observed between 36 and 84 hours after initiation of the fertilization treatments (Fig. 1; 317 Supplementary material Fig. S2 and Table S3). The emissions of N2O were increased by AMF 318 in the control treatment, but decreased in the case of urea addition when compared to –M 319 (Supplementary material Table S3). Unlike the N2O, the CO2 emissions were not affected by 320 the AMF treatment (Fig. 1) and were increased by both urea and urea+DCD application 321 14 (F=40.65, p<0.001).The cumulative N2O and CO2 emissions can be found in Supplementary 322 material Fig. S2. 323 324 3.2. Plant growth and N uptake 325 The inoculation with AMF reduced the shoot biomass but increased the root biomass of 326 the Brachiaria decumbens Stapf. plants (Table 1). Furthermore, the biomass was significantly 327 higher in the treatments fertilized with urea and urea+DCD (Table 1). The positive effect of 328 AMF on root biomass was detected only when plants were fertilized with urea or urea+DCD 329 (Table 1). The N content of plant shoots was significantly lower in the plants colonized by AMF 330 while the N content of plant roots was increased by the AMF root colonization (Table 1). The 331 fertilizer application increased significantly the shoot N content without differences between 332 urea and urea+DCD (LSD, p<0.05). The negative impact of AMF on the shoot N content was 333 only evident when urea or urea+DCD were applied. 334 The total amount of N in the shoot biomass was higher in the -M pots, particularly when 335 plants were fertilized with urea or urea+DCD (Table 1). However, the root N uptake was higher 336 in the +M pots (Table 1). In addition, no significant differences between urea or urea+DCD 337 (LSD, p<0.05) were found neither in the shoot nor in the root N uptake. 338 339 3.3. Soil properties and microbial parameters 340 The presence of AMF did not affect the NH4+ in soil, but reduced the content of NO3- 341 (Table 2). However, the mineral N contents were strongly affected by the type of fertilizer: the 342 highest content of NH4+ was found when urea was applied in combination with DCD while the 343 highest concentration of NO3- was detected when urea was applied without nitrification 344 inhibitor. In both cases, higher contents were found after seven days than after 14 days. 345 Furthermore, reduced amount of NO3- was found in mycorrhizal treatment amended with DCD 346 15 when compared the control pots with DCD (Table 2). Urease activity was only affected by 347 sampling time with higher activity after 14 days than after seven days (Table 2). 348 The MBN was higher in +M than in the -M pots while the MBC was not affected by the 349 presence of AMF (Table 2). However, in urea-amended pots, the MBC was higher in the +M/N 350 than in –M/N. Urea application increased the PNR in the –M/N treatment respect to +M/N (Fig. 351 2, Supplementary material Table S3). 352 3.4. Functional genes abundance 353 The quantification of amoA gene copies revealed a higher abundance of AOA than AOB 354 (5.4x105 number of copies g-1 soil of AOB vs. 2.87x109 number of copies g-1 soil of AOA). The 355 amoA-AOA copy numbers were unaffected by the three studied factors (AMF, fertilization and 356 time) (Supplementary material Table S4), while the abundance of amoA-AOB was significantly 357 increased by the AMF inoculation and by the application of urea (LSD, p<0.05). 358 16 4. Discussion 359 In this study, we demonstrated that the presence of AMF can have a substantial impact 360 on N2O emissions from tropical grasslands at least shortly after fertilization. Although the effect 361 of AMF on N2O emissions has been addressed in earlier studies, the majority of researchers 362 focused on N2O production under rather anaerobic conditions (Bender et al., 2014; Lazcano et 363 al., 2014) while few suggested that the reduced N2O production rates could be related to the 364 out-competition of slow-growing nitrifiers by AMF hyphae (Storer et al., 2017). In this study, 365 we evaluated the impact of native AMF of tropical grass on N2O emissions after the application 366 of urea. We hypothesized that the inoculation of B. decumbens by AMF would reduce the N2O 367 release from soil due to reduced amounts of NH4+ directly available to AOs and consequent 368 reduced growth of both AOA and AOB. 369 370 4.1. Nitrous oxide production pathways 371 The majority of N2O is released during nitrification and denitrification processes 372 (Butterbach-Bahl et al., 2013). While denitrification is the dominant N2O-producing process 373 under oxygen-limited condition (i.e. high moisture content), nitrification can be highly relevant 374 under aerobic conditions (Dobbie et al., 1999). Furthermore, two nitrification-related pathways 375 could be responsible for N2O emissions: (i) the ammonia oxidation with the importance 376 increasing with raising O2 concentrations, and (ii) nitrifier denitrification taking place under 377 lower O2 concentrations (Zhu et al., 2013). Nevertheless, the N2O production in soil is subject 378 to fluctuations and spatial variability and both processes likely occur simultaneously. Under 379 some circumstances, both nitrification-related pathways can account for important amounts of 380 produced N2O. For example, nitrification produced between 83 and 95% of total released N2O 381 in soils ranging between 45 and 50% of WFPS (Huang et al., 2014) while nitrifier denitrification 382 17 was responsible for 34-50% of total N2O at lower O2 concentrations (Zhu et al., 2013), despite 383 being not considered a strictly anaerobic process (Shaw et al., 2006). 384 The application of urea had the strongest effect on N2O production rates whereas the 385 N2O production after urea+DCD application remained comparable to N2O production rates in 386 control soil without fertilization. Furthermore, in the urea-amended pots, the AMF strongly 387 suppressed the cumulative N2O emissions (by 46%). Several studies have demonstrated that 388 AMF interact with soil biota and can influence the N2O production in the soil (Bender et al., 389 2014; Lazcano et al., 2014; Storer et al., 2017) and reduce leaching losses (Martínez-García et 390 al., 2017). Under elevated water content, N2O emission rates were reduced by AMF (when 391 compared to non-mycorrhizal plants) in the study of Bender et al. (2014) and of Lazcano et al. 392 (2014), which the authors related to the reduced abundance of denitrifiers and denitrification 393 rates, and to higher water uptake of AMF plants, respectively. On the other hand, Storer et al. 394 (2017) did not observe any effect of NO3- application on N2O production and strong effect of 395 NH4+, indicating that N2O emissions were released during nitrification, rather than 396 denitrification. In the present study, we observed increased emissions after fertilization with 397 urea and no increase of N2O production when urea was applied together with DCD, indicating 398 that DCD-suppressed nitrification was the direct reason of lower N2O emissions, or that lack of 399 NO3- in soil solution prevented N2O release by soil denitrifiers. Storer et al. (2017) suggested 400 that the reduction of N2O emissions is linked to the reduced abundance of AOs resulting from 401 the superiority of AMF in NH4+ uptake and low competitive capacity of AOB (Verhagen et al., 402 1995). 403 4.2. The abundance of ammonium oxidizers 404 Contrary to our hypothesis, the number of amoA-AOB gene copies increased in the 405 presence of AMF after urea application (Table 2) while no change was observed in case of 406 AOA, which, however, outnumbered the abundance of AOB by three orders. Similar positive 407 18 effect of AMF on AOs has been observed also by Amora-Lazcano et al. (1998). Although AOA 408 and AOB share the amoA gene responsible for oxidation of ammonia, it remains unclear 409 whether AOA share the genes required for nitrifier denitrification as in case of AOB 410 (Stieglmeier et al., 2014). Furthermore, it should be pointed out that soil N transformation 411 processes can occur simultaneously and functionally different microbial groups can share by-412 products (Hu et al., 2015), which makes separation between N2O-forming pathways 413 challenging. Thus, the increased AOB abundance could lead to increased nitrifier denitrification 414 followed by N2O reduction by heterotrophic microbes in anaerobic microsites releasing N2 415 rather than N2O. Such situation may occur inside of soil aggregates with high microbial activity 416 resulting from input of high amount of easily decomposable C originated from hyphae 417 exudation or decomposing senescence AMF hyphae. A similar mechanism has also been 418 proposed by Storer et al. (2017) suggesting that reduced N2O emissions are not necessarily 419 reflected in reduced abundance of AOs or PNR activity, but could be caused by increased N2O 420 consumption as indicated by Domeignoz-Horta et al. (2017). Furthermore, the variations of the 421 N2O emissions has been identified to be dependent on both the activity and diversity of one 422 clade of nosZ (nosZII) encoding the nitrous oxide reductase, the only known N2O consuming 423 mechanism (Domeignoz-Horta et al., 2017). Thus, especially in a short-term experiment such 424 as the present one, the total abundance of genes is less relevant in N2O production when 425 compared to the gene expression and the abundance of active nitrifying and denitrifying 426 populations. 427 4.3. Soil moisture content and nitrification rate 428 Mycorrhizal hyphae are well known to play a pivotal role in stabilization of soil aggregates 429 (Rillig, 2004) which in turn affects soil-water relations. AMF thin and dense mycelium can 430 penetrate to smaller soil pores which can substantially increase the water uptake of the host 431 plants which can reduce anaerobic conditions (Ruiz-Lozano and Azcon, 1995). Furthermore, 432 19 in case of positive mycorrhizal effect, plant water uptake can be increased due to higher biomass 433 production. Mycorrhizal effect on water removal from the pots and increasing availability of 434 O2 was suggested by Bender et al. (2014). Also in the study of Lazcano et al. (2014) the 435 emissions of N2O were decreased by AMF and this drop seemed to be more related to increased 436 use of water than improved N uptake, as AMF plants showed higher photosynthesis and 437 stomatal conductance compared to non-mycorrhizal tomatoes. The changes in soil moisture 438 have substantial effect on greenhouse gases emissions especially at high soil-water content and 439 low O2 concentrations (Hu et al., 2015). Nevertheless, we did not observe any difference 440 between the moisture of +M and -M pots, possibly due to lower biomass production of AMF-441 infected plants (Supplementary material Table S5). 442 443 4.4. Short-term plant-microbe competition and nitrogen immobilization 444 Nitrification, and, consequently, the N2O production, depend not only on the abundance 445 and activity of AOs, but also on the supply of NH4+ which can be reduced as a result of the 446 AMF and plant N uptake. The role of AMF in plant nutrition has been repeatedly demonstrated 447 (Hodge et al., 2010; Hodge and Storer, 2014) and seems to be dependent on the amount of soil 448 N content. Thus, the presence of the extensive extraradical mycelium of AMF can reduce N2O 449 emissions by direct immobilization of N within AMF biomass as well as by improved plant N 450 nutrition resulting from N transfer from fungus to the host root. The increased content of MBN 451 in AMF pots and its role in N2O emissions reduction indicate that short-term immobilization 452 could be the key mechanism of the suppression of N2O production. Most of the plant 453 growth/nutrient parameters were negatively affected by AMF presence, suggesting that reduced 454 N2O emissions could be caused by N immobilization in the AMF mycelium rather than in plant 455 biomass, especially during the first week after the fertilization, when the majority of N2O was 456 released. The competition-related stress could also explain the plant biomass partitioning 457 20 towards increased production of roots when compared to aboveground biomass. Furthermore, 458 the percentage of the root colonization with mycorrhizal arbuscules, which are the exchange 459 sites between the host plants and AMF (Gianinazzi et al., 1979) and indicators of actively 460 functioning mycorrhizae, was rather low accounting only for 8.5% of the root area, while the 461 percentage of root colonization by AMF vesicles was 29.7% (Supporting Information Table 462 S2). Vesicles serve as storage organs where AMF accumulate lipids and glycolipids, which 463 could be considered beneficial only to the fungal partner. This high occurrence of vesicles and 464 rather low abundance of arbuscules within the host plant roots can indicate unidirectional 465 benefits, confirming the stress of plants resulting from the competition for N. Although the 466 ability of both plant and fungal partner to up- or down-regulate the intensity of the symbiosis 467 has been observed in vitro (Kiers et al., 2011), our understanding of such regulation under 468 realistic field conditions remains limited. Nevertheless, plants may have gained other benefits 469 from the symbiosis even at the cost of slightly reduced biomass production. Long-term studies 470 are required in order to understand the outcome of symbiosis and the implications of the benefits 471 gained by both partners in the long-run for the plant production, N use efficiency and N2O 472 emissions, as short-term immobilization could result in enhanced N2O emissions after the 473 senescence of the hyphae. 474 The allometric biomass allocation has been observed in several studies to be affected by 475 AMF infection, especially in non-stressed plants grown from seeds (Veresoglou et al., 2012) 476 where increased allocation to shoots can be viewed as a sign of improved nutrition resulting 477 from improved nutrient uptake by AMF mycelium. In our case, on the contrary, only root 478 biomass production was increased by AMF infection, suggesting that plants were submitted to 479 stress originating from increased N competition after N amendment, regardless the DCD 480 application. On the other hand, a positive AMF effect (difference in plant N uptake by AMF-481 plants compared to non-AMF plants) could be expected when AMF are C-limited and plant is 482 21 N- (or other nutrient) limited (Corrêa et al., 2015). Thus, after N application to N-limited soils, 483 rapid uptake of N by AMF and growth of mycelium could be expected, but only until C-484 limitation of fungi. We did not observed any effect of on AMF on root biomass in control 485 treatments (without N fertilization) probably because AMF growth remained N limited and did 486 not induce physiological changes in plants. 487 Nevertheless, besides microbial N immobilization, the increased root N uptake seemed 488 to be an important driver of N2O production mitigation. Nevertheless, it has been observed that 489 while plant roots are more successful in acquisition of mineral N in the long-term, soil 490 microorganisms, including AMF, often outcompete plants in the short-term (Kuzyakov and Xu, 491 2013). Thus, the N immobilization by microbes and AMF may be temporal and plant roots may 492 gain the advantage in the long-run as they compete for the same N each time the microbes and 493 fungi die. Furthermore, the potential improvement of plant nutrition by AMF in the long-term 494 may result in enhanced N uptake by plants and increased biomass production which can 495 contribute to the reduction of N2O emissions. 496 5. Conclusions 497 We investigated the interactive effect of native arbuscular mycorrhizal fungi and urea 498 application on N2O emissions, plant growth and the abundance of AOs. Furthermore, using 499 DCD nitrification inhibitor we could identify nitrification-related pathway as the source of N2O 500 emissions in this experiment. The production of N2O was increased by urea application without 501 DCD and AM pots released only 54% of the N2O produced in non-AMF pots over a period of 502 two weeks. The negative plant growth response to AMF presence indicated the competition 503 between plants and AMF, which is further confirmed by higher MBN content in AMF pots. 504 Nevertheless, the abundance of amoA gene of AOB was higher in mycorrhizal pots when 505 compared to control. Such a short-term immobilization of N in the AMF mycelium and other 506 soil biota can clearly reduce the N supply available for nitrification and subsequent N2O 507 22 production. Nevertheless, this N will likely be released after the senescence of the AMF hyphae 508 and become remobilized and potentially taken up by plants or used by soil microorganisms. 509 Considering the increase of abundance of AOB in AMF-inoculated treatments, the possibility 510 of increased N2O emissions and the possible changes in competitive ability of mycorrhizal 511 plants in later stages after fertilization or after repeated N addition deserve attention in future 512 experiments. 513 514 Acknowledgements 515 This study was undertaken as part of the LivestockPlus project funded by CGIAR 516 Research Program (CRP) on Climate Change, Agriculture and Food Security (CCAFS), which 517 is a strategic partnership of CGIAR and Future Earth. In addition, this work was also done as 518 part of the Livestock CRP. We thank all donors that globally support the work of the program 519 through their contributions to the CGIAR system. Nikola Teutscherova thanks Cátedra Rafael 520 Dal-Re/TRAGSA for their support and Eduardo Vázquez thanks the Spanish Ministry of 521 Education for his FPU fellowship. 522 523 Conflict of interest 524 No conflict of interest has been declared. 525 526 References 527 Alvim, M.., Botrl, M. de A., Verneque, R. da S., Salvati, J.A., 1990. Aplicacao de nitrogenio 528 em acessos de Brachiaria. 1. Efeito sobre a producao de materia seca. Pasturas Trop. 12, 529 2–6. 530 Amora-Lazcano, E., Vázquez, M.M., Azcón, R., 1998. Response of nitrogen-transforming 531 microorganisms to arbuscular mycorrhizal fungi. Biol. Fertil. Soils 27, 65–70. 532 23 doi:10.1007/s003740050401 533 Bender, S.F., Plantenga, F., Neftel, A., Jocher, M., Oberholzer, H.-R., Köhl, L., Giles, M., 534 Daniell, T.J., van der Heijden, M.G., 2014. Symbiotic relationships between soil fungi 535 and plants reduce N2O emissions from soil. ISME J. 8, 1336–1345. 536 doi:10.1038/ismej.2013.224 537 Bollmann, A., Bär-Gilissen, M.J., Laanbroek, H.J., 2002. Growth at low ammonium 538 concentrations and starvation response as potential factors involved in niche 539 differentiation among ammonia-oxidizing bacteria. Appl. Environ. Microbiol. 68, 4751–540 4757. doi:10.1128/AEM.68.10.4751-4757.2002 541 Bremner, J.M., Mulvaney, C.S., 1982. Nitrogen-Total, in: Page, A.L., Miller, R, H., D.R., K. 542 (Eds.), Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties 543 (Agronomy Series n 9) ASA. SSSA Madison, Wisconsin, pp. 595–624. 544 Brookes, P.C., Landman, A., Pruden, G., Jenkinson, D.S., 1985. Chloroform fumigation and 545 the release of soil nitrogen: A rapid direct extraction method to measure microbial 546 biomass nitrogen in soil. Soil Biol. Biochem. 17, 837–842. doi:10.1016/0038-547 0717(85)90144-0 548 Butterbach-Bahl, K., Baggs, E.M., Dannenmann, M., Kiese, R., Zechmeister-Boltenstern, S., 549 2013. Nitrous oxide emissions from soils: how well do we understand the processes and 550 their controls? Philos. Trans. R. Soc. Lond. B. Biol. Sci. 368, 20130122. 551 doi:10.1098/rstb.2013.0122 552 Byrnes, R.C., Nùñez, J., Arenas, L., Rao, I., Trujillo, C., Alvarez, C., Arango, J., Rasche, F., 553 Chirinda, N., 2017. Biological nitrification inhibition by Brachiaria grasses mitigates soil 554 nitrous oxide emissions from bovine urine patches. Soil Biol. Biochem. 107, 156–163. 555 doi:10.1016/j.soilbio.2016.12.029 556 Cavagnaro, T.R., Barrios-Masias, F.H., Jackson, L.E., 2012. Arbuscular mycorrhizas and their 557 24 role in plant growth, nitrogen interception and soil gas efflux in an organic production 558 system. Plant Soil 353, 181–194. doi:10.1007/s11104-011-1021-6 559 Corrêa, A., Cruz, C., Ferrol, N., 2015. Nitrogen and carbon/nitrogen dynamics in arbuscular 560 mycorrhiza: the great unknown. Mycorrhiza 25, 499–515. doi:10.1007/s00572-015-561 0627-6 562 Davidson, E.A., 1991. Fluxes of nitrous oxide and nitric oxide from terrestrial ecosystems., 563 in: Microbial Production and Consumption of Greenhouse Gases: Methane, Nitrogen 564 Oxides, and Halomethanes. pp. 219–235. 565 De Morais, R.V., Da Fonseca, D.M., Do Nascimento, D., Ribeiro, J.I., Fagundes, J.L., 566 Moreira, L.D.M., Mistura, C., Martuscello, J.A., 2006. Demografia de perfilhos basilares 567 em pastagem de Brachiaria decumbens adubada com nitrogênio. Rev. Bras. Zootec. 35, 568 380–388. doi:10.1590/S1516-35982006000200007 569 Dobbie, K.E., McTaggart, I.P., Smith, K.A., 1999. Nitrous oxide emissions from intensive 570 agricultural systems: Variations between crops and seasons, key driving variables, and 571 mean emission factors. J. Geophys. Res. Atmos. 104, 26891–26899. 572 doi:10.1029/1999JD900378 573 Domeignoz-Horta, L.A., Philippot, L., Peyrard, C., Bru, D., Breuil, M.-C., Bizouard, F., 574 Justes, E., Mary, B., Léonard, J., Spor, A., 2017. Peaks of in situ N2O emissions are 575 influenced by N2O-producing and reducing microbial communities across arable soils. 576 Glob. Chang. Biol. 24, 360–370. doi:10.1111/gcb.13853 577 Fellbaum, C.R., Gachomo, E.W., Beesetty, Y., Choudhari, S., Strahan, G.D., Pfeffer, P.E., 578 Kiers, E.T., Bucking, H., 2012. Carbon availability triggers fungal nitrogen uptake and 579 transport in arbuscular mycorrhizal symbiosis. Proc. Natl. Acad. Sci. 109, 2666–2671. 580 doi:10.1073/pnas.1118650109 581 Firestone, M.K., Davidson, E., 1989. Microbiological basis of NO and N2O production and 582 25 consumption in soil., in: Exchange of Trace Gases Between Terrestrial Ecosystems and 583 the Atmosphere. pp. 7–21. doi:10.1017/CBO9781107415324.004 584 Forster JC, 1995. Soil Nitrogen, in: Alef K, N.P. (Ed.), Methods in Applied Soil Microbiology 585 and Biochemistry. pp. 79–87. 586 Gianinazzi, S., Gianinazzi‐ Pearson, V., Dexheimer, J., 1979. Enzymatic studies on the 587 metabolism of vesicular‐ arbuscular mycorrhiza. III. Ultrastructural localization of acid 588 and alkaline phosphatase in onion roots infected by Glomus mosseae (NICOL. & 589 GERD.). New Phytol. 82, 127–132. doi:10.1111/j.1469-8137.1979.tb07566.x 590 Glibert, P.M., Harrison, J., Heil, C., Seitzinger, S., 2006. Escalating worldwide use of urea - A 591 global change contributing to coastal eutrophication. Biogeochemistry. 592 doi:10.1007/s10533-005-3070-5 593 Govindarajulu, M., Pfeffer, P.E., Jin, H., Abubaker, J., Douds, D.D., Allen, J.W., Bücking, 594 H., Lammers, P.J., Shachar-Hill, Y., 2005. Nitrogen transfer in the arbuscular 595 mycorrhizal symbiosis. Nature 435, 819–823. doi:10.1038/nature03610 596 Hart, S.C., Stark, J.M., Davidson, E.A., Firestone, M.K., 1994. Nitrogen mineralisation, 597 immobilisation and nitrification, in: MEthods of Soil Analysis: Microbiological and 598 Biochemical Properties, Part 2. ASA and SSSA, Madison, pp. 985–1013. 599 Hetrick, B., Wilson, G., Cox, T., 1992. Mycorrhizal dependece of modern wheat varieties, 600 landraces, and ancestors. Can. J. Bot. 70, 2032–2040. doi:10.1139/b92-253 601 Hodge, A., Helgason, T., Fitter, A.H., 2010. Nutritional ecology of arbuscular mycorrhizal 602 fungi. Fungal Ecol. 3, 267–273. doi:10.1016/j.funeco.2010.02.002 603 Hodge, A., Storer, K., 2014. Arbuscular mycorrhiza and nitrogen: Implications for individual 604 plants through to ecosystems. Plant Soil 386, 1–19. doi:10.1007/s11104-014-2162-1 605 Hu, H.W., Chen, D., He, J.Z., 2015. Microbial regulation of terrestrial nitrous oxide 606 formation: Understanding the biological pathways for prediction of emission rates. 607 26 FEMS Microbiol. Rev. doi:10.1093/femsre/fuv021 608 Huang, T., Gao, B., Hu, X.K., Lu, X., Well, R., Christie, P., Bakken, L.R., Ju, X.T., 2014. 609 Ammonia-oxidation as an engine to generate nitrous oxide in an intensively managed 610 calcareous Fluvo-aquic soil. Sci. Rep. 4. doi:10.1038/srep03950 611 Joergensen, R.G., 1996. The fumigation-extraction method to estimate soil microbial biomass: 612 Calibration of the kEC value. Soil Biol. Biochem. 28, 25–31. doi:10.1016/0038-613 0717(95)00102-6 614 Johnson, N.C., Wilson, G.W.T., Wilson, J.A., Miller, R.M., Bowker, M.A., 2015. 615 Mycorrhizal phenotypes and the Law of the Minimum. New Phytol. 205, 1473–1484. 616 doi:10.1111/nph.13172 617 Kandeler, E., Gerber, H., 1988. Short-term assay of soil urease activity using colorimetric 618 determination of ammonium. Biol. Fertil. Soils 6. doi:10.1007/BF00257924 619 Kandeler, E., Stemmer, M., Klimanek, E.-M., 1999. Response of soil microbial biomass, 620 urease and xylanase within particle size fractions to long-term soil management. Soil 621 Biol. Biochem. 31, 261–273. doi:10.1016/S0038-0717(98)00115-1 622 Kiers, E.T., Duhamel, M., Beesetty, Y., Mensah, J.A., Franken, O., Verbruggen, E., Fellbaum, 623 C.R., Kowalchuk, G.A., Hart, M.M., Bago, A., Palmer, T.M., West, S.A., 624 Vandenkoornhuyse, P., Jansa, J., Bucking, H., 2011. Reciprocal Rewards Stabilize 625 Cooperation in the Mycorrhizal Symbiosis. Science (80-. ). 333, 880–882. 626 doi:10.1126/science.1208473 627 Kuzyakov, Y., Xu, X., 2013. Competition between roots and microorganisms for nitrogen: 628 Mechanisms and ecological relevance. New Phytol. doi:10.1111/nph.12235 629 Lazcano, C., Barrios-Masias, F.H., Jackson, L.E., 2014. Arbuscular mycorrhizal effects on 630 plant water relations and soil greenhouse gas emissions under changing moisture 631 regimes. Soil Biol. Biochem. 74, 184–192. doi:10.1016/j.soilbio.2014.03.010 632 27 Martínez-García, L.B., De Deyn, G.B., Pugnaire, F.I., Kothamasi, D., van der Heijden, 633 M.G.A., 2017. Symbiotic soil fungi enhance ecosystem resilience to climate change. 634 Glob. Chang. Biol. 23, 5228–5236. doi:10.1111/gcb.13785 635 Miranda, K.M., Espey, M.G., Wink, D.A., 2001. A Rapid, Simple Spectrophotometric 636 Method for Simultaneous Detection of Nitrate and Nitrite. Nitric Oxide 5, 62–71. 637 doi:10.1006/niox.2000.0319 638 Okano, Y., Hristova, K.R., Leutenegger, C.M., Jackson, L.E., Denison, R.F., Gebreyesus, B., 639 Lebauer, D., Scow, K.M., 2004. Application of Real-Time PCR To Study Effects of 640 Ammonium on Population Size of Ammonia-Oxidizing Bacteria in Soil. Appl. Environ. 641 Microbiol. 70, 1008–1016. doi:10.1128/AEM.70.2.1008-1016.2004 642 Rao, I.M., Kerridge, P.C., Macedo, M.C.M., 1996. Nutritional requirements of Brachiaria and 643 adaptation to acid soils. 644 Read, D.J., Perez-Moreno, J., 2003. Mycorrhizas and nutrient cycling in ecosystems - a 645 journey towards relevance? New Phytol. 157, 475–492. doi:10.1046/j.1469-646 8137.2003.00704.x 647 Rillig, M.C., 2004. Arbuscular mycorrhizae, glomalin, and soil aggregation. Can. J. Soil Sci. 648 84, 355–363. doi:10.4141/S04-003 649 Ruiz-Lozano, J.M., Azcon, R., 1995. Hyphal contribution to water uptake in mycorrhizal 650 plants as affected by the fungal species and water status. Physiol. Plant. 95, 472–478. 651 doi:10.1111/j.1399-3054.1995.tb00865.x 652 Salsac, L., Chaillou, S., Morot-Gaudry, J.-F., Lesaint, C., Jolivet, E., 1987. Nitrate and 653 ammonium nutrition in plants. Plant Physiol. Biochem. 25, 805–812. 654 Sano, E.E., Barcellos, A.O., Bezerra, H.S., 2002. Assessing the spatial distribution of 655 cultivated pastures in the Brazilian savanna. Pasturas Trop. 656 Shaw, L.J., Nicol, G.W., Smith, Z., Fear, J., Prosser, J.I., Baggs, E.M., 2006. Nitrosospira spp. 657 28 can produce nitrous oxide via a nitrifier denitrification pathway. Environ. Microbiol. 8, 658 214–222. doi:10.1111/j.1462-2920.2005.00882.x 659 Sieverding, E., 1991. Vesicular-arbuscular Mycorrhiza Management in Tropical Agrosystems. 660 GTZ, Eschborn, Germany. 661 Smith, S.E., Read, D., 2008. Mycorrhizal symbiosis, Third. ed. Academic Press, Elsevier, 662 Cambridge. 663 Smith, S.E., Read, D., Smith, S.E., Read, D., 2008. 17 – Mycorrhizas in agriculture, 664 horticulture and forestry, in: Mycorrhizal Symbiosis. p. 611–XVIII. doi:10.1016/B978-665 012370526-6.50019-2 666 Stieglmeier, M., Mooshammer, M., Kitzler, B., Wanek, W., Zechmeister-Boltenstern, S., 667 Richter, A., Schleper, C., 2014. Aerobic nitrous oxide production through N-nitrosating 668 hybrid formation in ammonia-oxidizing archaea. ISME J. 8, 1135–1146. 669 doi:10.1038/ismej.2013.220 670 Storer, K., Coggan, A., Ineson, P., Hodge, A., 2017. Arbuscular mycorrhizal fungi reduce 671 nitrous oxide emissions from N 2 O hotspots. New Phytol. doi:10.1111/nph.14931 672 Tanaka, Y., Yano, K., 2005. Nitrogen delivery to maize via mycorrhizal hyphae depends on 673 the form of N supplied. Plant, Cell Environ. 28, 1247–1254. doi:10.1111/j.1365-674 3040.2005.01360.x 675 Tao, R., Li, J., Guan, Y., Liang, Y., Hu, B., Lv, J., Chu, G., 2018. Effects of urease and 676 nitrification inhibitors on the soil mineral nitrogen dynamics and nitrous oxide (N2O) 677 emissions on calcareous soil. Environ. Sci. Pollut. Res. 1–10. doi:10.1007/s11356-018-678 1226-9 679 Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil 680 microbial biomass C. Soil Biol. Biochem. 19, 703–707. doi:10.1016/0038-681 0717(87)90052-6 682 29 Veresoglou, S.D., Menexes, G., Rillig, M.C., 2012. Do arbuscular mycorrhizal fungi affect 683 the allometric partition of host plant biomass to shoots and roots? A meta-analysis of 684 studies from 1990 to 2010. Mycorrhiza 22, 227–235. doi:10.1007/s00572-011-0398-7 685 Veresoglou, S.D., Sen, R., Mamolos, A.P., Veresoglou, D.S., 2011. Plant species identity and 686 arbuscular mycorrhizal status modulate potential nitrification rates in nitrogen-limited 687 grassland soils. J. Ecol. 99, 1339–1349. doi:10.1111/j.1365-2745.2011.01863.x 688 Verhagen, F.J.M., Laanbroek, H.J., Woldendrop, J.W., 1995. Competition for ammonium 689 between plant roots and nitrifying and heterotrophic bacteria and the effects of protozoan 690 grazing. Plant Soil 170, 241–250. doi:10.1007/BF00010477 691 Vierheilig, H., Coughlan, A.P., Wyss, U.R.S., Recherche, C. De, 1998. Ink and Vinegar , a 692 Simple Staining Technique for Arbuscular-Mycorrhizal Fungi 64, 5004–5007. 693 Whiteside, M.D., Garcia, M.O., Treseder, K.K., 2012. Amino Acid Uptake in Arbuscular 694 Mycorrhizal Plants. PLoS One 7. doi:10.1371/journal.pone.0047643 695 Zhu, X., Burger, M., Doane, T.A., Horwath, W.R., 2013. Ammonia oxidation pathways and 696 nitrifier denitrification are significant sources of N2O and NO under low oxygen 697 availability. Proc. Natl. Acad. Sci. 110, 6328–6333. doi:10.1073/pnas.1219993110 698 699 700 701 702 703 704 705 706 707 30 Table 1 Plant biomass production, N content and N uptake. Means are followed by standard 708 error between parentheses (n=4). The outputs Linear Mixed Model are shown at the bottom of 709 the table. 710 Shoot biomass Root biomass Shoot N content Root N content Shoot N uptake Root N uptake (g pot-1) (%) (mg pot -1) day 7 -M/Ctr 15.10 (0.40) 1.67 (0.14) 1.16 (0.04) 1.35 (0.18) 173.9 (2.8) 22.31 (3.20) -M/N 15.11 (0.50) 1.56 (0.22) 2.37 (0.09) 1.24 (0.06) 357.3 (15.1) 19.33 (3.04) -M/DCD 14.20 (0.18) 1.23 (0.13) 2.23 (0.11) 1.38 (0.03) 315.2 (12.4) 17.07 (1.84) +M/Ctr 13.16 (0.10) 1.16 (0.14) 1.24 (0.04) 1.57 (0.08) 163.1 (5.9) 18.37 (2.90) +M/N 12.50 (1.09) 1.70 (0.12) 2.07 (0.08) 1.44 (0.07) 259.0 (29.1) 24.35 (1.68) +M/DCD 13.81 (1.07) 1.94 (0.12) 1.89 (0.17) 1.63 (0.07) 264.7 (41.8) 31.37 (0.72) day 14 -M/Ctr 15.96 (0.12) 1.67 (0.15) 1.05 (0.05) 1.10 (0.09) 167.1 (7.7) 17.99 (0.35) -M/N 16.31 (0.42) 1.40 (0.11) 2.29 (0.07) 1.22 (0.08) 372.8 (7.9) 17.12 (1.75) -M/DCD 16.55 (0.99) 1.80 (0.17) 2.36 (0.05) 1.15 (0.03) 390.2 (22.2) 20.51 (1.93) +M/Ctr 13.24 (0.48) 1.38 (0.12) 1.31 (0.11) 1.36 (0.03) 171.4 (9.6) 18.83 (1.60) +M/N 16.08 (0.55) 2.15 (0.20) 2.14 (0.12) 1.42 (0.11) 343.5 (23.0) 30.26 (2.77) +M/DCD 14.52 (0.31) 1.96 (0.28) 2.03 (0.02) 1.41 (0.14) 294.7 (8.8) 27.31 (3.57) Effects F-value (p-value) M 21.48 (***) 4.641 (*) 6.673 (*) 18.51 (***) 18.26 (***) 27.47 (***) F 1.068 (n.s.) 4.771 (*) 167.6 (***) 0.490 (n.s.) 91.23 (***) 5.578 (**) T 16.85 (***) 5.874 (*) 0.607 (n.s.) 8.647 (**) 9.949 (**) 0.020 (n.s.) MxF 0.926 (n.s.) 13.63 (***) 9.252 (***) 0.105 (n.s.) 4.016 (*) 11.59 (***) MxT 0.000 (n.s.) 0.401 (n.s.) 1.154 (n.s.) 0.019 (n.s.) 0.355 (n.s) 0.392 (n.s.) FxT 2.425 (n.s.) 0.527 (n.s.) 1.037 (n.s.) 1.710 (n.s.) 2.379 (n.s.) 0.518 (n.s.) MxFxT 2.924 (n.s.) 4.975 (*) 0.267 (n.s.) 0.011 (n.s.) 2.273 (n.s.) 4.425 (*) -M/Ctr no mycorrhiza control; -M/N no mycorrhiza and urea application; -M/DCD no 711 mycorrhiza and urea with DCD application; +M/Ctr arbuscular mycorrhiza control; +M/N 712 arbuscular mycorrhiza and urea application; +M/DCD arbuscular mycorrhiza and urea with 713 DCD application. M mycorrhiza; F fertilizer; T time 714 *, **, *** indicate p<0.05, p<0.01 and p<0.001, respectively 715 716 717 31 Table 2 Potential urease activity, NH4+-N and NO3--N content, microbial biomass C (MBC) 718 and microbial biomass N (MBN). Means are followed by standard error between parentheses 719 (n=4). The outputs of general linear model are shown at the bottom of the table. 720 Urease* NH4+-N NO3--N Nmin MBC MBN (mg kg-1) day 7 -M/Ctr 7.09 (1.08) 1.81 (0.54) 1.04 (0.60) 2.85 (0.78) 90.56 (20.53) 17.32 (2.73) -M/N 9.24 (0.30) 1.97 (0.64) 18.19 (2.73) 20.16 (2.11) 76.87 (19.77) 16.51 (4.93) -M/DCD 11.49 (1.65) 15.77 (1.20) 2.04 (0.84) 17.81 (1.67) 101.67 (8.06) 27.16 (3.32) +M/Ctr 7.81 (0.62) 2.43 (0.84) 1.02 (0.25) 3.45 (0.93) 73.14 (9.71) 25.18 (5.52) +M/N 8.34 (1.14) 7.39 (2.63) 19.78 (3.63) 27.16 (4.39) 109.0 (17.83) 25.80 (0.27) +M/DCD 6.66 (1.90) 21.57 (5.34) 0.02 (0.01) 21.60 (5.33) 82.27 (15.75) 28.53 (3.86) day 14 -M/Ctr 10.08 (1.25) 2.24 (0.53) 2.59 (1.15) 4.84 (1.39) 95.83 (0.83) 16.47 (6.70) -M/N 12.03 (2.59) 0.80 (0.36) 5.36 (2.29) 6.16 (2.11) 70.73 (17.46) 20.04 (4.08) -M/DCD 8.90 (0.47) 5.71 (2.87) 4.41 (0.07) 10.12 (2.90) 103.4 (20.56) 27.31 (5.89) +M/Ctr 9.76 (0.22) 4.57 (2.62) 0.23 (0.16) 4.80 (2.61) 90.62 (16.47) 24.90 (5.86) +M/N 12.05 (1.19) 1.74 (0.37) 9.06 (1.78) 10.79 (2.03) 139.17 (8.00) 39.16 (10.78) +M/DCD 13.22 (2.61) 2.98 (1.21) 1.30 (0.38) 4.28 (1.47) 150.3 (25.28) 18.29 (5.62) Factor F-value (p-value) M 0.037 (n.s.) 2.371 (n.s.) 6.195 (*) 0.000 (n.s.) 3.432 (n.s) 4.294 (*) F 1.549 (n.s.) 13.82 (***) 73.38 (***) 21.16 (***) 1.770 (n.s.) 1.286 (n.s.) T 9.129 (***) 13.55 (***) 0.410 (n.s.) 12.87 (***) 4.187 (*) 0.176 (n.s.) MxF 0.051 (n.s.) 1.428 (n.s.) 7.997 (***) 2.058 (n.s.) 3.553 (*) 3.126 (n.s.) MxT 3.138 (n.s.) 0.531 (n.s.) 0.250 (n.s.) 1.080 (n.s.) 4.073 (*) 0.051 (n.s.) FxT 0.189 (n.s.) 8.243 (***) 14.88 (***) 7.514 (**) 0.661 (n.s.) 1.705 (n.s.) MxFxT 3.366 (*) 0.443 (n.s.) 3.378 (n.s.) 1.399 (n.s.) 0.681 (n.s.) 0.536 (n.s.) * Potential urease activity (mg NH4+-N g-1 h-1) 721 -M/Ctr no mycorrhiza control; -M/N no mycorrhiza and urea application; -M/DCD no 722 mycorrhiza and urea with DCD application; +M/Ctr arbuscular mycorrhiza control; +M/N 723 arbuscular mycorrhiza and urea application; +M/DCD arbuscular mycorrhiza and urea with 724 DCD application. M mycorrhiza; F fertilizer; T time 725 *, **, *** indicate p<0.05, p<0.01 and p<0.001, respectively 726 727 728 729 32 730 731 732 733 Figure 1 N2O (A) and CO2 (B) emission rates from pots planted with B.decumbens. Vertical 734 dotted lines indicate times of destructive sampling. Error bars indicate standard errors (n=6). –735 M/Ctr no mycorrhiza control; -M/N no mycorrhiza and urea application; -M/DCD no 736 mycorrhiza and urea with DCD application; +M/Ctr arbuscular mycorrhiza control; +M/N 737 arbuscular mycorrhiza and urea application; +M/DCD arbuscular mycorrhiza and urea with 738 DCD application. M effect of mycorrhizal inoculation; F effect of fertilizer. *,**,*** indicate 739 statistically significant differences (Linear Mixed Model) at p<0.05, p<0.01 and p<0.001, 740 respectivelly, n.s. not significant. 741 0 1 2 3 4 5 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 N 2 O e m is si o n s ra te s (µ g N 2 O -N k g -1 h -1 ) Time since fertilization (hours) (A) -M/Ctr -M/N -M/DCD +M/Ctr +M/N +M/DCD 0 2 4 6 8 10 12 14 16 0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 C O 2 em is si o n s ra te s (µ g C O 2 k g -1 h -1 ) Time since fertilization (hours) (B) Day 7 Day 14 Day 7 Day 14 M ** F *** MxF *** M n.s. F *** MxF n.s. 33 742 34 743 744 745 Figure 2 The amoA gene abundance of AOB (A) and the potential nitrification rate (PNR) (B). 746 Bars represent standard error of the mean (n=4). –M/Ctr no mycorrhiza control; -M/N no 747 mycorrhiza and urea application; -M/DCD no mycorrhiza and urea with DCD application; 748 +M/Ctr arbuscular mycorrhiza without fertilization (control); +M/N arbuscular mycorrhiza and 749 urea application; +M/DCD arbuscular mycorrhiza and urea with DCD application. M effect of 750 mycorrhizal inoculation; F effect of fertilizer. *,**,*** inidicate statistically significant 751 differences (Linear Mixed Model) at p<0.05, p<0.01 and p<0.001, respectivelly, n.s. not 752 significant. 753 754 755 756 0.E+00 5.E+05 1.E+06 2.E+06 2.E+06 3.E+06 3.E+06 4.E+06 4.E+06 -M/Ctr -M/N -M/DCD +M/Ctr +M/N +M/DCD am o A -A O B (g en e co p ie s g -1 ) (A)day 7 after fertilization day 14 after fertilization 0 2 4 6 8 10 12 14 -M/Ctr -M/N -M/DCD +M/Ctr +M/N +M/DCD P N R (µ g N -N O 3 - g -1 d ay -1 ) (B) M ** F ** MxF n.s. M n.s. F * MxF * 35 757