Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Plant breeding wikipedia , lookup
Crop rotation wikipedia , lookup
Plant defense against herbivory wikipedia , lookup
Renewable resource wikipedia , lookup
No-till farming wikipedia , lookup
Meadow vole wikipedia , lookup
Tropical rainforest wikipedia , lookup
Human impact on the nitrogen cycle wikipedia , lookup
Biological Dynamics of Forest Fragments Project wikipedia , lookup
1 No effects of Epichloë endophyte infection on nitrogen cycling in meadow 2 fescue (Schedonorus pratensis) grassland 3 4 Juha Mikola 1, Marjo Helander 2,3, Kari Saikkonen 3 5 6 1 7 Finland 8 2 Section of Ecology, Department of Biology, University of Turku, 20014 Turku, Finland 9 3 Management and Production of Renewable Resources, Natural Resources Institute Finland 10 11 Department of Environmental Sciences, University of Helsinki, Niemenkatu 73, 15140 Lahti, (Luke), Itäinen Pitkäkatu 3, 20520 Turku, Finland 12 Abstract 13 Background and aims Systemic Epichloë endophytes produce alkaloids that protect their grass 14 hosts against pathogens and herbivores. These alkaloids, together with other endophyte induced 15 changes in litter quality, may decelerate the decomposition of infected grass litter, but so far no 16 study has tested whether the effects on decomposition rate translate into changes in N cycling in 17 infected grasslands. Here we test if the Epichloë uncinata infection of meadow fescue, Schedonorus 18 pratensis decelerates litter decomposition and N release, increases soil C and N accumulation and 19 lowers the availability of mineral N in the soil under infected grass. 20 Methods To analyze grass litter and soil attributes, samples were collected from endophyte 21 infected (E+) and non-infected (E-) field plots, established seven years earlier. At the time of the 22 study, the frequency of E+ plants was 80-90 % and 0-3 % in the E+ and E- plots, respectively. 23 Litter decomposition rate and litter N release were examined using litter mesh bags, placed on field 24 ground. Soil mineral N availability was estimated using ion exchange resin bags that were buried in 25 the soil. 26 Results Epichloë uncinata infection did not affect meadow fescue litter N%, litter mass loss or 27 litter N release. Neither did soil C and N content and resin NH4 and NO3 contents differ between the 28 E+ and E- grass plots. E+ litter did not decompose faster in E+ than E- plots, i.e. no home-field 29 advantage was observed. 30 Conclusions We did not find evidence that Epichloë uncinata infection would decelerate N cycling 31 and reduce N mineralization in meadow fescue grasslands. This suggests that the infection may not 32 decrease the benefit of the endophyte-grass symbiosis by reducing soil fertility. 33 34 Keywords Epichloë uncinata, soil, plant litter, decomposition, systemic endophyte, mineralization, 35 symbiosis. 36 Introduction 37 Plant litter decomposition and nutrient mineralization are fundamental processes in all ecosystems. 38 A great share of the variation in these processes within and among ecosystems can be explained by 39 plant trait variation (Wardle 2002). Due to the adaptation of plants and their foliage to different 40 habitats, plant species differ in the quality of leaf litter they produce and such differences can have 41 significant consequences on litter decomposition and nutrient cycling in terrestrial ecosystems 42 (Hobbie 1992, Wardle et al. 1997). Significant intra-specific genetic variation has also been shown 43 to exist in both woody (Madritch et al. 2006, Silfver et al. 2007) and herbaceous species (Crutsinger 44 et al. 2009), which allows natural selection to drive litter decomposition through the effects on 45 foliage traits in plant populations (Whitham et al. 2006). Much of the interspecific variation in plant 46 traits and litter quality is due to plant adaptation to trophic interactions (Wardle 2002), but 47 interactions can also modify litter quality in the ecological time scale. For instance, herbivore 48 feeding can induce secondary metabolite production in plant leaves (Nykänen and Koricheva 2004) 49 and because these metabolites can remain through leaf senescence and reduce litter decomposition 50 rate (Findlay et al. 1996, Schweitzer et al. 2005), herbivore-induced changes in green leaf chemistry 51 can have after-life effects on litter decomposition. Although all herbivore effects on litter 52 decomposition are not explainable by induced production of secondary metabolites (Silfver et al. 53 2015) and the effects can vary across plant genotypes (Schweitzer et al. 2005, Silfver et al. 2015), 54 these effects show how organisms that interact with live plants can significantly contribute to plant 55 litter decomposition. 56 Like the herbivores, foliar endophytes can modify plant green leaf chemistry (Huitu et al. 57 2014). The systemic Epichloë endophytes (including the fungi earlier classified in Neotyphodium; 58 Leuchtmann et al. 2014), which as common among the cool season Poaceae grasses (subfamily 59 Pooideae) (Schardl et al. 2011), are particularly well known for their abundant production of 60 alkaloids (Lehtonen et al. 2005, Saikkonen et al. 2013a) that protect their grass hosts against 61 pathogens and herbivores (Clay 1988, Clay and Schardl 2002). Like the secondary metabolites of 62 plant leaves, these alkaloids remain in senescent grass leaves (Siegrist et al. 2010) and are therefore 63 assumed to have effects on litter decomposition and nutrient cycling (Saikkonen et al. 2015). There 64 are few studies that have examined the effects of Epichloë infection on grass litter decomposition, 65 but so far all published studies have found a weak, negative effect on litter mass loss (Omacini et al. 66 2004, Lemons et al. 2005, Siegrist et al. 2010). There is also evidence of lower microbial activity 67 and accumulation of organic matter under endophyte infected grass (Franzluebbers et al. 1999). 68 These effects are assumed to be due to the alkaloids having adverse effects on soil decomposers 69 (Franzluebbers et al. 1999), but this assumption has also been questioned in a study, where high 70 alkaloid concentrations of green grass leaves had only a small negative effect on leaf decomposition 71 in comparison to green leaves with no alkaloids (Siegrist et al. 2010). Further investigations of the 72 effects of Epichloë infection on grass litter decomposition are therefore needed (Omacini et al. 73 2012), and so far no study has tested whether the Epichloë effects on litter mass loss translate to 74 changes in nutrient cycling in plant communities with Epichloë infected grasses. These effects 75 might be particularly interesting as the symbiotic interaction between the grass and its endophyte 76 fungus appears to be mutualistic in fertile soils only (Saikkonen et al. 2006). The Epichloë fungi are 77 commonly considered as strong plant mutualists as their fitness depends on host fitness and the 78 fungi can significantly increase the vigor, environmental tolerance and pathogen and herbivore 79 resistance of the host (Clay 1988, Clay and Schardl 2002, Saikkonen et al. 2004, 2010). However, if 80 the endophyte infection could significantly decrease nutrient return from grass litter, the fertility of 81 the soil along with the benefits of the infection might decrease to the level, where selection would 82 start acting against the infection. It should be noted though that any effects of litter quality on litter 83 nutrient release can quickly vanish if the decomposer communities are able to adapt to the 84 chemistry of the litter. In such case, the ability of local decomposer communities to degrade litters 85 of different quality would not differ, a phenomenon called a home-field advantage (Austin et al. 86 2014). Home-field advantage has been shown to evolve under different plant species (Vivanco and 87 Austin 2008) and plant genotypes (Madritch and Lindroth 2011), and the faster decomposition of 88 Epichloë infected litter in infected than non-infected tall fescue plots (Lemons et al. 2005) suggests 89 that this phenomenon can also be induced by the chemistry of endophyte-infected litter. 90 Here we report results from a study, where we examined the effects of the systemic Epichloë 91 uncinata endophyte on grassland nitrogen (N) cycling in infected (E+) and non-infected (E-) 92 meadow fescue, Schedonorus pratensis (ex. Lolium pratense and Festuca pratensis) field plots. 93 Meadow fescue is a widely used perennial forage grass in Nordic countries, but also grows in 94 meadows, roadsides and wastelands (Hämet-Ahti et al., 1988). Both agricultural and wild 95 populations are often colonized by E. uncinata, a heritable systemic fungus, which grows 96 asymptomatically throughout the plant and is transmitted to new plant generations in plant seed 97 (Saikkonen et al. 1998, Cheplick and Faeth 2009, Leuchtmann et al. 2014). Epichloë uncinata is 98 known to produce loline alkaloids and their derivatives (Lehtonen et al. 2005). These appear to be 99 non-toxic to large mammal herbivores (Clay and Schardl 2002), but can be noxious to invertebrates 100 and small vertebrates (Conover 2003, Saikkonen et al. 2006, Huitu et al. 2008). Infected plants also 101 have higher silicon concentration than uninfected plants (Huitu et al. 2014), but whether the 102 differences in leaf chemistry affect litter decomposition and N cycling in meadow fescue grasslands 103 has not earlier been investigated. Based on the earlier findings of reduced litter mass loss in 104 Epichloë infected Italian ryegrass, Lolium multiflorum (Omacini et al. 2004) and tall fescue, 105 Schedonorus arundinaceus (syn. Festuca arundinacea and Lolium arundinaceum) (Lemons et al. 106 2005, Siegrist et al. 2010), we hypothesized that (1) E. uncinata infection will decelerate the 107 decomposition rate of meadow fescue shoot litter, and as a result, (2) more organic C and N will 108 accumulate in the soil in E+ than E- plots (Franzluebbers et al. 1999), (3) less N will be released 109 from E+ than E- litter and (4) less mineralized N (NH4 and NO3) will be available in the soil under 110 E+ than E- plants. However, since microbial communities can express home field advantage 111 (Austin et al. 2014), we also predicted that the mass loss and N release of E+ litter would be higher 112 in the E+ than E- plots. Finally, since the Epichloë infected grass often grow more vigorously than 113 non-infected grass (Clay 1998), we followed soil temperature in our E+ and E- plots to assess the 114 possibility that (6) any observed effects of E. uncinata infection on litter decomposition and N 115 cycling could simply be explained by differences in the degree of shade and soil temperature 116 between the E+ and E- plots. 117 118 Materials and methods 119 Experimental site and endophyte treatment 120 The experiment was conducted in Jokioinen, South-West Finland (60° 49´ N, 23° 30´ E) in an old 121 agricultural field, which was tilled and fertilized with cow manure (30 000 kg ha-1). Twenty metal 122 fenced enclosures (25 m × 39 m) were established in May 2006: ten of the enclosures (later called 123 plots) were sown with endophyte-free (E-; 0% endophyte frequency) and another ten with 124 endophyte infected (E+; 79% infection frequency) seeds of the meadow fescue (Schedonorus 125 pratensis) cultivar ‘Kasper’ (20 kg seeds ha-1). The E+ and E- treatments were randomly assigned 126 to 10 plot pairs (used as replicate blocks in statistical analyses). The seeds were obtained from seed 127 production farms via the Finnish Food Safety Authority (EVIRA, Seed Certification Unit, Loimaa, 128 Finland). In June 2007, the plots were fertilized with a commercial fertilizer (16:9:22 N:P:K with 129 micronutrients; Kemira, product number 0647334). In 2012, six years after establishment, the 130 frequency of E+ plants was 80-90 % and 0-3 % in the E+ and E- plots, respectively, and the aerial 131 cover of meadow fescue had decreased from 100 % to 75% in E− and 98% in E+ plots due to weed 132 invasion (Saikkonen et al. 2013b). 133 134 Sampling and analyses 135 For the samples and measurements of the present study, a triangular area (each side 3.5 m) with 136 high meadow fescue coverage was selected in each plot in mid-May 2013. Overwintered, erect 137 meadow fescue shoot litter, composed of senescent leaves and stems, was collected from each of 138 the 20 plots (not from the area of the triangle) at the same time and stored at room temperature until 139 placed into litter mesh bags (7 cm × 11 cm, mesh size 1 mm, 5 g fresh mass in each bag). Six bags 140 were established from each of the 20 litter samples: two of the bags were later placed on the plot, 141 where the litter was collected (i.e. E- litter placed on an E- plot, E+ litter placed on an E+ plot), two 142 were placed on the adjacent plot in the same replicate block (i.e. E- litter placed on an E+ plot, E+ 143 litter placed on an E- plot), and the remaining two bags were placed outside the experimental area 144 amongst a dense species-rich herbaceous vegetation. The latter area was used as a comparison site, 145 where the decomposers had not been subjected to, and potentially adapted to either E- or E+ litter 146 and is called as a “neutral” plot. The litter bags were placed on soil surface (in E+ and E- plots in 147 the middle of the triangular subplot) on 11th of June 2013, covered using a white mesh and 148 collected on 21st of August 2013. The collected bags were kept frozen (-18 °C) until the litter in 149 each bag was dried (48 h, 70 °C), weighed and ground and its N concentration analyzed using a 150 LECO CNS-2000 analyzer (LECO Corporation, USA). To be able to calculate litter mass loss and 151 the amount of N released from each litter bag during the decomposition period, subsamples of the 152 original 20 field samples were weighed before and after drying (to determine their dry mass 153 content), ground and analyzed for N concentration. The amount of N released from each litter bag 154 was then calculated using the initial and final dry masses and N concentrations. 155 To analyze the effect of the grass endophyte infection on the soil N and C content, three 156 random soil cores (depth 10 cm, diameter 3 cm) were collected from each subplot in mid-May 157 2013. After removing visible roots, the samples were pooled within each subplot, dried (48 h, 70 158 °C) and the soil C and N concentrations were determined using the LECO CNS-2000 analyzer. The 159 availability of mineral N (NH4+ and NO3-) in the soil was estimated using mesh bags (5 cm × 5 cm, 160 mesh size 200 µm) filled with 1 g of ion exchange resin (Amberlite® MB-150 Mixed Bed 161 Exchanger, Rohm and Haas, France). Nine bags were buried in the depth of 5 cm in each triangle 162 on 11th of June 2013.The bags were collected 5, 7 and 14 weeks later, three random bags at each 163 date. In the laboratory, the bags were rinsed using distilled water, the three replicate bags of each 164 date were pooled and their NH4+ and NO3- contents extracted in 50 ml of 2M KCl. The KCl solution 165 was filtered through a glass microfiber filter (Whatman, GE Healthcare Europe GmbH) and the 166 NH4 and NO3 concentrations were analyzed using a Lachat QuikChem 8000 analyzer (Zallweger 167 Analytics, Inc., Lachat Instruments Division, USA). 168 Soil temperature was measured in mid-July, early and late August and mid-September 2013 in 169 the top and 5-10 cm layer. For each measurement in each layer, the value is based on five readings 170 at random spots within the triangle. 171 172 Statistical analyses 173 The effects of grass endophyte infection and plot endophyte status on plant and soil variables were 174 analyzed using ANOVA models. The replicate block was included in the models to explain field 175 variation and repeated measures ANOVA was used for those response variables, which had 176 repeated measures in time (resin NH4 and NO3 contents, soil moisture and temperature) or space, 177 i.e. repeated measurements in E+, E- and neutral plots (litter mass loss and litter N release). The 178 homogeneity of variances was tested using the Levene’s test and the normality using model 179 residuals and the Shapiro-Wilk test. None of the response variables needed transformations to fulfil 180 these requirements. When the sphericity of data was violated in the repeated measures ANOVA, the 181 degrees of freedom were corrected using the Greenhouse-Geisser ε. 182 183 Results 184 The endophyte infection did not affect the grass litter attributes, i.e. shoot litter N%, litter mass loss 185 and litter N release (Table 1). Litter N release was higher in the neutral than E+ and E- plots, while 186 litter mass loss was not affected by the identity of the placement plot (Table 2). Grass endophyte 187 status × placement plot identity interactions were not significant for either litter N release (F=0.09, 188 P=0.913) or litter mass loss (F=0.10, 0.821). Litter N% correlated positively with litter N release (in 189 E+ plots r=0.54, P=0.015, n=20; E- plots r=0.53, P=0.016; neutral plots r=0.72, P<0.001) and litter 190 mass loss (in E+ plots r=0.58, P=0.007; E- plots r=0.79, P<0.001; neutral plots r=0.53, P=0.017). 191 Soil temperature and N and C contents did not differ between the E+ and E- plots (Table 3). 192 The resin NH4 content varied across the summer (F=7.20, P=0.005; July 72±6, August 105±9, 193 September 78±8 mg g-1 resin), while the NO3 content did not (F=0.10, P=0.904), and neither was 194 affected by plot endophyte status (Table 3). The mean NO3 content of the three resin harvests of a 195 plot correlated positively with plot litter N% (r=0.58, P=0.008, n=20), and the resin NH4 and NO3 196 contents correlated positively with each other in August (r=0.50, P=0.023) and September (r=0.74, 197 P<0.001). The August temperatures of the 5-10 cm soil layer correlated negatively with the August 198 resin NH4 (r=-0.67, P=0.001, n=20) and NO3 (r=-0.59, P=0.006) contents and the September 199 temperatures with the September NH4 (r=-0.51, P=0.021) and NO3 (r=-0.44, P=0.052) contents. 200 201 Discussion 202 We hypothesized that the meadow fescue grass infected by Epichloë uncinata would produce litter 203 that decomposes more slowly and releases less N than the litter produced by non-infected grass and 204 that this would lead to more organic C and N accumulating and less mineral N being available in 205 the soil under infected than non-infected plants. These hypotheses were, however, falsified by our 206 data: although the mean N release was lower from E+ (0.26 mg N g-1 dry litter) than E- (1.06 mg) 207 litter as we predicted, this difference was not statistically significant and the other variables showed 208 no signs of endophyte effects. 209 The absence of the endophyte effect on litter mass loss in our study is in contrast to earlier 210 observations with other fungus-grass combinations. Omacini et al. (2004) found that the litter of E+ 211 Italian ryegrass had 18% lower mean decay rate than the E- litter. Similarly, Lemons et al. (2005) 212 showed that E+ tall fescue (S. arundinaceus) litter had 4.4% and 7.9% more mass remaining than E- 213 litter after 197 and 256 d of decomposition. In the latter study, the endophyte effect got stronger 214 with time and no difference was observed after 135 d of decomposition and ~33% mass loss. Our 215 meadow fescue litter was subjected to decomposition for 70 d with an average mass loss of 33%. If 216 the endophyte effect on litter mass loss only appeared at the later stage of decomposition as in 217 Lemons et al. (2005), this would suggest that we missed the effect in our shorter test. However, late 218 effects do not seem to be a rule as the significant reduction in litter decay rate in Omacini et al. 219 (2004) was found in an 83-d study with a mean mass loss of ~23%. Another explanation for the 220 discrepancy between ours and earlier results might be the fact that we used overwintered litter and 221 the alkaloids had partly degraded during the winter (see Kallenbach et al. 2003). Using 222 overwintered litter to examine litter effects on N cycling is, however, justified as most of the litter 223 enters the soil at this stage. Moreover, the absence of the endophyte effect on litter decomposition 224 agrees well with our finding that soil C and N contents and N mineralization, which tell of longer- 225 term endophyte effects, did not respond to the endophyte status of the field plots. The last piece of 226 evidence of the absence of endophyte effects on N cycling in our field site comes from an earlier 227 investigation carried out in the same site; in this study, the green leaf N% did not differ between the 228 infected and non-infected meadow fescue (Huitu et al. 2014). If N cycling differed significantly 229 between the E+ and E- plots in our site, the grass green leaf N% would most likely respond to the 230 difference. Lastly, it is good to note that the theory of the effect of litter decomposition rate on soil 231 organic matter accumulation is currently under revision (Cotrufo et al. 2013). In contrast to the 232 earlier belief, slow decomposition of low-quality litter is now suggested to decrease the 233 accumulation of soil organic matter due to lower accumulation of persistent microbial products 234 (Cotrufo et al. 2013). This is not highly relevant in our study as the Epichloë infection had no effect 235 on soil C and N accumulation, but it challenges the interpretation that alkaloids would be the cause 236 for organic matter accumulation under endophyte infected grass (Franzluebbers et al. 1999). 237 Litter quality is always a combination of attributes that enhance decomposition such as litter 238 N% (Silfver et al. 2007) and those that decelerate decomposition such as lignin concentration 239 (Melillo et al. 1982), but also depends on their interaction (Berg 2014). Besides producing 240 alkaloids, the Epichloë fungi are known to have effects on many attributes of grass green leaf 241 chemistry (Vázquez-de-Aldana 2013, Huitu et al. 2014, Saikkonen et al. 2015) and the variation in 242 the relative magnitude of these effects may be one cause for the variation that appears in Epichloë 243 effects on litter decomposition. In our study, the Epichloë infection did not affect litter N%, but the 244 plot-to-plot variation in litter N% correlated positively with litter mass loss, litter N release and NO3 245 availability in the plot soil. This shows that litter N%, even when originating from the limited 246 spatial variation of a homogeneous agricultural soil, can be a main control of litter decomposition 247 and N cycling in grasslands. Since litter N% was not affected by grass endophyte infection in our 248 study, litter N% could not mask any potential alkaloid effects, but such mask is possible in other 249 studies as the endophyte infection has been shown to increase grass green leaf N% (Siegrist et al. 250 2010, Vázquez-de-Aldana 2013). In fact, Siegrist et al. (2010) speculated that the weak effect of 251 high alkaloid concentrations on decomposition in their study with green leaf material was due to the 252 high N% of the material, which potentially alleviated the adverse alkaloid effects. Considering that 253 the alkaloid concentrations of green leaves vary widely among grass-endophyte symbioses 254 (TePaske et al. 1993) and are low in the senescent leaves and litter of infected grasses (Eichenseer 255 et al. 1991, Siegrist et al. 2010), leaf N% provides a noteworthy, additional mechanism to test in the 256 studies of Epichloë endophyte effects on litter decomposition and grassland N cycling. It is likely 257 though that variation in litter N% will also have its limitations in predicting the endophyte effects 258 on decomposition rates; for example, while the increasing atmospheric CO2 concentrations have 259 predictable effects on litter N% (reduced) and lignin concentrations (elevated), these effects do not 260 appear to lead to slower litter decomposition rates as expected (Norby et al. 2001). 261 We did not find evidence of a home-field advantage (Austin et al. 2014), i.e. significant ‘litter 262 endophyte status × plot endophyte status’ interaction effects on litter decomposition or litter N 263 release, neither were there any signs of plot endophyte status affecting litter decomposition. These 264 results provide further evidence of the absence of Epichloë uncinata effects on the activity or 265 structure of the decomposer community in our meadow fescue grassland. Instead, we found that 266 vegetation composition affected litter N release (the release was greater in the “neutral” than 267 experimental plots) and that soil mineral N availability was linked to soil temperature. These results 268 suggest two potential long-term mechanisms for endophyte effects on litter decomposition. First, 269 since the grass endophyte infection can reduce the species richness of a plant community (Clay and 270 Holah 1999) and prohibit weed invasion to a grass field (Saikkonen et al. 2013b), the species 271 composition of infected and non-infected grass fields (wild or established) may eventually differ 272 insomuch that this will have effects on the decomposition processes. In our study, we purposely 273 eliminated this mechanism by choosing subplots (the triangles) that were dominated by meadow 274 fescue. Secondly, the more vigorous growth of endophyte infected grasses (Clay 1988) and the 275 concomitant changes in plant species richness (Clay and Holah 1999, Saikkonen et al. 2013b) may 276 impact soil temperature, which will then reflect in N mineralization. Such effects would require a 277 complex mixture of plant-plant competition, herbivory and community dynamics, but would still 278 origin from the endophyte symbiosis. 279 To conclude, we could not find evidence that Epichloë uncinata infection would decelerate N 280 cycling and reduce N mineralization in meadow fescue grassland. This suggests that the infection 281 may not decrease the benefit of the endophyte-grass symbiosis by reducing soil fertility as we 282 speculated. Instead of the endophyte infection, the spatial variation of litter N% was a good 283 predictor for the variation in litter mass loss, litter N release and soil mineral N concentrations in 284 our field. When such clear link between litter N%, litter decomposition and N mineralization is 285 combined with the earlier findings of low alkaloid concentrations in endophyte infected grass litter 286 (Siegrist et al. 2010), varying effects of the infection on litter decomposition (Omacini et al. 2012) 287 and the positive effects of the endophyte on grass amino acid and N concentrations (Lyons et al. 288 1990, Siegrist et al. 2010), it appears that the endophyte effect on litter N% might be a worthwhile 289 mechanism to examine to understand the endophyte effects on litter mass loss and N cycling in 290 grassland ecosystems. 291 292 Acknowledgements 293 We thank Hillamari Lehikoinen, Serdar Dirihan and Tuija Koivisto for assisting with litter and resin 294 bags, Merja Myllys for measuring soil temperatures and Viivi Toivio, Santeri Savolainen and 295 Marianne Lehtonen for processing and analyzing the resin and litter samples. The study was 296 supported by the Academy of Finland (grants no. 137909 and 281354) to Kari Saikkonen. 297 298 References 299 Austin AT, Vivanco L, González-Arzac A, Pérez LI (2014) There’s no place like home? An 300 exploration of the mechanisms behind plant litter–decomposer affinity in terrestrial ecosystems. 301 New Phytol 201:307-314 302 303 304 305 306 307 Berg B (2014) Decomposition patterns for foliar litter - A theory for influencing factors. Soil Biol Biochem 78:222-232 Cheplick GP, Faeth S (2009) Ecology and Evolution of the Grass Endophyte Symbiosis. Oxford University Press, New York. Clay K (1988) Fungal endophytes of grasses: a defensive mutualism between plants and fungi. Ecology 69:10-16 308 309 310 311 312 313 314 315 316 Clay K, Holah J (1999) Fungal endophyte symbiosis and plant diversity in successional fields. Science 285:1742-1744 Clay K, Schardl C (2002) Evolutionary origins and ecological consequences of endophyte symbiosis with grasses. Am Nat 160:S99-S127 Conover MR(1998) Impact of consuming tall fescue leaves with the endophytic fungus, Acremonium coenophialum, on meadow voles. J Mammal 79:457-463 Crutsinger GM, Sanders NJ, Classen AT (2009) Comparing intra-and inter-specific effects on litter decomposition in an old-field ecosystem. Basic Appl Ecol 10:535-543 Eichenseer H, Dahlman DL, Bush LP (1991) Influence of endophyte infection, plant age and 317 harvest interval on Rhopalosiphum padi survival and its relation to quantity of N-formyl and N- 318 acetyl loline in tall fescue. Entomol Exp Appl 60:29–38 319 Franzluebbers AJ, Nazih N, Stuedemann JA, Fuhrmann JJ, Schomberg HH, Hartel PG (1999) Soil 320 carbon and nitrogen pools under low- and high-endophyte-infected tall fescue. Soil Sci Soc Am J 321 63:1687-1694 322 323 324 325 Findlay S, Carreiro M, Krischik V, Jones CG (1996) Effects of damage to living plants on leaf litter quality. Ecol Appl 6:269–275 Hämet-Ahti L, Suominen J, Ulvinen T, Uotila P (eds) (1998) Retkeilykasvio (Field flora of Finland). Finnish Museum of Natural History, Botanical Museum, Helsinki. 326 Hobbie SE (1992) Effects of plant species on nutrient cycling. Trends Ecol Evol 7:336-339 327 Keinänen M, Julkunen-Tiitto R, Mutikainen P, Walls M, Ovaska J, Vapaavuori E (1999) Trade-offs 328 in phenolic metabolism of silver birch: effects of fertilization, defoliation, and genotype. Ecology 329 80:1970–1986 330 331 Huitu O, Helander M, Lehtonen P, Saikkonen K (2008) Consumption of grass endophytes alters the ultraviolet spectrum of vole urine. Oecologia 156:333–340 332 Huitu O, Forbes KM, Helander M, Julkunen-Tiitto R, Lambin X, Saikkonen K, Stuart P, Sulkama 333 S, Hartley S (2014) Silicon, endophytes and secondary metabolites as grass defenses against 334 mammalian herbivores. Front Plant Sci 5:478. doi: 10.3389/fpls.2014.00478 335 336 337 338 339 340 341 342 343 344 345 Kallenbach RL, Bishop-Hurley GJ, Massie MD, Rottinghaus GE, West CP (2003) Herbage mass, nutritive value, and ergovaline concentration of stockpiled tall fescue. Crop Sci 43:1001-1005 Lehtonen P, Helander M, Wink M, Sporer F, Saikkonen K (2005) Transfer of endophyte-origin defensive alkaloids from a grass to a hemiparasitic plant. Ecol Lett 8:1256-1263 Lemons A, Clay K, Rudgers JA (2005) Connecting plant–microbial interactions above and belowground: a fungal endophyte affects decomposition. Oecologia 145:595–604 Leuchtmann A, Bacon CW, Schardl CL, White JF Jr, Tadych M (2014) Nomenclatural realignment of Neotyphodium species with genus Epichloë. Mycologia 106:202–215 Lyons PC, Evans JJ, Bacon CW (1990) Effects of the fungal endophyte Acremonium coenophialum on nitrogen accumulation and metabolism in tall fescue. Plant Physiol 92:726-732 Madritch M, Donaldson JR, Lindroth RL (2006) Genetic identity of Populus tremuloides litter 346 influences decomposition and nutrient release in a mixed forest stand. Ecosystems 9:528-537 347 Madritch MD, Lindroth RL (2011) Soil microbial communities adapt to genetic variation in leaf 348 349 350 351 352 353 354 litter inputs. Oikos 120:1696–1704 Melillo JM, Aber JD, Muratore JF (1982) Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63:621-626 Norby RJ, Cotrufo MF, Ineson P, O’Neill EG, Canadell JG (2001) Elevated CO2, litter chemistry, and decomposition: a synthesis. Oecologia 127:153-165 Nykänen H, Koricheva J (2004) Damage-induced changes in woody plants and their effects on insect herbivore performance: a meta-analysis. Oikos 104:247-268 355 Omacini M, Chaneton E, Ghersa CM, Otero P (2004) Do foliar endophytes affect grass litter 356 decomposition? A microcosm approach using Lolium multiflorum. Oikos 104:581-590 357 Omacini M, Semmartin M, Pérez LI, Gundel PE (2012) Grass–endophyte symbiosis: A neglected 358 aboveground interaction with multiple belowground consequences. Appl Soil Ecol 61:273-279 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 Saikkonen K, Faeth SH, Helander M, Sullivan TJ (1998) Fungal endophytes: a continuum of interactions with host plants. Annu Rev Ecol Syst 29:319-343 Saikkonen K, Wäli P, Helander M, Faeth SH (2004) Evolution of endophyte–plant symbioses. Trends Plant Sci 9:275-280 Saikkonen K, Lehtonen P, Helander M, Koricheva J, Faeth SH (2006) Model systems in ecology: dissecting the endophyte–grass literature. Trends Plant Sci 11:428-433 Saikkonen K, Saari S, Helander M (2010) Defensive mutualism between plants and endophytic fungi? Fungal Divers 41:101-113 Saikkonen K, Gundel PE, Helander M (2013a) Chemical ecology mediated by fungal endophytes in grasses. J Chem Ecol 39:962-968 Saikkonen K, Ruokolainen K, Huitu O, Gundel PE, Piltti T, Hamilton CE, Helander M (2013b) Fungal endophytes help prevent weed invasions. Agric Ecosyst Environ 165:1-5 Saikkonen K, Mikola J, Helander M (2015) Endophytic phyllosphere fungi and nutrient cycling in terrestrial ecosystems. Curr Sci 109:121-126 Schardl CL, Young CA, Faulkner JR, Florea S, Pan J (2011) Chemotypic diversity of epichloae, fungal symbionts of grasses. Fungal Ecol 5:331-344 375 Schweitzer JA, Bailey JK, Hart SC, Wimp GM, Chapman SK, Whitham TG (2005) The interaction 376 of plant genotype and herbivory decelerate leaf litter decomposition and alter nutrient dynamics. 377 Oikos 110:133-145 378 Siegrist JA, McCulley RL, Bush LP, Phillips TD (2010) Alkaloids may not be responsible for 379 endophyte-associated reductions in tall fescue decomposition rates. Funct Ecol 24:460-468 380 Silfver T, Mikola J, Rousi M, Roininen H, Oksanen E (2007) Leaf litter decomposition differs 381 among genotypes in a local Betula pendula population. Oecologia 152:707-714 382 Silfver T, Paaso U, Rasehorn M, Rousi M, Mikola J (2015) Genotype × herbivore effect on leaf 383 litter decomposition in Betula pendula saplings: ecological and evolutionary consequences and 384 the role of secondary metabolites. PLoS ONE 10:e0116806 385 386 387 TePaske MR, Powell RG, Clement SL (1993) Analyses of selected endophyte-infected grasses for the presence of loline-type and ergot-type alkaloids. J Agric Food Chem 41:2299-2303 Vázquez-de-Aldana BR, García-Ciudad A, García-Criado B, Vicente-Tavera S, Zabalgogeazcoa I 388 (2013) Fungal endophyte (Epichloë festucae) alters the nutrient content of Festuca rubra 389 regardless of water availability. PLoS ONE 8:e84539 390 391 392 393 394 395 Vivanco L, Austin AT (2008) Tree species identity alters forest litter decomposition through longterm plant and soil interactions in Patagonia, Argentina. J Ecol 96:727-736 Wardle DA, Zackrisson O, Hörnberg G, Gallet C (1997) The influence of island area on ecosystem properties. Science 277:1296-1299 Wardle DA (2002) Communities and Ecosystems: Linking the Aboveground and Belowground Components. Princeton University Press, Princeton, U.S.A. 396 Whitham TG, Bailey JK, Schweitzer JA, Shuster SM, Bangert RK, LeRoy CJ, Lonsdorf EV, 397 Allan GJ, DiFazio SP, Potts BM, Fischer DG, Gehring CA, Lindroth RL, Marks JC, Hart SC, 398 Wimp GM, Wooley SC (2006) A framework for community and ecosystem genetics: from genes 399 to ecosystems. Nat Rev Genet 7:510-523 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 Table 1. The means (± SE, n=10) of the attributes of E+ and E- meadow fescue litter and the F- and P-statistics of ANOVA of the effect of the Epichloë uncinata infection. Litter N% Litter mass loss (% of dry mass) b Litter N release (mg N g-1 dry litter) b E+ grass 0.96 ± 0.04 33.2 ± 1.3 0.26 ± 0.28 E- grass 1.02 ± 0.10 31.8 ± 2.3 1.06 ± 0.48 F 0.45 0.64 2.40 P 0.520 0.446 0.156 b means are based on the average of values obtained in E+, E- and neutral plots; statistics from repeated measures ANOVA Table 2. The mean attributes (± SE, n=20) of meadow fescue litter placed on E+, E- and neutral plots and the F- and P-statistics of ANOVA of the effect of the placement plot identity. a Litter mass loss (% of dry mass) Litter N release (mg N g-1 litter) a a E+ plot 32.3 ± 0.01 0.41 ± 0.29 E- plot Neutral plot 33.0 ± 0.02 32.2 ± 0.01 0.66 ± 0.33 0.91 ± 0.30 F 0.47 4.34 P 0.556 0.029 means include both E+ and E- grass; statistics from repeated measures ANOVA Table 3. The means (± SE, n=10) of soil attributes and resin NH4 and NO3 contents in E+ and Eplots and the F- and P-statistics of ANOVA of the effect of the plot endophyte status. Soil N content (% of dry mass) Soil C content (% of dry mass) Resin NH4 content (mg g-1) a Resin NO3 content (mg g-1) a Soil temperature (°C, surface) b Soil temperature (°C, 5-10 cm) b a E+ plot 0.38 ± 0.02 5.34 ± 0.23 83 ± 8 232 ± 39 17.2 ± 0.3 15.6 ± 0.2 E- plot 0.36 ± 0.01 5.16 ± 0.20 87 ± 8 233 ± 56 17.2 ± 0.3 15.6 ± 0.2 F 0.51 0.28 0.10 <0.01 0.07 0.18 P 0.493 0.609 0.758 0.995 0.798 0.685 means are based on the average of July, August and September values; statistics from repeated measures ANOVA b means are based on the average of July, August (early and late) and September values; statistics from repeated measures ANOVA