Invasive plant species do not create more negative soil conditions for other plants than natives

Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

Contents lists available at ScienceDirect

Perspectives in Plant Ecology, Evolution and Systematics journal homepage: www.elsevier.com/locate/ppees

Research article

Invasive plant species do not create more negative soil conditions for other plants than natives Corina Del Fabbro ∗ , Daniel Prati Institute of Plant Sciences, University of Bern, Altenbergrain 21, 3013 Bern, Switzerland

a r t i c l e

i n f o

Article history: Received 6 June 2014 Received in revised form 12 February 2015 Accepted 12 February 2015 Available online 19 February 2015 Keywords: Allelopathy Biological invasions Multi-species experiment Novel weapons hypothesis Soil biota

a b s t r a c t A major task in ecology is to establish the degree of generality of ecological mechanisms. Here we present results from a multi-species experiment that tested whether a set of invasive species altered the soil conditions to the detriment of other species by releasing allelopathic compounds or inducing shifts in soil biota composition, and whether this effect was more pronounced relative to a set of closely related native species. We pre-cultivated soil with 23 exotic invasive, 19 related native and 6 related exotic garden species and used plain soil as a control. To separate allelopathy from effects on the soil biota, we sterilized half of the soil. Then, we compared the effect of soil pre-cultivation and sterilization on germination and growth of four native test species in two experiments. The general effect of soil sterilization was positive. The effect of soil pre-cultivation on test species performance was neutral to positive, and sterilization reduced this positive effect. This indicates general absence of allelopathic compounds and a shift toward a less antagonistic soil biota by cultivation species. In both experiments, pre-cultivation effects did not differ systematically between exotic invasive, exotic garden or native species. Our results do not support the hypothesis that invasive plants generally inhibit the growth of others by releasing allelopathic compounds or accumulating a detrimental soil biota. © 2015 Geobotanisches Institut ETH, Stiftung Ruebel. Published by Elsevier GmbH. All rights reserved.

Introduction Several mechanisms have been proposed to explain the success of plant invaders. Many of them involve natural enemies (e.g., Keane and Crawley, 2002) or other interspecific interactions (e.g., Mitchell et al., 2006), changes in the competitive balance between plants (e.g., Blossey and Notzold, 1995), or changes in resource availability (Davis et al., 2000). During recent years, several studies have highlighted the critical role of soil ecology for the study of plant invasions and the importance of belowground mechanisms for plant invasion success has been increasingly recognized (Callaway and Aschehoug, 2000; Inderjit and van der Putten, 2010; van der Putten et al., 2007; Wolfe and Klironomos, 2005). This has led to the formulation of two major hypotheses. The first hypothesis is the novel weapons hypothesis, which states that invasive plants release biochemical compounds, so-called allelopathic compounds, which are harmful to native

∗ Corresponding author. Tel.: +41 31 631 49 25. E-mail address: [email protected] (C. Del Fabbro).

species (Callaway and Ridenour, 2004; Rabotnov, 1981). This hypothesis assumes that allelopathic effects of invasive plant species are especially important in the invaded range as the recipient community’s species are not adapted to the biochemical compounds of plant invaders (Hierro and Callaway, 2003). To have a long-term effect, these allelopathic compounds must be persistent in the soil, resulting in a legacy effect (Kaur et al., 2009). Evidence for this mechanism relies on a few single species experiments. For instance, Centaurea diffusa and Ageratina adenophora exert stronger allelopathic effects on plant species from their invaded than from their native range (Callaway and Aschehoug, 2000; Inderjit et al., 2011). Another example is Alliaria petiolata exhibiting stronger allelopathic effects in the invaded than in the native range, both by harming other plants directly (Prati and Bossdorf, 2004) and indirectly by disrupting mycorrhizal associations in the invaded range (Callaway et al., 2008). Such biogeographical comparisons are important to understanding the role of co-evolution in plant invasions. However, this approach neglects the fact that native species might exert the same effects on other native species. Such a comparison is crucial to understand which processes contribute to the success of invasive over native

http://dx.doi.org/10.1016/j.ppees.2015.02.002 1433-8319/© 2015 Geobotanisches Institut ETH, Stiftung Ruebel. Published by Elsevier GmbH. All rights reserved.

88

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

species (Hamilton et al., 2005), but a systematic comparison of allelopathy in invasive and native species is still lacking. A second hypothesis is the accumulation of local pathogens, stating that invasive species alter the soil microbial community to the disadvantage of native species (Eppinga et al., 2006). This hypothesis makes the same prediction of outcome as the novel weapons hypothesis but assuming a different underlying mechanism. It is well known that plant species can influence the structure and composition of the soil microbial community, resulting in unique soil communities under different plant species (Bever et al., 1997; Bezemer et al., 2006; van der Putten et al., 2007). These soil communities may differ in density or composition of mutualistic or pathogenic microorganisms and thereby affect the performance of other plant species (Bever, 2003; Mangla et al., 2008). This idea has received a lot of attention in invasion ecology. However, evidence that invasive species accumulate pathogens that harm native species relies on only a few study systems (de la Pena et al., 2010; Mangla et al., 2008), not all of which show the same pattern (te Beest et al., 2009). A major task in ecology is to establish the degree of generality of a mechanism. Although details of the mechanisms differ between the two above mentioned hypotheses, they both predict that invasive plants affect the soil to the detriment of native species, either directly or indirectly. Furthermore, these mechanisms are not restricted to invasive species and may play important roles during range expansion and competition among native species. In invasion ecology, many studies focused either on the effect of a single invasive species, or the response of a single native species, although native species have been shown to differ in sensitivity to belowground alterations, both in terms of allelopathic compounds and the soil microbial community (Abhilasha et al., 2008; Gomez-Aparicio and Canham, 2008). Thus, the general importance of both the novel weapons and the accumulation of local pathogens hypotheses in explaining invasions remains unclear. To assess the generality of a hypothesis, meta-analytical approaches are often used, which combine different studies that often vary considerably in the details of methods (Kulmatiski et al., 2008). Multi-species experiments offer an alternative to meta-analyses by comparing the response of several species in a common experiment and thereby reducing the heterogeneity among studies commonly associated with meta-analyses (Schlaepfer et al., 2010). Furthermore, these experiments allow estimating the variation among species groups more accurately than meta-analyses as they are unaffected by publication bias. Most studies on mechanisms of plant invasions have focused on invasive plants without testing if native plants exert the same effects on other plants as invasives. Thus, one way to study the relative importance of allelopathy and accumulation of local pathogens is to compare the effect of invasive species with that of closely related native species. Even though closely related species may not always occupy the same habitat type or directly compete with each other, the strength of this method is that it accounts for phylogenetic interdependence among species (van Kleunen et al., 2010; Westoby et al., 1995). Here, we tested the common prediction of the novel weapons and the accumulation of local pathogen hypotheses, that invasive species generally create more negative soil conditions for native plants compared with their native congeners. Specifically, we asked the following questions: (1) Do invasive species generally exert a stronger allelopathic effect on native plants compared to the invasives’ native congeners? (2) Do invasive species generally promote a soil community more harmful to native plants compared with the invasives’ native congeners? (3) Do native species respond consistently to soil pre-cultivation with invasive and closely related native species?

To answer these questions, we used the soil pre-cultivation approach, which allows assessing the net effect of changes in the soil microbial community composition caused by plant species (Bever et al., 1997; Wolfe and Klironomos, 2005). We pre-cultivated soil with invasive and native species and then compared the performance of plants with that of plants in control soil that was not pre-cultivated. This soil pre-cultivation approach has been acknowledged as the most useful to investigate the role of the soil microbial community in plant invasion success (Wolfe and Klironomos, 2005). Additionally, we compared the effect of soil precultivation in sterilized and unsterilized soil to separate effects of soil microbial community composition and allelopathy. Methods Investigated plant species To test the role of the soil microbial community and allelopathy in plant invasion success, we pre-cultivated soil with 48 plant species, which we call cultivation species hereafter. Among these, 23 were invasive in Europe, 19 were closely related natives and 6 were closely related exotic species cultivated in gardens (Appendix Table S1). The 23 invasive species are established in more than 40% of the European countries (DAISIE, 2008) and most of them appear on the ‘black list’ of noxious plant invaders in Switzerland (CPS/SKEW, 2007). To account for possible phylogenetic effects, we selected for each invasive species a congeneric or confamiliar native species or a closely related non-invasive exotic garden species to compare their pre-cultivation effect. To test the effect of the cultivation species on the soil, we assessed germination and growth of four native species, which we call test species hereafter (Appendix Table S1). Our test species include three herbs (Campanula rotundifolia, Capsella bursapastoris, Daucus carota) and one grass (Poa annua), which all have a broad ecological range (Landolt, 2010). We chose test species with different phylogenetic backgrounds to test potential allelopathic and microbial effects of cultivation species. Moreover, test species are not closely related to any cultivation species. Furthermore, all test species often co-occur with our cultivation species, thus, representing species actually interacting with the cultivation species in the field. Soil pre-cultivation and sterilization For soil pre-cultivation, we grew the 48 cultivation species in 20 L pots in a greenhouse near Berne, Switzerland, from May to October 2010. Depending on the size of the species, we planted 1–5 individuals in each pot and replaced dead plants during the first 2 weeks. As a control, we placed pots containing only bare soil in the same greenhouse (non-cultivated soil). The substrate used (Ntotal : 1.5 g/kg, Ptotal : 40 mg/kg, Ktotal : 188 mg/kg) was a 3:1 mixture of sand and soil collected from an area that did not contain any plant species used in this study in Switzerland, the introduced range for the invaders. We replicated cultivation species and control pots 4 times each, resulting in a total pot number of (48 + 1) × 4 = 196. Pots were randomly arranged in the greenhouse and regularly watered with tap water. We harvested soil from all cultivation species and from the control at the end of October. The aboveground parts of plants were cut at soil level and soil was freed from roots by sieving with a 1 cm mesh. Above- and belowground biomass of cultivation species was dried at 80 ◦ C and weighed. Plants in four pots of three cultivation species died during soil pre-cultivation (Quercus palustris, Pseudotsuga menziesii, two Senecio jacobea) and soil was replaced

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

with that of the remaining replicates to achieve equal sample size for the following experiments. We kept the soils in plastic bags and cold-stored them for 4 weeks. To distinguish direct allelopathic effects of soil pre-cultivation from indirect effects mediated by changes in the soil microbial community, the soil from each pot including the control soil was split and one half was sterilized using ionizing radiation in dose of 40 kGy (LEONI Business Unit Irradiation Services, Däniken, Switzerland). This dose is twice as high as known to eliminate the majority of soil microorganisms (McNamara et al., 2003) and has only minimal effects on soil structure (e.g., Bank et al., 2008; Berns et al., 2008; Herbert et al., 2005). To ensure that sterilization was also effective in the center of the bags containing the soil, bags were aligned lengthways in order to reduce the distance of radiation waves to reach the center of soil bags. To minimize possible effects of sterilization on soil nutrient status (McNamara et al., 2003; Troelstra et al., 2001) and differences in nutrient uptake among the cultivation species, we fertilized the soils during the following experiments (for the actual amounts see below). We then compared the effects of soil pre-cultivation by invasive, native or non-native garden species and sterilization on germination and growth of four native test species in two experiments (Fig. 1). Germination experiment We tested the effects of soil pre-cultivation and sterilization on germination success of each of the four test species in a full-factorial experiment. There were four replicates per cultivation species and control soil, two sterilization treatments, four test species and two replicates per test species resulting in a total number of 3136 Petri dishes. The germination experiment was set up in a greenhouse in Bern, Switzerland, and ran for 3 weeks starting from 20th December 2010. Petri dishes (5.5 cm diameter) were filled with pre-cultivated soil and 20, 10, 15 and 30 seeds of Capsella, Poa, Daucus and Campanula were sown, respectively. We used different numbers of seeds based on differences in germination rate among the four test species (data not shown). Petri dishes were randomly placed in trays, except that Petri dishes with sterile and unsterilized soil were kept on different trays to avoid microbial contamination. Then, trays were randomly distributed in a greenhouse (day/night regime: 16/8, light regime: minimum 45 ␮mol m−2 s−1 , temperature regime: 20/18 ◦ C). Every second day, Petri dishes were re-randomized and watered with tap water. They were fertilized once with a modified Hoagland solution (per Petri dish: 2 mg N, 0.75 mg P, 3.25 mg K, and micronutrients in non-limiting amounts). Every 2 days we recorded germination, which we defined as cotyledons unfolded (herbs) or visible (grass). From these data, we calculated germination success as number of seedlings at harvest divided by the number of seeds sown and germination rate as the number of days until 50% of seedlings counted at harvest had emerged. After 3 weeks, the seedlings were harvested by test species and their biomass was dried at 80 ◦ C for at least 72 h and weighed. Growth experiment To test the effects of soil pre-cultivation and sterilization on plant growth, we conducted an experiment using the same design as the germination experiment, resulting in a total of 3136 pots. The growth experiment was set-up in a greenhouse near Berne, Switzerland, and run for 8–11 weeks starting from December 2010. Test species were germinated in sterilized soil and transplanted into 0.5 L pots (1 L for D. carota) containing either sterilized or unsterilized pre-cultivated or control soil. One week

89

after transplantation, 15 dead seedlings were replaced. Plants were kept in a greenhouse (day/night regime: 12/12, light: minimum 200 ␮mol m−2 s−1 , temperature regime: 12/24 ◦ C). Pots were placed on trays and randomized as in the germination experiment, except that pots were re-randomized every 2 weeks. Plants were watered as needed with tap water and fertilized once with a modified Hoagland solution (per plant: 4 mg N, 1.5 mg P, and 6.5 mg K). After 8, 9, 10 and 11 weeks, C. bursa-pastoris, P. annua, D. carota and C. rotundifolia were harvested, respectively. The aboveground biomass was clipped at soil level. Roots were gently washed and above- and belowground biomass was dried at 80 ◦ C for 72 h and weighed. Data analysis For both the germination and the growth experiment we tested for the absolute effect of soil sterilization with linear mixed effect models containing ‘soil pre-cultivation’ (i.e., precultivated vs. control soil), ‘sterilization’ and ‘test species’ as fixed factors and ‘cultivation species’, ‘cultivation pot’ and ‘cultivation pot × sterilization’ as a random effects to account for the split-plot design of our experiment. Three-way interactions were removed as these were never significant. If necessary, data were log- (seedling and root biomass) or square-root (total and aboveground biomass, germination success and germination rate) transformed to meet model assumptions. To compare the effect of soil pre-cultivation of different cultivation species we used a meta-analytical approach. We used the standardized mean difference Hedges’ g as effect size. It is calculated as the difference of the mean of an experimental and a control group divided by a pooled variance (Borenstein et al., 2009). We computed Hedges’ g for each combination of cultivation species, test species and sterilization treatment separately, using the respective combination of the control soil (i.e., soil without pre-cultivation) as control group. A main advantage of using Hedges’ g to analyze multispecies experiments is that different test species can be directly compared because their performance is assessed relative to their control. Moreover, the effect sizes of different cultivation species are easy to interpret: a Hedges’ g < 0 indicates a negative effect of soil pre-cultivation, while g > 0 indicates a positive and g = 0 no effect. Furthermore, the Hedges’ g exclusively shows the effect of the cultivation species on the response of the test species. Therefore, it eliminates possible side effects, for instance changes in nutrient availability by sterilization or microbial recruitment after sterilization during test species growth, because such effects are likely the same in soil with or without pre-cultivation. We deemed this approach as preferable to using mixed effect models as such analyses would have complicated the presentation and interpretation of results because we would have had to look at the higher-order interactions between treatment and soil pre-cultivation (pre-cultivated vs. control soil) and compare the effect in pre-cultivated and control soil. Effect sizes of all measured traits were then analyzed with mixed models to test the effects of plant origin (invasive, native or garden species), soil sterilization and test species as fixed factors. The factors ‘congeneric/confamiliar species’ and ‘cultivation species’ nested in ‘congeneric/confamiliar species’ were included as random terms in the model to account for possible phylogenetic effects and the experimental design. The effect of cultivation species identity on effect sizes was analyzed with fixed effects model ANOVAs with ‘cultivation species’, ‘sterilization’ and ‘test species’ as fixed factors. In all models, effect sizes were weighted inversely proportionally to their variance (Borenstein et al., 2009). Three-way interactions were removed as these were all non-significant. Furthermore, as different species may have depleted nutrients during

90

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

Fig. 1. Scheme of our experimental design showing 48 species (23 invasive, 19 native, 6 garden), which were used to pre-cultivate soil and bare soil as control. After ionic sterilization of half of the soil, the effect of soil pre-cultivation was tested on four native test species in a germination experiment and a growth experiment.

pre-cultivation differently, we included the biomass of cultivation species as a co-variable in the analyses. However, this effect was never significant, suggesting that the differential nutrient up-take by cultivation species was negligible, and we therefore excluded it from the models. To test if experimental and control group significantly differed, t-tests for effect sizes were conducted. For all figures and tables, model estimates and standard errors of effect sizes extracted from the models were used to display the results. All statistical analyses were performed in R 2.14.0 with the package ‘nlme’ (R Development Core Team, 2011).

Results Germination experiment Soil sterilization generally increased the performance of test species in the germination experiment (germination success: F1,193 = 64.53, P < 0.001; germination rate: F1,193 = 97.83, P < 0.001; seedling biomass: F1,193 = 37.04, P < 0.001). However, this effect also occurred in control soil, as there was no interaction between sterilization and pot (germination success: F1,193 = 0.01, P = 0.92; germination rate: F1,193 = 0.25, P = 0.62; seedling biomass: F1,193 = 0.38, P = 0.54; Fig. 2). The analysis of effect sizes of pre-cultivated vs. control soils showed that plant origin of the cultivation species (invasive vs. native vs. garden) did not affect test species performance in the germination experiment (Table 1). Within congeneric or con-familial cultivation species, test species did not respond consistently to plant origin (Appendix Fig. S1): some invasive species had a less positive (or more negative) effect on test species compared with their congeneric native, whereas others had the same or a more positive effect. Our results do not support the notion that

invasive species generally affect native species more negatively than congeneric native species in the early life stage. In unsterilized soil, soil pre-cultivation had no effect on germination success, a negative effect on germination rate and a positive effect on seedling biomass of the test species (Fig. 3). Soil sterilization did not influence the effect of soil pre-cultivation on germination success and germination rate, but eliminated the positive effect on seedling biomass (Table 1 and Fig. 3). Cultivation species affected germination rate and seedling biomass, but not germination success of test species differently (Appendix Table S2). The effect of soil pre-cultivation on seedling biomass of test species differed among cultivation species in both sterile (F47,141 = 2.07, P < 0.001) and unsterilized soil (F47,141 = 2.19, P < 0.001). The four native test species responded differently to soil precultivation (Table 1 and Fig. 4A). Campanula and Poa generally responded positively and negatively, respectively, and the response of Daucus and Campanula depended on sterilization (significant interaction of sterilization × test species for all variables, Table 1). The effect of soil pre-cultivation on germination of Campanula and Daucus was higher and lower in sterilized than in unsterilized soil, respectively, while the other species grew equally well in sterilized and unsterilized pre-cultivated soil.

Growth experiment Soil sterilization generally affected test species growth positively (total biomass: F1,194 = 146.53, P < 0.001; aboveground biomass: F1,194 = 271.54, P < 0.001; root biomass: F1,194 = 4.51, P = 0.03). Sterilization equally increased biomass of test species in control, i.e., not pre-cultivated, soil as indicated by the nonsignificant interaction of pot and sterilization (total biomass:

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

91

Table 1 Effect of status (invasive vs. native vs. garden), sterilization, test species identity and interactions of these factors on effect sizes (Hedges’ g) of traits measured in the germination and the growth experiment. Shown are degrees of freedom in the numerator (df1) and denominator (df2) as well as F-values and levels of significance (*P < 0.05, **P < 0.01, ***P < 0.001). Three-way interactions are not shown as they were never significant. df1

Intercept Status Sterilization Test species Status × sterilization Status × test species Sterilization × test species

1 2 1 3 2 6 3

df2

321 26 321 321 321 321 321

Germination experiment

Growth Experiment

Germination success

Germination rate

Seedling biomass

Total biomass

Aboveground biomass

Root biomass

0.2 0.3 0.5 93.2*** 1.8 0.5 21.8***

23.6*** 2.4 2.4 105.9*** 1.2 0.9 3.5*

30.6*** 0.2 14.3*** 188.7*** 2.0 2.1 38.6***

114.8*** 1.0 149.3*** 63.3*** 0.3 1.1 27.0***

60.6*** 0.7 233.7*** 47.0*** 0.2 1.1 17.9***

186.3*** 1.2 11.8*** 96.5*** 0.1 1.1 34.3***

the soil microbial community accumulated by plant species plays a major role in plant establishment success. As the effect of soil precultivation was not negative in sterilized soil, this indicates that allelopathic compounds were generally lacking. The four native test species differed in sensitivity to soil pre-cultivation and sterilization in terms of growth (Table 1). The biomass of Capsella and Poa was positively affected by soil pre-cultivation, whereas Campanula and Daucus did not grow differently in pre-cultivated and non-cultivated soil (Fig. 4B). Soil sterilization diminished the positive effect of soil pre-cultivation for Capsella and Poa, but did not affect the neutral pre-cultivation effect for Campanula and Daucus. Discussion The effect of soil sterilization

Fig. 2. The effect of soil sterilization on germination success (A) and total biomass (B) of test species grown in soil pre-cultivated with one of 48 cultivation species or control soil in the germination and growth experiment, respectively. Germination success was calculated as number of germinated seeds at the end of the experiment divided by total number of seeds supplied.

F1,194 = 1.15, P < 0.29; aboveground biomass: F1,194 = 2.75, P < 0.1; root biomass: F1,194 = 0.06, P = 0.81; Fig. 2). The analysis of effect sizes of pre-cultivated vs. control soils showed that invasive, native and garden species did not consistently differ in their cultivation effect in any trait measured in the growth experiment (Table 1 and Appendix Fig. S1). There was no interaction of plant origin with sterilization or test species for all responses. Thus, non-native plants were neither more allelopathic than natives, nor did they promote a more harmful soil biota. Soil pre-cultivation had a positive effect on test species performance (Fig. 3). This positive effect was stronger in unsterilized than in sterilized soil (Table 1). Cultivation species differed in their effect on test species growth (Appendix Table S2). This difference was less pronounced in sterilized than in non-sterilized soil for all responses. The difference in effect of soil pre-cultivation on plant performance in sterilized and unsterilized soil found suggests that

Soil sterilization positively affected test species performance both in pre-cultivated soil where the soil microbial community had been influenced by plant species and in control soil where soil microbes had not been cultured. This positive response to sterilization may be due to a general negative impact of the soil microbial community. During the growth process of the test species microbes were likely introduced in both pre-cultivated and control soil, either by air or with the seeds of test species. These microbes may have generally acted as pathogens and thus affected plant performance negatively. Alternatively, sterilization itself increased nutrient availability and changed soil structure, leading to a better performance of plants in sterilized soil (Troelstra et al., 2001). However, as we fertilized all soils we minimized effects of nutrient availability. Furthermore, we used ionic radiation for soil sterilization, which is known to keep possible side effects to a minimum (e.g., Bank et al., 2008; Berns et al., 2008; McNamara et al., 2003). Cultivation species changed the soil biota composition, but were generally not allelopathic Soil sterilization led to more similar performance of test species in pre-cultivated and control soil in terms of biomass. Even though we did not directly measure changes in soil biota because it would have been prohibitively expensive given our sample size, the response of the test species indicates that effects on the soil microbial community by plant species are generally more important for future plant growth than the exudation of allelopathic compounds. If the cultivation species were allelopathic, pre-cultivated soil would still have a negative effect compared to control soil. Our results suggest that allelopathic compounds of the cultivation species acted on germination rate, but this does not seem to translate into disadvantages in later growth. The soil microbial community promoted by cultivation species had variable effects on test species performance. The early stages

92

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

Fig. 3. Effect of soil pre-cultivation by different cultivation species in unsterilized and sterilized soil on germination success (A) and total biomass (B) in the germination and growth experiment, respectively. Shown are model estimates and standard errors of the effect sizes (Hedges’ g). If the effect size is larger than zero, the response is larger in soil pre-cultivated with the respective cultivation species than in control soil (zero line). Stars indicate significant differences between pre-cultivated and control soil (t-tests, P < 0.05). Numbers below bars indicate the different cultivation species: 1. Impatiens parviflora, 2. Amaranthus retroflexus, 3. Lonicera japonica, 4. Lupinus polyphyllus, 5. Impatiens glandulifera, 6. Ambrosia artemisiifolia (*), 7. Impatiens balfouri, 8. Heracleum mantegazzianum (*), 9. Prunus serotina, 10. Matricaria discoidea, 11. Acer negundo (*), 12. Quercus rubra, 13. Conyza canadensis, 14. Robinia pseudoacacia (*), 15. Mahonia aquifolium (*), 16. Buddleja davidii, 17. Bunias orientalis, 18. Pseudotsuga menziesii, 19. Solidago gigantea (*), 20. Fallopia japonica, 21. Solidago canadensis, 22. Populus × canadensis, 23. Senecio inaequidens (*), 24. Impatiens noli-tangere, 25. Berberis vulgaris (*), 26. Senecio jacobea, 27. Heracleum sphondylium, 28. Solidago virgaurea, 29. Acer platanoides, 30. Populus tremula, 31. Vicia cracca (*), 32. Prunus padus, 33. Senecio vernalis (*), 34. Polygonum bistorta (*), 35. Quercus robur, 36. Lonicera xylosteum, 37. Inula conyza, 38. Picea abies (*), 39. Sinapis arvensis (*), 40. Laburnum anagyroides (*), 41. Artemisia vulgaris (*), 42. Matricaria recutita, 43. Impatiens walleriana, 44. Berberis thunbergii (*), 45. Amaranthus tricolor (*), 46. Quercus palustris, 47. Artemisia ludoviciana, 48. Acer saccharinum (*).

of plant life cycles were not differently affected by the soil biota in pre-cultivated and control soil, whereas biomass was generally higher when the soil biota was promoted by the cultivation species. This pattern concurs with the observation that Carpobrotus edulis affected growth of one, but not germination of two other species via changes in soil microbial community composition (de

la Pena et al., 2010). In contrast to our test species, biomass of a test species was negatively influenced by the soil microbial community pre-cultivated by C. edulis. The importance of the soil microbial community for plant growth has been demonstrated before (e.g., De Deyn et al., 2004; Wardle et al., 2004), but this is the first report of multiple plant species consistently promoting soil biota,

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

93

Fig. 4. Effect of soil pre-cultivation in unsterilized and sterilized soil on germination success (A) and total biomass (B) of the four test species across all cultivation species in the germination and the growth experiment, respectively. Shown are model estimates and standard errors of the effect sizes (Hedges’ g). If the effect size is larger than zero, the response is larger in pre-cultivated soil than in control soil. Stars above bars indicate significant differences between pre-cultivated and control soil (t-tests); stars above lines indicate significant differences between sterilized and unsterilized soil according to Tukey post hoc test (*P < 0.05, **P < 0.01, ***P < 0.001).

which is less harmful to other species than soil biota of unconditioned soil. This indicates that plant species generally shift the soil microbial community composition to a more mutualistic or less pathogenic stage compared to uncultivated soil. However, as we only investigated the effect of heterospecific, not conspecific, soil pre-cultivation on test species growth, this does not suggest that plants only promote mutualists. Invasive species did not exhibit stronger allelopathic effects than their native congeners Our results do not support the hypothesis that invasive plant species have stronger persistent allelopathic effects on native plants compared with their native congeners, as predicted by the novel weapons hypothesis (Callaway and Ridenour, 2004). This hypothesis predicts a negative effect of soil pre-cultivation by

invasive species and a less negative or even positive effect of native congeners on test species, at least in sterilized soil. However, in both unsterilized and sterilized soil the effect of soil pre-cultivation on biomass fractions and germination success was generally neutral to positive for all plants, independent of their origin. This indicates that allelopathic compounds reducing the growth of other native plant species were generally lacking for all plant origins. Alternatively, the pre-cultivation period of 5 months was too short for allelopathic compounds to build up sufficiently in the soil. However, when we harvested the cultivation species, their root systems were well established throughout the pots, which should have enabled them to exude chemical compounds in large amount into the soil. It is also possible that potential allelopathic compounds produced during soil pre-cultivation were not persistent enough to affect test species growth (Kaur et al., 2009). Alternatively, common species as used in our study may be insensitive to allelopathic

94

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95

compounds, although other studies showed that common species do respond to allelopathic compounds (e.g., Abhilasha et al., 2008; Scharfy et al., 2011). Moreover, in our experiment, germination rate was strongly negatively affected by soil pre-cultivation in sterilized soil, indicating that our test species actually responded to allelopathic compounds. Nevertheless, this effect did not differ between invasive, garden and native cultivation species and did not affect later plant life stages. Some studies have found support for the novel weapons hypothesis (Callaway and Aschehoug, 2000; Kim and Lee, 2011; Prati and Bossdorf, 2004; Scharfy et al., 2011), but there are also studies that have not (Blair et al., 2006; Del Fabbro et al., 2014; Dostal, 2011; Lind and Parker, 2010), thus the generality of this hypothesis appears in doubt. We tested one prediction of the novel weapon hypothesis more closely approximating field conditions than most previous studies attempted and found no evidence for it. Rather, our results indicate that even though allelopathy might explain the success of some invasive species and possibly give invaders an advantage in the short term, it does not enable invasive species to persistently affect native species in the long term. Our conclusion may be restricted to the particular soil type, high nutrient conditions, and the specific greenhouse environment of this experiment. We have chosen nutrient-rich conditions because most invasive plant species in Europe invade such habitats (Chytry et al., 2008). However, it would be worthwhile to test whether the absence of a general negative effect of invasive species holds true under a broader range of environmental conditions, including, for instance, nutrient-poor soils.

Invasive species did not promote a more harmful soil biota than their native congeners Our results do not support the hypothesis that invasive plants promote a soil microbial community more harmful to native plants compared with their native congeners (Eppinga et al., 2006; Mangla et al., 2008: accumulation of local pathogens hypothesis). In all experiments, invasive, native and garden congeners did not seem to consistently differ in their effect on soil properties important for the performance of all test species in both sterilized and unsterilized soil. However, as we did not directly measure soil biota, there could be hidden effects. For instance, invasive and native species may very well cultivate different microbial species. Nevertheless, our results indicate that even if there are such differences, the net effects of invasive and native species on others are similar. It does not necessarily follow that invasive species may not generally profit from the soil biota compared to natives. If invasive species experience a more positive or less negative conspecific than heterospecific soil feedback they would experience a positive net soil feedback. Nevertheless, our results indicate that plant origin does not seem to be a key factor in determining the promotion of a soil biota detrimental or beneficial to other plants in general. Despite the non-significance of plant origin, cultivation species identity was important for test species growth in both experiments, although this effect was strongly diminished in sterilized soil. This implies that cultivation species accumulated soil microbial communities that differed in either density or composition of generalistic mutualists or pathogens (Bever et al., 1997) and confirms previous observations of the uniqueness of soil microbial communities establishing under different plant species (Bever et al., 1996; Bezemer et al., 2006; Grayston et al., 1998; Kardol et al., 2007; Kowalchuk et al., 2002). Growth form of cultivation species influenced test species growth as well. We found that soil microbes accumulated by tree species had a more positive effect on test species than herb and shrub species in the growth experiment

(F2,17 = 5.44, P = 0.01), extending the observation that grasses and forbs accumulate different soil microbes (Bezemer et al., 2006). Test species show different sensitivity to soil pre-cultivation Our four test species showed large differences in both magnitude and direction of their response to soil pre-cultivation and sterilization. Surprisingly, none of our test species showed a general sensitivity to soil pre-cultivation or the soil microbial community. Rather, if soil pre-cultivation or sterilization did not affect the biomass of one test species, they affected its germination success and vice versa. Our results suggest that responses of test species to soil pre-cultivation are highly species specific. However, as we only used four test species, future studies should include a larger set of test species in order to explain the variation in response among them. Conclusions In general, invasive species do not create more negative soil conditions for other plants than related native species. Thus, plant origin is neither a key factor for the accumulation of persistent allelopathic compounds, nor for the accumulation of a soil microbial community beneficial or detrimental to other plants. Plant species seem to have a larger effect on the soil microbial community than on abiotic soil properties. This alteration is species specific and affects other species unequally, potentially changing community composition. The strength of our experiment is that we compared the effect and response of a large set of plant species, which enables us to derive general conclusions. However, the ability to generalize can only be achieved at the cost of precision (Levins, 1966). Thus, further studies should focus on investigating the mechanisms of the soil biota responses found in our study. Acknowledgments We would like to thank the many students for help during the experiments, Stephanie Frei for help with counting germinated seedlings and members of the Fischer lab for helpful comments on earlier versions of the manuscript. Financial support was provided be the Swiss National Science Foundation SNSF (grant no. 31003A 127561 to Daniel Prati). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ppees.2015.02.002. References Abhilasha, D., Quintana, N., Vivanco, J., Joshi, J., 2008. Do allelopathic compounds in invasive Solidago canadensis s.l. restrain the native European flora? J. Ecol. 96, 993–1001. Bank, T.L., Kukkadapu, R.K., Madden, A.S., Ginder-Vogel, M.A., Baldwin, M.E., Jardine, P.M., 2008. Effects of gamma-sterilization on the physico-chemical properties of natural sediments. Chem. Geol. 251, 1–7. Berns, A.E., Philipp, H., Narres, H.D., Burauel, P., Vereecken, H., Tappe, W., 2008. Effect of gamma-sterilization and autoclaving on soil organic matter structure as studied by solid state NMR, UV and fluorescence spectroscopy. Eur. J. Soil Sci. 59, 540–550. Bever, J.D., 2003. Soil community feedback and the coexistence of competitors: conceptual frameworks and empirical tests. New Phytol. 157, 465–473. Bever, J.D., Morton, J.B., Antonovics, J., Schultz, P.A., 1996. Host-dependent sporulation and species diversity of arbuscular mycorrhizal fungi in a mown grassland. J. Ecol. 84, 71–82. Bever, J.D., Westover, K.M., Antonovics, J., 1997. Incorporating the soil community into plant population dynamics: the utility of the feedback approach. J. Ecol. 85, 561–573.

C. Del Fabbro, D. Prati / Perspectives in Plant Ecology, Evolution and Systematics 17 (2015) 87–95 Bezemer, T.M., Lawson, C.S., Hedlund, K., Edwards, A.R., Brook, A.J., Igual, J.M., Mortimer, S.R., Van der Putten, W.H., 2006. Plant species and functional group effects on abiotic and microbial soil properties and plant–soil feedback responses in two grasslands. J. Ecol. 94, 893–904. Blair, A.C., Nissen, S.J., Brunk, G.R., Hufbauer, R.A., 2006. A Lack of evidence for an ecological role of the putative allelochemical (±)-catechin in spotted knapweed invasion success. J. Chem. Ecol. 32, 2327–2331. Blossey, B., Notzold, R., 1995. Evolution of increased competitive ability in invasive nonindigenous plants – a hypothesis. J. Ecol. 83, 887–889. Borenstein, M., Hedges, L.V., Higgins, J.P.T., Rothstein, H.R., 2009. Introduction to Meta-Analysis. John Wiley & Sons, Chichester, UK. Callaway, R.M., Aschehoug, E.T., 2000. Invasive plants versus their new and old neighbors: a mechanism for exotic invasion. Science 290, 521–523. Callaway, R.M., Cipollini, D., Barto, K., Thelen, G.C., Hallett, S.G., Prati, D., Stinson, K., Klironomos, J., 2008. Novel weapons: invasive plant suppresses fungal mutualists in America but not in its native Europe. Ecology 89, 1043–1055. Callaway, R.M., Ridenour, W.M., 2004. Novel weapons: invasive success and the evolution of increased competitive ability. Front. Ecol. Environ. 2, 436–443. Chytry, M., Maskell, L.C., Pino, J., Pysek, P., Vila, M., Font, X., Smart, S.M., 2008. Habitat invasions by alien plants: a quantitative comparison among Mediterranean, subcontinental and oceanic regions of Europe. J. Appl. Ecol. 45, 448–458. CPS/SKEW, 2007. The Swiss Commission for Wild Plant Conservation. Black List and Watch List, Available from: http://www.cps-skew.ch/english/ invasive alien plants/black list watch list.html (accessed 03.01.12). DAISIE, 2008. European Invasive Alien Species Gateway, Available from: http://www.europe-aliens.org (accessed 02.01.12). Davis, M.A., Grime, J.P., Thompson, K., 2000. Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol. 88, 528–534. De Deyn, G.B., Raaijmakers, C.E., Van der Putten, W.H., 2004. Plant community development is affected by nutrients and soil biota. J. Ecol. 92, 824–834. de la Pena, E., de Clercq, N., Bonte, D., Roiloa, S., Rodriguez-Echeverria, S., Freitas, H., 2010. Plant–soil feedback as a mechanism of invasion by Carpobrotus edulis. Biol. Invas. 12, 3637–3648. Del Fabbro, C., Gusewell, S., Prati, D., 2014. Allelopathic effects of three plant invaders on germination of native species: a field study. Biol. Invas., 1035–1042, http://dx.doi.org/10.1007/s10530-013-0555-3. Dostal, P., 2011. Plant competitive interactions and invasiveness: searching for the effects of phylogenetic relatedness and origin on competition intensity. Am. Nat. 177, 655–667. Eppinga, M.B., Rietkerk, M., Dekker, S.C., De Ruiter, P.C., Van der Putten, W.H., 2006. Accumulation of local pathogens: a new hypothesis to explain exotic plant invasions. Oikos 114, 168–176. Gomez-Aparicio, L., Canham, C.D., 2008. Neighbourhood analyses of the allelopathic effects of the invasive tree Ailanthus altissima in temperate forests. J. Ecol. 96, 447–458. Grayston, S.J., Wang, S.Q., Campbell, C.D., Edwards, A.C., 1998. Selective influence of plant species on microbial diversity in the rhizosphere. Soil Biol. Biochem. 30, 369–378. Hamilton, M.A., Murray, B.R., Cadotte, M.W., Hose, G.C., Baker, A.C., Harris, C.J., Licari, D., 2005. Life-history correlates of plant invasiveness at regional and continental scales. Ecol. Lett. 8, 1066–1074. Herbert, R.B., Malmstrom, M., Ebena, G., Salmon, U., Ferrow, E., Fuchs, M., 2005. Quantification of abiotic reaction rates in mine tailings: evaluation of treatment methods for eliminating iron- and sulfur-oxidizing bacteria. Environ. Sci. Technol. 39, 770–777. Hierro, J.L., Callaway, R.M., 2003. Allelopathy and exotic plant invasion. Plant Soil 256, 29–39. Inderjit, Evans, H., Crocoll, C., Bajpai, D., Kaur, R., Feng, Y.L., Silva, C., Carreon, J.T., Valiente-Banuet, A., Gershenzon, J., Callaway, R.M., 2011. Volatile chemicals from leaf litter are associated with invasiveness of a Neotropical weed in Asia. Ecology 92, 316–324. Inderjit and van der Putten, W.H., 2010. Impacts of soil microbial communities on exotic plant invasions. Trends Ecol. Evol. 25, 512–519.

95

Kardol, P., Cornips, N.J., van Kempen, M.M.L., Bakx-Schotman, J.M.T., van der Putten, W.H., 2007. Microbe-mediated plant–soil feedback causes historical contingency effects in plant community assembly. Ecol. Monogr. 77, 147–162. Kaur, H., Kaur, R., Kaur, S., Baldwin, I.T., Inderjit, 2009. Taking ecological function seriously: soil microbial communities can obviate allelopathic effects of released metabolites. PLoS ONE 4. Keane, R.M., Crawley, M.J., 2002. Exotic plant invasions and the enemy release hypothesis. Trends Ecol. Evol. 17, 164–170. Kim, Y.O., Lee, E.J., 2011. Comparison of phenolic compounds and the effects of invasive and native species in East Asia: support for the novel weapons hypothesis. Ecol. Res. 26, 87–94. Kowalchuk, G.A., Buma, D.S., de Boer, W., Klinkhamer, P.G.L., van Veen, J.A., 2002. Effects of above-ground plant species composition and diversity on the diversity of soil-borne microorganisms. Anton. Leeuw. Int. J. G. 81, 509–520. Kulmatiski, A., Beard, K.H., Stevens, J.R., Cobbold, S.M., 2008. Plant–soil feedbacks: a meta-analytical review. Ecol. Lett. 11, 980–992. Landolt, E., 2010. Flora Indicativa: Ecological Indicator Values and Biological Attributes of the Flora of Switzerland and the Alps. Haupt Verlag, Bern, CH. Levins, R., 1966. Strategy of model building in population biology. Am. Sci. 54, 421–431. Lind, E.M., Parker, J.D., 2010. Novel weapons testing: are invasive plants more chemically defended than native plants? PLoS ONE 5. Mangla, S., Inderjit, Callaway, R.M., 2008. Exotic invasive plant accumulates native soil pathogens which inhibit native plants. J. Ecol. 96, 58–67. McNamara, N.P., Black, H.I.J., Beresford, N.A., Parekh, N.R., 2003. Effects of acute gamma irradiation on chemical, physical and biological properties of soils. Appl. Soil Ecol. 24, 117–132. Mitchell, C.E., Agrawal, A.A., Bever, J.D., Gilbert, G.S., Hufbauer, R.A., Klironomos, J.N., Maron, J.L., Morris, W.F., Parker, I.M., Power, A.G., Seabloom, E.W., Torchin, M.E., Vazquez, D.P., 2006. Biotic interactions and plant invasions. Ecol. Lett. 9, 726–740. Prati, D., Bossdorf, O., 2004. Allelopathic inhibition of germination by Alliaria petiolata (Brassicaceae). Am. J. Bot. 91, 285–288. R Development Core Team (2011). R: A language and environmentfor statistical computing. Vienna, Austria. ISBN: 3-900051-07-0, available online at http://www.R-project.org/ Rabotnov, T.A., 1981. Importance of the evolutionary approach to the study of allelopathy. Sov. J. Ecol.+ 12, 127–130. Scharfy, D., Funk, A., Venterink, H.O., Gusewell, S., 2011. Invasive forbs differ functionally from native graminoids, but are similar to native forbs. New Phytol. 189, 818–828. Schlaepfer, D.R., Glattli, M., Fischer, M., van Kleunen, M., 2010. A multi-species experiment in their native range indicates pre-adaptation of invasive alien plant species. New Phytol. 185, 1087–1099. te Beest, M., Stevens, N., Olff, H., van der Putten, W.H., 2009. Plant–soil feedback induces shifts in biomass allocation in the invasive plant Chromolaena odorata. J. Ecol. 97, 1281–1290. Troelstra, S.R., Wagenaar, R., Smant, W., Peters, B.A.M., 2001. Interpretation of bioassays in the study of interactions between soil organisms and plants: involvement of nutrient factors. New Phytol. 150, 697–706. van der Putten, W.H., Klironomos, J.N., Wardle, D.A., 2007. Microbial ecology of biological invasions. ISME J. 1, 28–37. van Kleunen, M., Dawson, W., Schlaepfer, D., Jeschke, J.M., Fischer, M., 2010. Are invaders different? A conceptual framework of comparative approaches for assessing determinants of invasiveness. Ecol. Lett. 13, 947–958. Wardle, D.A., Bardgett, R.D., Klironomos, J.N., Setala, H., van der Putten, W.H., Wall, D.H., 2004. Ecological linkages between aboveground and belowground biota. Science 304, 1629–1633. Westoby, M., Leishman, M.R., Lord, J.M., 1995. On misinterpreting the phylogenetic correction. J. Ecol. 83, 531–534. Wolfe, B.E., Klironomos, J.N., 2005. Breaking new ground: soil communities and exotic plant invasion. Bioscience 55, 477–487.