Artificial particles and soil communities interactively change heterospecific plant-soil feedbacks

doi:10.21203/rs.3.rs-4918961/v1

Background and aims

Microplastics affect plant growth and change abiotic and biotic soil properties, such as soil structure or soil-community composition. However, how microplastics affect plant-soil interactions, such as plant-soil feedbacks (PSFs), is still poorly understood. Here, we tested how artificial particles affect heterospecific PSFs, depending on an intact or depleted soil community.

Methods

We conducted a two-phase-greenhouse experiment using Centaurea jacea to condition soil containing an intact or initially depleted soil community in the first phase. Subsequently, we grew individuals of Crepis biennis and Eragrostis minor in all combinations of soil conditioning, soil-community status, and different material treatments including no particle addition, glass particles, or three microplastics individually and mixed. Effects of soil community, material treatment and their interaction on PSFs were assessed based on plant biomass and rootmorphology traits.

Results

Particles in general, microplastics and glass, increased PSF strength based on plant biomass. PSFs tended to be negative with the intact but positive with the initially depleted soil community. Overall, particle-addition effects on PSFs were stronger in the initially depleted community, indicating interactive effects of artificial particles in the soil and soil biota. Interactive particle and soil-community effects generally depended on material type and concentration.

Conclusion

Our findings indicate that artificial particles can affect heterospecific PSFs, and that these effects are likely to be partly mediated by the soil community. Further, they highlight the need for studies assessing potential ecological implications of microplastics modifying plant-soil interactions.

Microplastic

plant-soil feedback

plastic pollution

plant-soil interactions

Plastic pollution is a global issue receiving increasing social, political, and scientific attention (Bank et al., 2021; MacLeod et al., 2021; Rochman et al., 2013). The potential scale of this problem is obvious when considering that in 2021 almost 400 million tons of new plastics were produced, in Europe alone (Plastics Europe, 2022). Due to insufficient recycling strategies and improper disposal, ever increasing quantities of plastic waste are accumulating in the environment (Geyer et al., 2017; Jambeck et al., 2015). Plastic waste is slowly fragmented into smaller particles called microplastics, generally defined as particles < 5 mm (Thompson et al., 2009). Microplastics might be particularly problematic for the environment, but also for humans (MacLeod et al., 2021; Vethaak & Legler, 2021). Although research on microplastics was mainly limited to marine environments before (Andrady, 2011; Cole et al., 2011), research focusing on terrestrial ecosystems is increasing (Baho et al., 2021; Rillig, 2012; Rochman, 2018). Soils are the major sink for microplastics, making the plant-soil system especially prone to potential microplastic effects (Chia et al., 2021; Helmberger et al., 2020; Rillig & Lehmann, 2020; Zhou et al., 2021). Microplastics can affect plant functioning and alter soil communities (Speißer & van Kleunen, 2023; Sun et al., 2022), but impacts on plant-soil interactions are still poorly understood.

Plants change abiotic and biotic components of the soil they grow in, in turn affecting current plants but also subsequent ones, a process known as plant-soil feedback (PSF) (Bever, 1994; Ehrenfeld et al., 2005; Wardle, 2002). Such PSFs can either be conspecific, if the conditioning and the affected plants belong to the same species, or heterospecific, if the plants belong to different species (van der Putten et al., 2013). Generally, PSFs are driven by plants changing abiotic conditions, including nutrient availability and other soil properties, and biotic soil conditions, including the abundance of mutualists and antagonists (Bennett & Klironomos, 2019). As effects of such changes range from positive (improved plant performance) to negative (reduced plant performance), PSFs play important roles in ecological processes such as community assembly and succession, and plant invasion (Bennett et al., 2017; Chapin et al., 1994; Chen & van Kleunen, 2022; Klironomos, 2002; Van der Putten et al., 1993). However, human-driven global change is likely to affect PSFs (van der Putten et al., 2016), and microplastics in particular, due to effects on plant functioning, physical soil properties and soil communities are likely to interact with PSFs, but this interaction remains to be tested.

By now, it has become clear that microplastics can affect both plant and soil properties. Microplastics change the soil structure, which can result in reduced water availability and increased nutrient leaching (Ingraffia et al., 2022a; Ingraffia et al., 2022b; Kim et al., 2021b; Speißer & van Kleunen, 2023; Wan et al., 2019). Regarding possible alterations of PSFs, microplastic effects on soil communities might be of particular importance. Plastic particles in the soil can create a new type of habitat for microorganisms called the plastisphere, which is the environment and microbial community under direct influence of microplastics (Rillig et al., 2023). Microbial communities associated with microplastics might contain more pathogenic bacteria and fungi (Gkoutselis et al., 2021; Zhu et al., 2021), which could lead to more negative PSFs. On the other hand, there could be positive effects on mutualistic arbuscular mycorrhizal fungi (AMF), reducing negative PSF effects (Lehmann et al., 2020). Overall, experimental evidence for microplastic effects on AMF is very limited and effects are likely to be highly context dependent (Wang et al., 2020). However, although main biotic drivers of PSF, pathogenic bacteria and fungi and mutualistic AMF, are likely to, or have been shown to be affected by microplastics, studies directly testing microplastic effects on PSFs are missing, so far (but see Lozano & Rillig, 2022).

Microplastics can affect plants both directly, due to phytotoxic effects (Maity & Pramanick, 2020; Pignattelli et al., 2020) and indirectly, by changing soil characteristics, including biogeochemical cycling and nutrient availability (Ingraffia et al., 2022b; Leifheit et al., 2021a). Most likely, both direct and indirect mechanisms are responsible for observed effects of microplastics on plant performance and root growth (Lozano et al., 2021; Speißer & van Kleunen, 2023; van Kleunen et al., 2020). Moreover, effects of microplastics on plants could feed back to plant interactions with soils. For example, microplastics can change plant secondary metabolites, in turn affecting soil nematodes (Kim et al., 2021a). A first study looking at effects of microplastics on PSFs found that soils previously exposed to microplastics exhibited more positive or negative PSFs compared to soils without microplastic ancestry, depending on microplastic characteristics (Lozano & Rillig, 2022), emphasizing the need for more studies investigating microplastic effects on PSFs.

Microplastics affect key drivers of PSFs (plants conditioning the soil and soil communities) in various ways, making predictions about microplastic effects on PSFs difficult. To gain a better understanding of how microplastics affect heterospecific PSFs, we tested the hypothesis that microplastics affect PSFs depending on the plastic type and particle concentrations, by assessing how different microplastics affected soil-conditioning effects of Centaurea jacea on the growth of the forb Crepis biennis and the grass Eragrostis minor. Next to a control treatment without added particles and a chemically inert control using glass granules, we included three plastic types, i.e. low-density polyethylene (PE), ethylene propylene diene monomer (EPDM), and polyhydroxyalkanoate (PHA). We applied the different plastics individually and mixed in a low (0.5% vol) and a high (5% vol) concentration. We expected the effects of microplastics on PSFs to be mediated by both abiotic (e.g. soil structure, nutrient availability) and biotic (soil community) factors. Therefore, to disentangle biotic from abiotic effects, we grew C. jacea in substrate containing either an intact or depleted (sterilized) soil community. In the feedback phase, we grew individuals of C. biennis and E. minor in all possible combinations of conditioning, soil community, material type, and material concentration. Finally, due to the closer relatedness to C. jacea, we expected PSF effects and microplastic-induced changes to be stronger for C. biennis, especially with the intact soil community.

Study species and precultivation:

We used the perennial forb Centaurea jacea L. (Asteraceae) to condition the substrate in the first phase of the experiment (conditioning phase; Fig. 1). Centaurea jacea is native to Europe and is naturalized in other parts of the world, including North America and Australia (POWO, 2024). We chose this species because it has been shown to induce detectable PSF effects on conspecifics and heterospecifics (Xue et al., 2018a; Xue et al., 2018b). To examine heterospecific PSFs of C. jacea, and assess whether these are modified by microplastics, we used the distantly related Eragrostis minor Host (Poaceae) and the more closely related Crepis biennis L. (Asteraceae) as test species in the second phase of the experiment (feedback phase). Eragrostis minor is an annual grass that is native to large parts of Asia, the Mediterranean and parts of Africa, and is naturalized in Central Europe, Australia, and parts of North and South America (POWO, 2024). Crepis biennis is a biennial forb native to large parts of Europe and occurs in parts of North America as naturalized species (POWO, 2024). Seeds were sown separately by species into plastic trays (13.4 cm × 12.2 cm × 4.9 cm; TEKU® TK 1214, Pöppelmann GmbH & Co. KG, Lohne, Germany) filled with unsterilized potting soil (Einheitserde® CL P, Einheitserdewerke Werkverband e.V., Sinntal-Altengronau, Germany) and placed in a climatized greenhouse for germination two weeks before the beginning of the respective phase. That is, seeds of C. jacea were sown on 15 August 2022, and seeds of E. minor and C. biennis were sown on 10 October 2022. Seeds of C. jacea and C. biennis were obtained from a commercial seed company (Rieger-Hofmann GmbH, Blaufelden-Raboldshausen, Germany), and E. minor seeds were obtained from the botanical garden of the University of Konstanz.

Microplastics:

To test how different plastic types affect PSFs, we selected three plastic types differing in major characteristics, such as polymer type, surface structure and degradability. Polyethylene (PE) is the most commonly used plastic type (Geyer et al., 2017), and is characterized by a smooth surface and high resistance to biotic and abiotic degradation. In our experiment, we used granules (2.5–4 mm) of a non-additivated low-density PE (Lupolen 3020H, LyondellBasell Industries, Rotterdam, Netherlands), which is used to produce, amongst others, plastic bags, and food packaging. As second conventional plastic type, we selected ethylene propylene diene monomer (EPDM). EPDM is an elastomer that is often used for outdoor applications, due to its high resistance against UV-degradation and abrasion. For example, EPDM granules are frequently used in artificial turfs (e.g. soccer pitches), from which they can easily spread into the surroundings (van Kleunen et al., 2020). These EPDM granules are a good example of primary microplastics. Here, we used such EPDM granules (0.5–2.5 mm; Resedagrün RAL 6011, GranuElastic Höfer & Stankowska GbR, Frankfurt (Oder), Germany). As the usage of biodegradable plastic types is increasing continuously (Kumar et al., 2023), we also included the biodegradable polymer polyhydroxyalkanoate (PHA) in our experiment (PHI 002, NaturePlast, Mondeville, France). The PHA granules (2–4 mm) were similar to the PE granules in terms of their appearance. However, PHA is considered to degrade relatively quickly (Dilkes-Hoffman et al., 2019), and hence stronger short-term effects can be expected compared to conventional plastics (Qi et al., 2018). In addition to the different microplastics, to account for purely physical effects of adding particles to the substrate, we included glass granules (2–4 mm, Glasgranulat klar, Deco Stones, Vechelde, Germany) as a chemically inert control. To remove glass dust and other impurities, the glass granules were thoroughly washed prior to usage.

Conditioning phase:

For the conditioning phase, we created a substrate consisting of sand-vermiculite (1:1, v:v) mixed with field soil in a 1:1 volume-ratio, to include a natural soil community. The field soil was collected from the topsoil layer of a grassland community close to the botanical garden of the University of Konstanz (N: 47°69′19.56″, E: 9°17′78.42″) and was sieved (mesh size 15 mm × 15 mm) to remove stones and break up large soil aggregates. To be able to investigate the interactive effects of the soil community and our other treatments (soil conditioning, microplastics), we sterilized half of the substrate twice using a steam sterilizer, heating the substrate to 80°C for approximately three hours (Erddämpfer Sterilo, Harter Elektrotechnik, Schenkenzell, Germany). Subsequently, we filled 3-L square pots with the respective substrates, i.e. either with pure sterilized or non-sterilized substrate for the control pots or with sterilized or non-sterilized substrate containing a low (0.5%, v:v) or high (5%, v:v) concentration of PE, EPDM, PHA, a mix of all three plastics (equal volumetric proportions), or glass. To ensure accurate concentrations of the microplastic and glass particles, we prepared the substrate for each pot individually. As soils are recolonized after sterilization (Baweja, 1939; Li et al., 2019; Marschner & Rumberger, 2004), we refer to an initially depleted vs intact soil community, rather than sterilized vs intact.

On 29 August 2022, to condition the substrate, we planted four similar sized seedlings of C. jacea per pot into half of the pots, keeping the other half of the pots as unconditioned control. Subsequently, we placed the 220 pots (2 conditioning × 2 soil communities × (((4 plastics + glass) × 2 concentrations) + 1 control) × 5 replicates) in the same greenhouse compartment described above in a randomized block design. Within the five blocks, the pots were randomly assigned to fixed positions. On 24 October 2022, after a growth period of eight weeks, we harvested the plants by first cutting the shoots and subsequently carefully removing the roots from the substrate. We collected the shoots and roots from each pot individually, dried them at 70°C for at least 72 hours and weighed them using a digital scale. We report the effects of microplastics and soil community (depleted vs. intact) on C. jacea in the supplement (Supplementary Tables 1 & 2; Supplementary Fig. 1)

Feedback phase:

We thoroughly homogenized the substrate of each individual conditioning-phase 3-L pot, and subsequently redistributed it into two new 1-L pots. Into one of these two pots, we planted one seedling of C. biennis, and, into the other one, we planted a seedling of E. minor. This resulted in a total of 440 pots for the feedback phase (i.e., 220 pots per test species; Fig. 1), which again were arranged in the same greenhouse into five randomized replicate blocks. To be able to account for differences in initial size of the seedlings in the statistical analysis, for each seedling, we measured the length and width of the largest leaf, counted the number of leaves, and calculated a proxy of initial leaf area by multiplying the leaf length by the width of the largest leaf and by the number of leaves. After a growth period of ten weeks, we harvested both the aboveground and belowground parts of the plants, individually. The shoots of both species, as well as the roots of E. minor were directly dried at 70°C for at least 72 hours and subsequently weighed using a digital scale. For C. biennis, we stored the fresh roots in water-filled tubes at 8°C for a maximum of 72 hours prior to root-morphology analysis. For the root-morphology analysis of C. biennis, we scanned the individual root systems using a root scanner (modified Epson Expression 1100 XL and Epson Expression 1200 XL flatbed scanners) and analyzed the total root length, average root diameter, average link length (as a proxy for root ramification) and root volume (to calculate root-tissue density), using the WinRhizo™ Pro imaging software (Regent Instruments Inc., Canada). After scanning, the roots were dried and weighed as described above. As the root systems of E. minor were very dense and convoluted, we analyzed root-morphology traits only for C. biennis.

Statistical analysis:

All statistical analyses were performed in R 4.2.2 (R Core Team, 2022).

PSFs:

To test how the three different types of microplastics in different concentrations affect heterospecific plant-soil feedbacks (PSF), and how potential effects might be influenced by soil-community depletion, we first calculated PSF values as log-response ratios:

$$\:{PSF}_{X}=\text{l}\text{n}\left(\frac{{X}_{conditioned}}{\stackrel{-}{{X}_{unconditioned}}}\right)$$

That is, the PSF value for a given trait (X) was calculated as the natural logarithm of the quotient of the individual sample trait value from the conditioned soil and the mean of the control samples from the unconditioned soil (grouped by material treatment and soil community). For both response species, C. biennis and E. minor, we calculated PSFs for total biomass (PSF_biomass) and the proportion of root biomass to total biomass, i.e. root-weight ratio (PSF_RWR). For C. biennis, we also calculated PSFs for the root-morphology traits average root diameter (PSF_RD), specific root length (PSF_SRL), average link length (PSF_LL), and root-tissue density (PSF_RTD). We then fitted linear mixed-effects models using the lme function of the “nlme” package (Pinheiro et al., 2021), including PSF as response variable. We included material treatment (the one control and 10 combinations of material type and concentration), soil community (intact vs. initially depleted) and their interaction as fixed effects, and block as random effect. To improve homoscedasticity, we added variance structures for soil community for the PSF_biomass model of E. minor and for all C. biennis models except for the PSF_LL and PSF_RTD models, using the varIdent fuction in the “nlme” package. For the PSF_RWR model of E. minor, adding variance structures for material treatment resulted in the best model fit (based on AIC). Significance of fixed effects was assessed using log-likelihood-ratio tests (Zuur et al., 2009).

Contrast models:

To obtain more detailed information about differences in effects of the different material types and concentrations, we created orthogonal contrasts comparing specific material-treatment combinations (Supplementary Table 3). The first contrast tested the effect of particle addition by comparing the average of the grouped material treatments (low and high concentrations of glass and microplastics) to the control (no added material). The second contrast tested the overall effects of plastics by comparing the average of the grouped microplastics to the average of low and high glass. The third contrast tested the effect of mixing the plastic types by comparing the average of the low and high microplastic mix to the average of the grouped individual microplastics. The fourth contrast tested the effect of the biodegradable plastic by comparing the average of the low and high biodegradable PHA to the average of the grouped conventional plastics (EPDM and PE). The fifth contrast tested whether the two conventional plastics had different effects by comparing the average of the low and high EPDM to the average of the low and high PE. The sixth to tenth contrasts compared the high vs. low concentration of each individual material type. We then replaced the material treatment with the ten a priori chosen contrasts as explanatory variables in the linear mixed-effects models, together with soil community and the specific two-way interactions between soil community and each contrast.

Given the long-standing discussion about strictly binary decisions based on arbitrary p-value thresholds, we followed the recommendations of Muff et al. (2021), and wrote the results in a gradual evidence language. All results refer to effects on the calculated plant-soil feedback (PSF) values for the respective traits.

Heterospecific PSF effects on Crepis biennis:

Biomass responses:

PSF_biomass and PSF_RWR of C. biennis ranged from negative to positive values, and there was very strong evidence that both were interactively affected by the material treatment (combinations of different material types and concentrations) and the soil-community treatment (intact vs. initially depleted) (p < 0.001, respectively; Table 1, Figure 2). Overall, PSF_biomass tended to be positive for the initially depleted soil community, but negative for the intact soil community, and the material treatment further modified this pattern (Figure 2A). Based on the a priori chosen contrasts comparing specific material-treatment combinations, we identified different combinations driving these interactions (Table 2). For the initially depleted soil community, there was very strong evidence that particle addition generally resulted in a shift from a negative to a positive PSF_biomass (C1 × soil community: p < 0.001; particles absent: -0.67 ± 0.6, particles present: 1.14 ± 0.18). In contrast, for the intact soil community, PSF_biomass was generally negative, but more so if particles were added (particles absent: -0.30 ± 0.13, particles present: -0.49 ± 0.05; Figure 2A). Further, there was very strong evidence that the biodegradable microplastics led to a weaker positive PSF_biomass for the initially depleted community (C4 × soil community: p < 0.001; conventional: 1.44 ± 0.28, degradable: 0.16 ± 0.19), and a weaker negative PSF_biomass for the intact community (conventional: -0.52 ± 0.08, degradable: -0.37 ± 0.08). There was also strong evidence for concentration effects for glass and EPDM. For the initially depleted community, PSF_biomass was generally positive but less so for the high concentrations (C6 × soil community: p < 0.001; glass low: 2.1 ± 0.18, glass high: 0.29 ± 0.76; C7 × soil community: p = 0.002; EPDM low: 2.75 ± 0.16, EPDM high: 0.89 ± 0.68). For the intact community, PSF_biomass was generally negative but weaker for the high concentration (glass low: -0.46 ± 0.08, glass high: -0.4 ± 0.11; EPDM low: -0.71 ± 0.14, EPDM high: -0.31 ± 0.11). In addition, there was strong evidence for a concentration dependency in the plastic mix, with a negative PSF_biomass for the low concentration but a positive PSF_biomass for the high concentration (C10: p < 0.01; low: -0.24 ± 0.34, high: 0.84 ± 0.45), irrespective of the soil community.

Similar to the results for PSF_biomass, there was very strong evidence that particle addition led to a shift from negative to positive PSF_RWR (i.e. from a negative to a positive conditioning effect on biomass allocation to roots) for the initially depleted community (C1 × soil community: p < 0.001; particles absent: -0.45 ± 0.19, particles present: 0.15 ± 0.05). For the intact community, PSF_RWR was generally weak, but tended to be slightly positive without added particles (particles absent: 0.05 ± 0.09) and slightly negative with added particles (particles present: -0.07 ± 0.03; Figure 2B). In addition, there was very strong evidence that microplastics, compared to glass, led to a weaker positive PSF_RWR for the initially depleted community (C2 × soil community: p < 0.001; glass: 0.39 ± 0.05, microplastics: 0.08 ± 0.05), and a weaker negative PSF_RWR for the intact community (glass: -0.12 ± 0.07, microplastics: -0.06 ± 0.03). Further there was moderate evidence that the high EPDM concentration reduced PSF_RWR strength (from positive to neutral) for the initially depleted community (C7 × soil community: p = 0.03; low: 0.35 ± 0.04, high: 0.004 ± 0.22), and for the intact community (low: -0.07 ± 0.15, high: -0.03 ± 0.08). There was also moderate evidence that the high glass concentration led to a more positive PSF_RWR compared to the low concentration (C6: p = 0.02; low: 0.02 ± 0.10, high: 0.26 ± 0.09), irrespective of the soil community.

Root-morphology responses:

For all root traits of C. biennis, except for average link length, we found very strong evidence that PSF responses were interactively affected by material and soil-community treatments (Table 1, Supplementary Table 4). Overall, PSF_SRL tended to be negative when the soil community had initially been depleted, whereas it was frequently neutral or positive when the soil community was intact, although there were some exceptions to this (Figure 3A). Again, based on the contrast models, we found different specific combinations driving those interactions (Table 2). For the initially depleted community, we found very strong evidence that microplastic particles in the soil led to a less negative PSF_SRL compared to glass particles (C2 × soil community: p < 0.001; glass: ‑1.04 ± 0.14, microplastics: ‑0.54 ± 0.08). For the intact community, PSF_SRL tended to be positive in general, but less so in the presence of microplastics (glass: 0.19 ± 0.09; microplastics: 0.08 ± 0.05). Further, we found very strong evidence for EPDM (C7 × soil community: p < 0.001) and moderate evidence for PHA (C9 × soil community: p = 0.01) that low and high concentrations differed in their effects on PSF_SRL in interaction with the soil community. For the combination of EPDM and the initially depleted community, PSF_SRL was negative for the low concentration (-0.97 ± 0.11) but turned positive for the high concentration (0.18 ± 0.25). In contrast, in combination with the intact community, PSF_SRL was positive for the low EPDM concentration (0.41 ± 0.17) but rather neutral for the high concentration (0.03 ± 0.07). For PHA, the pattern was the other way around. Although PSF_SRL tended to be generally negative, for the initially depleted community, the high PHA concentration resulted in a more negative PSF_SRL (low: -0.32 ± 0.09, high: -0.90 ± 0.25). For the intact community, PSF_SRL tended to be negative for the low but rather neutral for the high concentration (low: -0.17 ± 0.06, high: -0.02 ± 0.15). In addition, we found strong evidence that, irrespective of the soil community, the biodegradable PHA led to a stronger negative PSF_SRL compared to the conventional plastics (C4: p < 0.01; conventional: -0.11 ± 0.09, degradable: -0.35 ± 0.10). As effects on SRL were generally inversely related to root thickness and root-tissue density, detailed results for PSF_RD and PSF_RTD are provided in the supplement (Supplementary Results, Supplementary Tables 4 & 5, Supplementary Figure 2).

We found moderate evidence that material treatment (p < 0.05), and very strong evidence that soil community (p < 0.001) independently affected PSF_LL (Table 1). Overall, PSF_LL tended to be slightly positive for the intact soil community (0.07 ± 0.03) but negative for the initially depleted community (-2.1 ± 0.03). Regarding the material effects, we found strong evidence that PSF_LL was negative for EPDM (‑0.16 ± 0.07) but neutral for PE (0.001 ± 0.05; C5: p < 0.01). In addition, there was moderate evidence that the high concentration of the mixed microplastics led to a shift from a slightly positive to a negative PSF_LL (C10: p < 0.05; low: 0.09 ± 0.08, high: -0.17 ± 0.06; Figure 3B).

Heterospecific PSF-effects on Eragrostis minor:

Biomass responses:

We found strong evidence that the material treatment affected PSF_biomass of E. minor (p < 0.01), but no evidence for soil-community effects (p = 0.4; Table 3). Based on the a priori chosen contrasts (Table 4), we found strong evidence that particles in general led to a more negative PSF_biomass (C1: p < 0.01; particles absent: -0.1 ± 0.1, particles present: -0.42 ± 0.04), and that the high PHA concentration led to a less negative PSF_biomass than the low PHA concentration (C9: p < 0.001; low: -0.71 ± 0.18, high: ‑0.17 ± 0.06; Figure 4A).

We found strong evidence that the material treatment and the soil community interactively affected PSF_RWR of E. minor (p = 0.01; Table 3). Overall, PSF_RWR tended to be weaker for the intact community treatment compared to the depleted one, but this was not true for all material-treatment combinations (Figure 4B). Based on the contrast models, we found very strong evidence that microplastics led to a less negative PSF_RWR with the initially depleted community (C2 × soil community: p < 0.001; glass: -0.26 ± 0.08, microplastic: -0.04 ± 0.05), and a weaker PSF_RWR with the intact community, compared to glass (glass: 0.12 ± 0.06, microplastic: -0.01 ± 0.05). In addition, there was moderate evidence that the high PHA concentration led to a weaker positive PSF_RWR with the initially depleted community (C9 × soil community: p < 0.05; low: 0.19 ± 0.16, high: 0.03 ± 0.09), and a shift from negative to positive PSF_RWR for the intact community (low: ‑0.18 ± 0.07, high: 0.16 ± 0.13; Figure 4B), compared to the low concentration. Further, there was moderate to strong evidence that different material-treatment groups differed from each other, irrespective of the initial state of the soil community. Particle addition in general tended to shift the direction of PSF_RWR from positive to slightly negative (C1: p < 0.05; particles absent: 0.09 ± 0.06, particles present: -0.04 ± 0.03). Also, PSF_RWR was negative with EPDM (-1.16 ± 0.06) but positive and overall weaker with PE (0.09 ± 0.07; C5: p < 0.01). In addition, PSF_RWR tended to be slightly positive with the low glass concentration (0.04 ± 0.08) but turned negative for the high concentration (-0.18 ± 0.09; C6: p < 0.05).

We tested how different types of microplastics, in combination with an intact or initially depleted soil community, affect heterospecific plant-soil feedbacks (PSFs; Fig. 1). We show that, for Crepis biennis, the direction of PSF strongly depended on the status of the soil community, with an intact community generally resulting in negative PSF and the initially depleted community rather leading to positive PSF (Fig. 2), indicating a positive effect of soil recolonization (Li et al., 2019). However, soil-community status during the conditioning phase was less influential regarding PSF on Eragrostis minor (Table 3). Interestingly, we found that particle addition in general (glass and microplastics) intensified PSFs based on total biomass for both species, although the overall direction was generally positive for C. biennis and negative for E. minor. Compared to glass particles, microplastics led to weaker PSFs based on root-weight ratio (RWR) and root-morphological traits, although there were some exceptions (Figs. 2, 3, 4). In most cases, particle addition and soil-community effects depended on each other, suggesting that artificial particles in the soil are likely to change heterospecific PSFs interactively with the soil community.

Despite effects on PSF, there was no evidence for overall particle-addition effects on plant productivity independent of soil conditioning or initial state of the soil community, neither for C. biennis nor E. minor (Supplementary Fig. 3). This indicates that particle-addition effects are generally highly context dependent (Krehl et al., 2022). In contrast, there was very strong evidence for material-treatment effects, in addition to but irrespective of soil-community effects, on both productivity and RWR of C. jacea in the conditioning phase (Supplementary Table 1, Supplementary Fig. 1). This suggests that particle addition could modify PSFs by changing the growth of conditioning plants. Regarding the PSF responses, when looking at the treatment combination most closely resembling natural conditions without pollution, i.e. no added particles and intact soil community, we found only moderate evidence for heterospecific PSF effects on C. biennis biomass (-0.33 ± 0.16, p = 0.049) and on RWR of E. minor (0.19 ± 0.09, p = 0.038), indicating that PSF effects were rather weak, overall. However, PSF effects changed with particle addition, and depended on whether the soil community was intact or initially depleted (Tables 1 & 3).

In line with our expectation, we found clear evidence that microplastics affected heterospecific PSFs, depending on plastic type and concentration. Moreover, this was true for particle presence in general. Glass and microplastic particles strengthened PSF_biomass of both response species, inducing a positive feedback in C. biennis but a negative one in E. minor. The similar pattern for PSF_RWR (see C1 in Figs. 2, 4A) suggests that effects on PSF_biomass might be strongly driven by changes in root biomass. That is, root biomass increased for C. biennis and decreased for E. minor, when grown in conditioned soil with artificial particles. The changed PSFs indicate that artificial particles in the soil in general, and not just microplastics, can affect PSFs. Such effects might arise because artificial particles change physical soil properties, affecting both plants and soil communities and, in turn, PSFs. For example, microplastics affect soil structure, bulk density, water flow and water holding capacity (de Souza Machado et al., 2019; de Souza Machado et al., 2018; Kim et al., 2021b). Such changes can result in a higher water evaporation and reduced soil moisture (Speißer & van Kleunen, 2023; Wan et al., 2019). Consequently, changed physical soil properties might affect plants and soil organisms directly (Krehl et al., 2022; Leifheit et al., 2021b), but also their interactions by, for example, changing root morphology or rhizosphere properties (de Souza Machado et al., 2019; Speißer & van Kleunen, 2023). Both root morphology and rhizosphere properties are key determinants of PSFs (Kuzyakov & Blagodatskaya, 2015; Kuzyakov & Razavi, 2019; Wilschut et al., 2019), so changing those properties could be one way in which artificial particles alter PSFs.

The different effects of glass and microplastics (see C2 in Fig. 2B, 3A, 4B) indicate that chemical components are also likely to be involved in how artificial particles affect PSFs. Considering the high number and variety of additives in many plastics (Jones, 2024; Wagner et al., 2024), this appears to be a plausible factor. In our study, however, EPDM was the only plastic type containing additives, and previous studies showed that its effects are dose-dependent and can affect root morphology (Speißer & van Kleunen, 2023; van Kleunen et al., 2020). At the same time, it was the plastic type showing the strongest concentration dependent effects for PSFs based on biomass and root traits (see C7 in Fig. 2, 3A, Supplementary Fig. 2), supporting the assumption that additives could be one factor of how microplastics change PSFs. However, considering the clear differences in effects between glass and all grouped microplastics, it seems unlikely that EPDM was the only driver of these different effects. The difference between conventional and biodegradable microplastics on PSF_biomass of C. biennis (see C4 in Fig. 2A) and PSF_RWR for both C. biennis and E. minor (Tables 2 & 4), and the concentration effects of PHA on PSF_SRL of C. biennis and PSF_biomass and PSF_RWR of E. minor (see C9 in Figs. 3A, 4) point towards another potential component.

Plastics consist mainly of carbon chains, which might serve as additional carbon source for microorganisms, and can change soil microbial composition and activity (Cao et al., 2023; Fei et al., 2020; Rong et al., 2021; Zheng et al., 2005). Altered microbial composition or activity could affect PSFs directly, but also indirectly by influencing plants, leading to follow-up effects on PSFs. Generally, the accessibility of carbon from plastics strongly depends on the plastic type (Zheng et al., 2005). PHA, as biodegradable plastic, can be degraded relatively quickly, up to approximately 0.1 mg×day^− 1×cm^− 2 (Dilkes-Hoffman et al., 2019), and many bacteria and fungi are able to degrade PHA (Jendrossek & Handrick, 2002). Additionally, despite the initial inertness of many plastics, aging due to environmental factors (e.g. UV, heat, mechanical abrasion) can induce physical and chemical changes making plastics more prone to further degradation (Shah et al., 2008). So, more carbon (and other compounds) could be released from plastics, in the long term. Consequently, aging might lead to stronger microplastic effects in the environment (Lozano et al., 2023; Speißer, 2023), which should also be considered for future investigations regarding PSFs.

Importantly, in most cases, particle effects on PSFs depended on whether the soil in the conditioning phase contained an intact or depleted soil community, which is in line with our second expectation that biotic factors can mediate particle effects. Interactive effects of microplastics and soil organisms are in line with previous findings that microplastics can affect soil microbial composition and activity (Fei et al., 2020; Rong et al., 2021). For C. biennis, with the initially depleted soil community, adding particles led to a shift from negative to positive PSF_biomass. In contrast, with the intact community, PSF_biomass was generally negative, but more so if particles were added to the soil (see C1 in Fig. 2A). The positive PSF_biomass for the combination of initially depleted community and particle addition is likely to be mainly driven by changes in root properties of C. biennis. The positive PSF_RWR (Fig. 2B) together with the positive PSF_RTD and negative PSF_SRL (Supplementary Fig. 2B, Fig. 3A) indicate that the plants did not just produce proportionally more roots, but also denser and heavier roots in conditioned soil with the initially depleted community and added particles (Supplementary Figs. 4, 5). Moreover, compared to glass, microplastics led to weaker PSF_RWR (see C2 in Fig. 2B), PSF_SRL and PSF_RTD (see C2 in Fig. 3A and Supplementary Fig. 2B), especially with the depleted community. So, overall, the plants invested more in roots in the conditioned soil with initially depleted soil community, but less so when microplastics were present instead of glass (Supplementary Figs. 6, 7). The differences between glass and microplastics could be explained by differing effects on soil communities due to distinct material properties. Indeed, microplastics can be associated with less diverse microbial communities, of specific taxonomical and functional composition (Luo et al., 2022; Shi et al., 2022; Sun et al., 2022; Zhang et al., 2019; Zhu et al., 2021). A potential explanation why these effects were stronger for the depleted soil community could be that the community structure in the intact community was more stable, so changes in composition and the resulting effects were less pronounced. However, as we did not analyze the community composition, future studies should clarify this.

When comparing the PSF responses of C. biennis and E. minor, mediating effects of the soil community were less pronounced for E. minor. There was no evidence that the initial state of the community affected PSF_biomass of E. minor directly or via modifying microplastic effects (Table 3, Fig. 4A), matching our third expectation that interactive effects between particle addition and soil community are stronger for C. biennis. However, PSF responses of both species generally differed, and E. minor experienced more negative PSF_biomass than C. biennis (-0.26 ± 0.16, -0.08 ± 0.41, respectively). This is in line with previous findings that short-lived species are more prone to negative PSFs (Kardol et al., 2006; Lemmermeyer et al., 2015; Xi et al., 2021). Accordingly, C. biennis, although just being biennial, had a clearly higher RWR than the annual E. minor (0.46 ± 0.01, 0.09 ± 0.002), matching previous findings that species with low relative root weight experience more negative PSFs (Wilschut et al., 2023). So, while the generally negative PSF_biomass of E. minor could probably be explained by the low RWR and other root traits associated with fast-growing plants (Wilschut et al., 2023; Xi et al., 2021), the pattern for C. biennis was more complex as PSF_biomass also depended on the initial state of the soil community (Fig. 2A).

With the intact soil community, PSF_biomass of C. biennis was overall negative but was positive for the initially depleted community (-0.39 ± 0.09, 0.23 ± 0.91, respectively). This pattern could reflect that generalists are responsible for the negative PSFs (Semchenko et al., 2022; Wilschut et al., 2023; Wilschut et al., 2019), which might have dominated the intact community. In contrast, with the initially depleted community, conditioning by C. jacea could have led to a less diverse community with perhaps less total, or mainly specialist pathogens. Taken on its own, this does not explain the shift to a positive PSF_biomass or why PSF responses generally tended to be stronger with the depleted community. However, soil sterilization itself can have positive effects on plant performance, when followed by recolonization of beneficial soil organisms (Li et al., 2019), although such effects might change over time (Marschner & Rumberger, 2004). Further, our analysis showed that soil-community effects also depended on particle addition, suggesting that artificial particles in the soil might further modify effects of differing soil communities on PSFs.

In conclusion, our study showed that artificial particles in the soil can affect heterospecific PSFs, and that these effects are likely to be partly mediated by soil-community composition. However, as we did not analyze the soil-community composition, further investigations are needed to gain a better understanding of the interplay of artificial particles and soil biota on PSFs. Our results also suggest that changes in PSFs are shaped by both physical and chemical mechanisms. In this context, future studies should also consider that effects of microplastics might change over time, which could have further implications for the plant-soil system. Our findings add evidence to first findings that microplastics could change plant-soil feedbacks (Lozano & Rillig, 2022) but also highlight the complexity of the mechanisms involved and the need for further studies assessing potential ecological implications of microplastics modifying plant-soil interactions.

AMF: Arbuscular mycorrhizal fungi

EPDM: Ethylene propylene diene monomer

LL: Average link length

PE: Polyethylene

PHA: Polyhydroxyalkanoate

PSF(s): Plant-soil feedback(s):

RD: Average root diameter

RTD: Root-tissue density

RWR: Root-weight ratio

SRL: Specific root length

Competing interests

The authors declare no competing interests.

Author contributions

BS conceived the idea. BS, MvK and RAW designed the experiment. SG performed the experiment. BS analyzed the data with additional inputs by MvK and RAW. BS wrote the manuscript with considerable contributions from MvK, RAW and SG.

Acknowledgements

We thank Kristina Schlötter, Otmar Ficht and Heinz Vahlenkamp for practical assistance and Rhea DeStefano and Sven Egger for their help during the experiment. We thank Yanjie Liu for reviewing the manuscript prior to submission.

Data availability

Upon acceptance, the data will be archived in a public repository and the associated DOI will be included in this article.

Andrady, A. L. (2011). Microplastics in the marine environment. Marine Pollution Bulletin, 62(8), 1596-1605. https://doi.org/10.1016/j.marpolbul.2011.05.030
Baho, D. L., Bundschuh, M., & Futter, M. N. (2021). Microplastics in terrestrial ecosystems: Moving beyond the state of the art to minimize the risk of ecological surprise. Global Change Biology, 27(17), 3969-3986. https://doi.org/https://doi.org/10.1111/gcb.15724
Bank, M. S., Swarzenski, P. W., Duarte, C. M., Rillig, M. C., Koelmans, A. A., Metian, M., et al. (2021). Global Plastic Pollution Observation System to Aid Policy. Environmental Science & Technology, 55(12), 7770-7775. https://doi.org/10.1021/acs.est.1c00818
Baweja, K. D. (1939). Studies of the Soil Fauna, with Special Reference to the Recolonization of Sterilized Soil. Journal of Animal Ecology, 8(1), 120-161. https://doi.org/10.2307/1257
Bennett, J. A., & Klironomos, J. (2019). Mechanisms of plant–soil feedback: interactions among biotic and abiotic drivers. New Phytologist, 222(1), 91-96. https://doi.org/https://doi.org/10.1111/nph.15603
Bennett, J. A., Maherali, H., Reinhart, K. O., Lekberg, Y., Hart, M. M., & Klironomos, J. (2017). Plant-soil feedbacks and mycorrhizal type influence temperate forest population dynamics. Science, 355(6321), 181-184. https://doi.org/doi:10.1126/science.aai8212
Bever, J. D. (1994). Feeback between Plants and Their Soil Communities in an Old Field Community. Ecology, 75(7), 1965-1977. https://doi.org/https://doi.org/10.2307/1941601
Cao, J., Li, C., Gao, X., Cai, Y., Song, X., Siddique, K. H. M., et al. (2023). Agricultural soil plastic as a hidden carbon source stimulates microbial activity and increases carbon dioxide emissions. Resources, Conservation and Recycling, 198, 107151. https://doi.org/https://doi.org/10.1016/j.resconrec.2023.107151
Chapin, F. S., Walker, L. R., Fastie, C. L., & Sharman, L. C. (1994). Mechanisms of Primary Succession Following Deglaciation at Glacier Bay, Alaska. Ecological Monographs, 64(2), 149-175. https://doi.org/https://doi.org/10.2307/2937039
Chen, D., & van Kleunen, M. (2022). Invasional meltdown mediated by plant–soil feedbacks may depend on community diversity. New Phytologist, 235(4), 1589-1598. https://doi.org/https://doi.org/10.1111/nph.18218
Chia, R. W., Lee, J.-Y., Kim, H., & Jang, J. (2021). Microplastic pollution in soil and groundwater: a review. Environmental Chemistry Letters. https://doi.org/10.1007/s10311-021-01297-6
Cole, M., Lindeque, P., Halsband, C., & Galloway, T. S. (2011). Microplastics as contaminants in the marine environment: A review. Marine Pollution Bulletin, 62(12), 2588-2597. https://doi.org/10.1016/j.marpolbul.2011.09.025
de Souza Machado, A. A., Lau, C. W., Kloas, W., Bergmann, J., Bachelier, J. B., Faltin, E., et al. (2019). Microplastics can change soil properties and affect plant performance. Environmental Science & Technology, 53(10), 6044-6052. https://doi.org/10.1021/acs.est.9b01339
de Souza Machado, A. A., Lau, C. W., Till, J., Kloas, W., Lehmann, A., Becker, R., et al. (2018). Impacts of microplastics on the soil biophysical environment. Environmental Science & Technology, 52(17), 9656-9665. https://doi.org/10.1021/acs.est.8b02212
Dilkes-Hoffman, L. S., Lant, P. A., Laycock, B., & Pratt, S. (2019). The rate of biodegradation of PHA bioplastics in the marine environment: A meta-study. Marine Pollution Bulletin, 142, 15-24. https://doi.org/https://doi.org/10.1016/j.marpolbul.2019.03.020
Ehrenfeld, J. G., Ravit, B., & Elgersma, K. (2005). Feedback in the plant-soil system. Annual Review of Environment and Resources, 30(1), 75-115. https://doi.org/10.1146/annurev.energy.30.050504.144212
Fei, Y., Huang, S., Zhang, H., Tong, Y., Wen, D., Xia, X., et al. (2020). Response of soil enzyme activities and bacterial communities to the accumulation of microplastics in an acid cropped soil. Science of The Total Environment, 707, 135634. https://doi.org/https://doi.org/10.1016/j.scitotenv.2019.135634
Geyer, R., Jambeck, J. R., & Law, K. L. (2017). Production, use, and fate of all plastics ever made. Science Advances, 3(7), e1700782. https://doi.org/doi:10.1126/sciadv.1700782
Gkoutselis, G., Rohrbach, S., Harjes, J., Obst, M., Brachmann, A., Horn, M. A., et al. (2021). Microplastics accumulate fungal pathogens in terrestrial ecosystems. Scientific Reports, 11(1), 13214. https://doi.org/10.1038/s41598-021-92405-7
Helmberger, M. S., Tiemann, L. K., & Grieshop, M. J. (2020). Towards an ecology of soil microplastics. Functional Ecology, 34(3), 550-560. https://doi.org/https://doi.org/10.1111/1365-2435.13495
Ingraffia, R., Amato, G., Bagarello, V., Carollo, F. G., Giambalvo, D., Iovino, M., et al. (2022a). Polyester microplastic fibers affect soil physical properties and erosion as a function of soil type. SOIL, 8(1), 421-435. https://doi.org/10.5194/soil-8-421-2022
Ingraffia, R., Amato, G., Iovino, M., Rillig, M., Giambalvo, D., & Frenda, A. S. (2022b). Polyester microplastic fibers in soil increase nitrogen loss via leaching and decrease plant biomass production and N uptake. Environmental Research Letters. https://doi.org/10.1088/1748-9326/ac652d
Jambeck, J. R., Geyer, R., Wilcox, C., Siegler, T. R., Perryman, M., Andrady, A., et al. (2015). Plastic waste inputs from land into the ocean. Science, 347(6223), 768-771. https://doi.org/doi:10.1126/science.1260352
Jendrossek, D., & Handrick, R. (2002). Microbial Degradation of Polyhydroxyalkanoates. Annual Review of Microbiology, 56(Volume 56, 2002), 403-432. https://doi.org/https://doi.org/10.1146/annurev.micro.56.012302.160838
Jones, N. (2024). More than 4,000 plastic chemicals are hazardous, report finds
Year-long effort compiles comprehensive database of chemicals in plastics. Nature. https://doi.org/https://doi.org/10.1038/d41586-024-00805-2
Kardol, P., Martijn Bezemer, T., & van der Putten, W. H. (2006). Temporal variation in plant–soil feedback controls succession. Ecology Letters, 9(9), 1080-1088. https://doi.org/https://doi.org/10.1111/j.1461-0248.2006.00953.x
Kim, S. W., Liang, Y., Lozano, Y. M., & Rillig, M. C. (2021a). Microplastics Reduce the Negative Effects of Litter-Derived Plant Secondary Metabolites on Nematodes in Soil [Original Research]. Frontiers in Environmental Science, 9(560). https://doi.org/10.3389/fenvs.2021.790560
Kim, S. W., Liang, Y., Zhao, T., & Rillig, M. C. (2021b). Indirect Effects of Microplastic-Contaminated Soils on Adjacent Soil Layers: Vertical Changes in Soil Physical Structure and Water Flow. Frontiers in Environmental Science, 9(152). https://doi.org/10.3389/fenvs.2021.681934
Klironomos, J. N. (2002). Feedback with soil biota contributes to plant rarity and invasiveness in communities. Nature, 417(6884), 67-70. https://doi.org/10.1038/417067a
Krehl, A., Schöllkopf, U., Májeková, M., Tielbörger, K., & Tomiolo, S. (2022). Effects of plastic fragments on plant performance are mediated by soil properties and drought. Scientific Reports, 12(1), 17771. https://doi.org/10.1038/s41598-022-22270-5
Kumar, R., Sadeghi, K., Jang, J., & Seo, J. (2023). Mechanical, chemical, and bio-recycling of biodegradable plastics: A review. Science of The Total Environment, 882, 163446. https://doi.org/https://doi.org/10.1016/j.scitotenv.2023.163446
Kuzyakov, Y., & Blagodatskaya, E. (2015). Microbial hotspots and hot moments in soil: Concept & review. Soil Biology and Biochemistry, 83, 184-199. https://doi.org/https://doi.org/10.1016/j.soilbio.2015.01.025
Kuzyakov, Y., & Razavi, B. S. (2019). Rhizosphere size and shape: Temporal dynamics and spatial stationarity. Soil Biology and Biochemistry, 135, 343-360. https://doi.org/https://doi.org/10.1016/j.soilbio.2019.05.011
Lehmann, A., Leifheit, E. F., Feng, L., Bergmann, J., Wulf, A., & Rillig, M. C. (2020). Microplastic fiber and drought effects on plants and soil are only slightly modified by arbuscular mycorrhizal fungi. Soil Ecology Letters. https://doi.org/10.1007/s42832-020-0060-4
Leifheit, E. F., Kissener, H. L., Faltin, E., Ryo, M., & Rillig, M. C. (2021a). Tire abrasion particles negatively affect plant growth even at low concentrations and alter soil biogeochemical cycling. Soil Ecology Letters. https://doi.org/10.1007/s42832-021-0114-2
Leifheit, E. F., Lehmann, A., & Rillig, M. C. (2021b). Potential Effects of Microplastic on Arbuscular Mycorrhizal Fungi. Frontiers in Plant Science, 12, Article 626709. https://doi.org/10.3389/fpls.2021.626709
Lemmermeyer, S., Lörcher, L., Kleunen, M. v., & Dawson, W. (2015). Testing the Plant Growth-Defense Hypothesis Belowground: Do Faster-Growing Herbaceous Plant Species Suffer More Negative Effects from Soil Biota than Slower-Growing Ones? The American Naturalist, 186(2), 264-271. https://doi.org/10.1086/682005
Li, K., DiLegge, M. J., Minas, I. S., Hamm, A., Manter, D., & Vivanco, J. M. (2019). Soil sterilization leads to re-colonization of a healthier rhizosphere microbiome. Rhizosphere, 12, 100176. https://doi.org/https://doi.org/10.1016/j.rhisph.2019.100176
López-Angulo, J., de la Cruz, M., Chacón-Labella, J., Illuminati, A., Matesanz, S., Pescador, D. S., et al. (2020). The role of root community attributes in predicting soil fungal and bacterial community patterns. New Phytologist, 228(3), 1070-1082. https://doi.org/https://doi.org/10.1111/nph.16754
Lozano, Y. M., Gordillo, H., Waldmann, W., & Rillig, M. C. (2023). Photodegradation modifies microplastic effects on soil properties and plant performance. Journal of Applied Ecology.
Lozano, Y. M., Lehnert, T., Linck, L. T., Lehmann, A., & Rillig, M. C. (2021). Microplastic Shape, Polymer Type, and Concentration Affect Soil Properties and Plant Biomass [Original Research]. Frontiers in Plant Science, 12. https://doi.org/10.3389/fpls.2021.616645
Lozano, Y. M., & Rillig, M. C. (2022). Legacy effect of microplastics on plant–soil feedbacks [Original Research]. Frontiers in Plant Science, 13. https://doi.org/10.3389/fpls.2022.965576
Luo, G., Jin, T., Zhang, H., Peng, J., Zuo, N., Huang, Y., et al. (2022). Deciphering the diversity and functions of plastisphere bacterial communities in plastic-mulching croplands of subtropical China. Journal of Hazardous Materials, 422, 126865. https://doi.org/https://doi.org/10.1016/j.jhazmat.2021.126865
MacLeod, M., Arp, H. P. H., Tekman, M. B., & Jahnke, A. (2021). The global threat from plastic pollution. Science, 373(6550), 61-65. https://doi.org/10.1126/science.abg5433
Maity, S., & Pramanick, K. (2020). Perspectives and challenges of micro/nanoplastics-induced toxicity with special reference to phytotoxicity. Global Change Biology, 26(6), 3241-3250. https://doi.org/https://doi.org/10.1111/gcb.15074
Marschner, P., & Rumberger, A. (2004). Rapid changes in the rhizosphere bacterial community structure during re-colonization of sterilized soil. Biology and Fertility of Soils, 40(1), 1-6. https://doi.org/10.1007/s00374-004-0736-4
Muff, S., Nilsen, E. B., O’Hara, R. B., & Nater, C. R. (2021). Rewriting results sections in the language of evidence. Trends in Ecology & Evolution. https://doi.org/https://doi.org/10.1016/j.tree.2021.10.009
Pignattelli, S., Broccoli, A., & Renzi, M. (2020). Physiological responses of garden cress (L. sativum) to different types of microplastics. Science of The Total Environment, 727, 138609. https://doi.org/https://doi.org/10.1016/j.scitotenv.2020.138609
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team. (2021). nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-153. https://CRAN.R-project.org/package=nlme
Plastics Europe (2022). Plastics - the Facts 2022. PlasticsEurope AISBL. Retrieved 5th October 2023 https://plasticseurope.org/knowledge-hub/plastics-the-facts-2022/
POWO (2024). Plants of the World Online. Facilitated by the Royal Botanic Gardens, Kew. Published on the Internet. Retrieved 05 April 2024 from http://www.plantsoftheworldonline.org/
Qi, Y., Yang, X., Pelaez, A. M., Lwanga, E. H., Beriot, N., Gertsen, H., et al. (2018). Macro- and micro- plastics in soil-plant system: Effects of plastic mulch film residues on wheat (Triticum aestivum) growth. Science of The Total Environment, 645, 1048-1056. https://doi.org/https://doi.org/10.1016/j.scitotenv.2018.07.229
R Core Team. (2022). R: A language and environment for statistical computing. In R Foundation for Statistical Computing. https://www.R-project.org/
Rillig, M. C. (2012). Microplastic in Terrestrial Ecosystems and the Soil? Environmental Science & Technology, 46(12), 6453-6454. https://doi.org/10.1021/es302011r
Rillig, M. C., Kim, S. W., & Zhu, Y.-G. (2023). The soil plastisphere. Nature Reviews Microbiology. https://doi.org/10.1038/s41579-023-00967-2
Rillig, M. C., & Lehmann, A. (2020). Microplastic in terrestrial ecosystems. Science, 368(6498), 1430-1431. https://doi.org/10.1126/science.abb5979
Rochman, C. M. (2018). Microplastics research - from sink to source. Science, 360(6384), 28-29. https://doi.org/doi:10.1126/science.aar7734
Rochman, C. M., Browne, M. A., Halpern, B. S., Hentschel, B. T., Hoh, E., Karapanagioti, H. K., et al. (2013). Classify plastic waste as hazardous. Nature, 494(7436), 169-171. https://doi.org/10.1038/494169a
Rong, L., Zhao, L., Zhao, L., Cheng, Z., Yao, Y., Yuan, C., et al. (2021). LDPE microplastics affect soil microbial communities and nitrogen cycling. Science of The Total Environment, 773, 145640. https://doi.org/10.1016/j.scitotenv.2021.145640
Semchenko, M., Barry, K. E., de Vries, F. T., Mommer, L., Moora, M., & Maciá-Vicente, J. G. (2022). Deciphering the role of specialist and generalist plant–microbial interactions as drivers of plant–soil feedback. New Phytologist, 234(6), 1929-1944. https://doi.org/https://doi.org/10.1111/nph.18118
Shah, A. A., Hasan, F., Hameed, A., & Ahmed, S. (2008). Biological degradation of plastics: A comprehensive review. Biotechnology Advances, 26(3), 246-265. https://doi.org/https://doi.org/10.1016/j.biotechadv.2007.12.005
Shi, Z., Xiong, L., Liu, T., & Wu, W. (2022). Alteration of bacterial communities and co-occurrence networks as a legacy effect upon exposure to polyethylene residues under field environment. Journal of Hazardous Materials, 426, 128126. https://doi.org/https://doi.org/10.1016/j.jhazmat.2021.128126
Speißer, B. (2023). Time matters: Why ecological effects of microplastics might change over time. Journal of Applied Ecology, 61(1), 4-6. https://doi.org/10.1111/1365-2664.14542
Speißer, B., & van Kleunen, M. (2023). Plants forage for soil patches free of plastic pollution but cannot bag the profits. Scientific Reports, 13(1), 18506. https://doi.org/10.1038/s41598-023-45662-7
Sun, Y., Shi, J., Wang, X., Ding, C., & Wang, J. (2022). Deciphering the Mechanisms Shaping the Plastisphere Microbiota in Soil. mSystems, 7(4), e00352-00322. https://doi.org/doi:10.1128/msystems.00352-22
Thompson, R. C., Moore, C. J., vom Saal, F. S., & Swan, S. H. (2009). Plastics, the environment and human health: current consensus and future trends. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1526), 2153-2166. https://doi.org/doi:10.1098/rstb.2009.0053
van der Putten, W. H., Bardgett, R. D., Bever, J. D., Bezemer, T. M., Casper, B. B., Fukami, T., et al. (2013). Plant–soil feedbacks: the past, the present and future challenges. Journal of Ecology, 101(2), 265-276. https://doi.org/https://doi.org/10.1111/1365-2745.12054
van der Putten, W. H., Bradford, M. A., Pernilla Brinkman, E., van de Voorde, T. F. J., & Veen, G. F. (2016). Where, when and how plant–soil feedback matters in a changing world. Functional Ecology, 30(7), 1109-1121. https://doi.org/https://doi.org/10.1111/1365-2435.12657
Van der Putten, W. H., Van Dijk, C., & Peters, B. A. M. (1993). Plant-specific soil-borne diseases contribute to succession in foredune vegetation. Nature, 362(6415), 53-56. https://doi.org/10.1038/362053a0
van Kleunen, M., Brumer, A., Gutbrod, L., & Zhang, Z. (2020). A microplastic used as infill material in artificial sport turfs reduces plant growth. PLANTS, PEOPLE, PLANET, 2(2), 157-166. https://doi.org/10.1002/ppp3.10071
Vethaak, A. D., & Legler, J. (2021). Microplastics and human health. Science, 371(6530), 672-674. https://doi.org/doi:10.1126/science.abe5041
Wagner, M., Monclús, L., Arp, H. P. H., Groh, K. J., Løseth, M. E., Muncke, J., et al. (2024). State of the science on plastic chemicals - Identifying and addressing chemicals and polymers of concern.
Wan, Y., Wu, C., Xue, Q., & Hui, X. (2019). Effects of plastic contamination on water evaporation and desiccation cracking in soil. Science of The Total Environment, 654, 576-582. https://doi.org/https://doi.org/10.1016/j.scitotenv.2018.11.123
Wang, F., Zhang, X., Zhang, S., Zhang, S., & Sun, Y. (2020). Interactions of microplastics and cadmium on plant growth and arbuscular mycorrhizal fungal communities in an agricultural soil. Chemosphere, 254, 126791. https://doi.org/https://doi.org/10.1016/j.chemosphere.2020.126791
Wardle, D. A. (2002). Communities and Ecosystems Linking the Aboveground and Belowground components. Princeton University Press. https://doi.org/doi:10.1515/9781400847297
Wilschut, R. A., Hume, B. C. C., Mamonova, E., & van Kleunen, M. (2023). Plant–soil feedback effects on conspecific and heterospecific successors of annual and perennial Central European grassland plants are correlated. Nature Plants. https://doi.org/10.1038/s41477-023-01433-w
Wilschut, R. A., van der Putten, W. H., Garbeva, P., Harkes, P., Konings, W., Kulkarni, P., et al. (2019). Root traits and belowground herbivores relate to plant–soil feedback variation among congeners. Nature Communications, 10(1), 1564. https://doi.org/10.1038/s41467-019-09615-x
Xi, N., Adler, P. B., Chen, D., Wu, H., Catford, J. A., van Bodegom, P. M., et al. (2021). Relationships between plant–soil feedbacks and functional traits. Journal of Ecology, 109(9), 3411-3423. https://doi.org/https://doi.org/10.1111/1365-2745.13731
Xue, W., Berendse, F., & Bezemer, T. M. (2018a). Spatial heterogeneity in plant–soil feedbacks alters competitive interactions between two grassland plant species. Functional Ecology, 32(8), 2085-2094. https://doi.org/https://doi.org/10.1111/1365-2435.13124
Xue, W., Bezemer, T. M., & Berendse, F. (2018b). Density-dependency and plant-soil feedback: former plant abundance influences competitive interactions between two grassland plant species through plant-soil feedbacks. Plant and Soil, 428(1), 441-452. https://doi.org/10.1007/s11104-018-3690-x
Zhang, M., Zhao, Y., Qin, X., Jia, W., Chai, L., Huang, M., et al. (2019). Microplastics from mulching film is a distinct habitat for bacteria in farmland soil. Science of The Total Environment, 688, 470-478. https://doi.org/https://doi.org/10.1016/j.scitotenv.2019.06.108
Zhang, Y., Ma, J., O'Connor, P., & Zhu, Y.-G. (2022). Microbial communities on biodegradable plastics under different fertilization practices in farmland soil microcosms. Science of The Total Environment, 809, 152184. https://doi.org/https://doi.org/10.1016/j.scitotenv.2021.152184
Zheng, Y., Yanful, E. K., & Bassi, A. S. (2005). A Review of Plastic Waste Biodegradation. Critical Reviews in Biotechnology, 25(4), 243-250. https://doi.org/10.1080/07388550500346359
Zhou, Y., He, G., Jiang, X., Yao, L., Ouyang, L., Liu, X., et al. (2021). Microplastic contamination is ubiquitous in riparian soils and strongly related to elevation, precipitation and population density. Journal of Hazardous Materials, 411, 125178. https://doi.org/10.1016/j.jhazmat.2021.125178
Zhu, D., Ma, J., Li, G., Rillig, M. C., & Zhu, Y.-G. (2021). Soil plastispheres as hotspots of antibiotic resistance genes and potential pathogens. The ISME Journal, 16(2), 521-532. https://doi.org/10.1038/s41396-021-01103-9
Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., & Smith, G. M. (2009). Mixed effects models and extensions in ecology with R. Springer. https://doi.org/10.1007/978-0-387-87458-6

Table 1 Results of linear mixed-effects models testing the effects of soil community (intact vs. initially depleted), material treatment (glass and different microplastics in low and high concentrations), and the interaction of both factors on PSF effects on Crepis biennis. Effects were assessed for the calculated PSF values based on total plant biomass (PSF_biomass), root-weight ratio (PSF_RWR), specific root length (PSF_SRL), and average link length (PSF_LL). Fixed effects were assessed using log-likelihood ratio tests (Zuur et al., 2009). Log-likelihood ratios (LLR) are approximately χ²-distributed. P values < 0.05 are indicated in bold, p values < 0.1 are indicated in italics.

		PSF_biomass		PSF_RWR		PSF_SRL		PSF_LL
Fixed effects	df	LLR	p	LLR	p	LLR	p	LLR	p
Initial leaf area	1	0.87	0.35	6.05	0.014	1.41	0.23	0.23	0.64
Material treatment (M)	10	12.81	0.23	14.16	0.166	13.79	0.18	18.43	0.048
Soil community (S)	1	37.59	< 0.001	6.25	0.012	48.43	< 0.001	47.73	< 0.001
M×S	10	45.69	< 0.001	34.58	< 0.001	48.61	< 0.001	13.89	0.18
Random effects		SD		SD		SD		SD
Block		0.0024		0.0772		0.0024		0.0286
Residual		0.3219		0.1567		0.2826		0.1936

Table 2 Results of linear mixed-effects models testing the effects of soil community (intact vs. initially depleted), orthogonal contrasts comparing specific material treatment combinations (material type, i.e. glass and different microplastics, and low vs. high concentrations; Supplementary Table 3), and the interaction of both factors on PSF effects on Crepis biennis. Effects were assessed for the calculated PSF values based on total plant biomass (PSF_biomass), root-weight ratio (PSF_RWR), specific root length (PSF_SRL), and average link length (PSF_LL). Fixed effects were assessed using log-likelihood ratio tests (Zuur et al., 2009). Log-likelihood ratios (LLR) are approximately χ²-distributed. P values < 0.05 are indicated in bold, p values < 0.1 are indicated in italics.

		PSF_biomass		PSF_RWR		PSF_SRL		PSF_LL
Fixed effects	df	LLR	p	LLR	p	LLR	p	LLR	p
Initial leaf area	1	0.87	0.35	6.05	0.014	1.41	0.23	0.23	0.64
C1 (particles vs no particles)	1	0.12	0.73	0.40	0.53	1.10	0.29	0.10	0.75
C2 (glass vs plastics)	1	0.43	0.51	0.13	0.72	0.00	0.95	0.24	0.62
C3 (individual vs mixed plastics)	1	0.87	0.35	0.53	0.46	1.24	0.27	0.07	0.79
C4 (degradable vs conventional)	1	0.76	0.38	0.18	0.67	7.51	0.006	0.73	0.39
C5 (PE vs EPDM)	1	0.14	0.71	0.78	0.38	0.48	0.49	6.94	0.008
C6 (low vs high glass)	1	0.02	0.88	9.19	0.002	3.02	0.08	1.90	0.17
C7 (low vs high EPDM)	1	2.86	0.091	0.00	0.98	0.11	0.74	0.25	0.61
C8 (low vs high PE)	1	0.04	0.85	2.54	0.11	0.35	0.56	1.16	0.28
C9 (low vs high PHA)	1	0.68	0.41	1.66	0.20	0.00	0.98	2.17	0.14
C10 (low vs high mix)	1	8.26	0.004	0.09	0.77	0.70	0.40	6.17	0.013
Soil community (S)	1	37.60	< 0.001	6.25	0.012	48.43	< 0.001	47.73	< 0.001
C1×S	1	16.94	< 0.001	16.67	< 0.001	3.62	0.057	0.16	0.69
C2×S	1	0.01	0.91	11.14	< 0.001	15.33	< 0.001	2.12	0.15
C3×S	1	1.99	0.16	0.04	0.84	1.36	0.24	1.73	0.19
C4×S	1	13.65	< 0.001	3.00	0.08	0.47	0.49	1.92	0.17
C5×S	1	3.27	0.071	0.04	0.85	0.00	0.98	2.50	0.11
C6×S	1	9.20	0.002	0.14	0.70	0.78	0.38	0.19	0.66
C7×S	1	13.64	< 0.001	4.61	0.031	28.17	< 0.001	1.39	0.24
C8×S	1	0.46	0.50	3.22	0.073	0.11	0.74	2.05	0.15
C9×S	1	0.00	0.98	0.83	0.36	6.22	0.013	2.09	0.15
C10×S	1	1.49	0.22	1.34	0.25	1.80	0.18	0.10	0.76
Random effects		SD		SD		SD		SD
Block		0.0024		0.0772		0.0024		0.0286
Residual		0.3219		0.1567		0.2826		0.1936

Table 3 Results of linear mixed-effects models testing the effects of soil community (intact vs. initially depleted), material treatment (glass and different microplastics in low and high concentrations), and the interaction of both factors on PSF effects on Eragrostis minor. Effects were assessed for the calculated PSF values based on total plant biomass (PSF_biomass) and root-weight ratio (PSF_RWR). Fixed effects were assessed using log-likelihood ratio tests (Zuur et al., 2009). Log-likelihood ratios (LLR) are approximately χ²-distributed. P values < 0.05 are indicated in bold.

		PSF_biomass		PSF_RWR
Fixed effects	df	LLR	p	LLR	p
Initial leaf area	1	0.15	0.7	6.92	0.009
Material treatment (M)	10	24.83	0.006	33.78	< 0.001
Soil community (S)	1	0.78	0.38	20.52	< 0.001
M×S	10	13.54	0.19	22.96	0.01
Random effects		SD		SD
Block		0.0056		0.1077
Residual		0.2673		0.1771

Table 4 Results of linear mixed-effects models testing the effects of soil community (intact vs. initially depleted), orthogonal contrasts comparing specific material treatment combinations (material type, i.e. glass and different microplastics, and low vs. high concentrations; Supplementary Table 3), and the interaction of both factors on PSF effects on Eragrostis minor. Effects were assessed for the calculated PSF values based on total plant biomass (PSF_biomass) and root-weight ratio (PSF_RWR). Fixed effects were assessed using log-likelihood ratio tests (Zuur et al., 2009). Log-likelihood ratios (LLR) are approximately χ²-distributed. P values < 0.05 are indicated in bold, p values < 0.1 are indicated in italics.

		PSF_biomass		PSF_RWR
Fixed effects	df	LLR	p	LLR	p
Initial leaf area	1	0.15	0.70	6.92	0.009
C1 (particles vs no particles)	1	7.78	0.005	3.95	0.047
C2 (glass vs plastics)	1	0.01	0.92	0.32	0.57
C3 (individual vs mixed plastics)	1	0.62	0.43	0.59	0.44
C4 (degradable vs conventional)	1	0.33	0.57	0.54	0.46
C5 (PE vs EPDM)	1	1.24	0.27	8.20	0.004
C6 (low vs high glass)	1	1.37	0.24	6.42	0.011
C7 (low vs high EPDM)	1	0.56	0.46	0.36	0.55
C8 (low vs high PE)	1	1.08	0.30	0.21	0.64
C9 (low vs high PHA)	1	14.25	< 0.001	0.53	0.47
C10 (low vs high mix)	1	0.56	0.46	0.99	0.32
Soil community (S)	1	0.78	0.38	20.52	< 0.001
C1×S	1	0.38	0.54	1.35	0.25
C2×S	1	6.84	0.008	16.26	< 0.001
C3×S	1	0.20	0.65	0.02	0.88
C4×S	1	0.02	0.88	3.62	0.057
C5×S	1	1.74	0.19	1.65	0.20
C6×S	1	1.27	0.26	1.83	0.18
C7×S	1	1.55	0.21	0.35	0.55
C8×S	1	1.80	0.18	2.30	0.13
C9×S	1	0.17	0.68	4.89	0.027
C10×S	1	0.39	0.53	0.22	0.64
Random effects		SD		SD
Block		0.0056		0.1077
Residual		0.2673		0.1771

mppsfSI.docx

Artificial particles and soil communities interactively change heterospecific plant-soil feedbacks

Status:

Version 1

Abstract

Background and aims

Methods

Results

Conclusion

Figures

Introduction

Methods

Study species and precultivation:

Microplastics:

Conditioning phase:

Feedback phase:

Statistical analysis:

Results

Discussion

Abbreviations

Declarations

Competing interests

Author contributions

Acknowledgements

Data availability

References

Tables

Supplementary Files

Status:

Version 1