effects of root decomposition on plant soil feedback of ... · effects of root decomposition on...

Effects of root decomposition on plant–soil feedback ofearly- and mid-successional plant species

Naili Zhang1,2, Wim H. Van der Putten2,3 and G. F. (Ciska) Veen2

1State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, the Chinese Academy of Sciences, 20 Nanxincun, Xiangshan, Beijing 100093, China; 2Department of

Terrestrial Ecology, Netherlands Institute of Ecology (NIOO–KNAW), Droevendaalsesteeg 10, Wageningen 6708 PB, the Netherlands; 3Laboratory of Nematology, Wageningen University

and Research Centre, Droevendaalsesteeg 1, Wageningen 6708 PB, the Netherlands

Author for correspondence:Naili Zhang

Tel: +86 10 62836958Email: [email protected]

Received: 31 January 2016

Accepted: 6 April 2016

New Phytologist (2016) 212: 220–231doi: 10.1111/nph.14007

Key words: decomposing roots, ex-arablefields, functional group, long-term effects,plant succession, plant–soil feedback (PSF),short–term effects.

Summary

� Plant–soil feedback (PSF) is an important driver of plant community dynamics. Many studies

have emphasized the role of pathogens and symbiotic mutualists in PSFs; however, less is

known about the contribution of decomposing litter, especially that of roots.� We conducted a PSF experiment, where soils were conditioned by living early- and mid-

successional grasses and forbs with and without decomposing roots of conspecific species

(conditioning phase). These soils were used to test growth responses of conspecific and

heterospecific plant species (feedback phase).� The addition of the roots of conspecifics decreased the biomass of both early- and mid-

successional plant species in the conditioning phase. In the feedback phase, root addition had

positive effects on the biomass of early-successional species and neutral effects on mid-

successional species, except when mid-successional grasses were grown in soils conditioned

by conspecifics, where effects were negative. Biomass of early- and mid-successional forbs

was generally reduced in soils conditioned by conspecifics.� We conclude that root decomposition may increase short-term negative PSF effects, but

that the effects can become neutral to positive over time, thereby counteracting negative

components of PSF. This implies that root decomposition is a key element of PSF and needs to

be included in future studies.

Introduction

Plant–soil feedbacks (PSFs) are plant-induced changes in bioticand abiotic soil properties that in turn influence the performanceof the same or other plant species (Bever et al., 1997; Van derPutten et al., 2013). PSF involves short-term processes mediatedby root-associated organisms, such as soil-borne pathogens, nitro-gen (N)-fixing bacteria or mycorrhizal fungi, and longer termprocesses such as decomposition mediated by saprotrophic organ-isms, which influence plant growth through litter decomposition,mineralization and consequent nutrient availability (Wardleet al., 2004; Poorter et al., 2012). A growing body of evidenceindicates that PSFs mediated by root-associated organisms play acritical role in driving plant growth and community compositionand dynamics (Augspurger & Kelly, 1984; Klironomos, 2002;Petermann et al., 2008; Bever et al., 2015); however, the role ofdecomposition mediated by saprotrophic soil organisms in driv-ing PSF is still poorly understood (Wardle, 2002; Kardol et al.,2015). Here, we studied plant–soil feedback effects with particu-lar focus on the role of root decomposition.

Soil pathogens often cause negative PSF effects, which reduceplant competitive abilities and promote co-existence (Manganet al., 2010; Terborgh, 2012; Bever et al., 2015). Symbiotic

mutualists, such as arbuscular mycorrhizal fungi, commonly gen-erate positive PSF effects on host plant growth (Klironomos,2002), but under specific conditions they may also function asparasites (Johnson et al., 1997; Castelli & Casper, 2003;Klironomos, 2003). Both pathogens and mycorrhizal fungi inter-act directly with living plant roots. In contrast, saprotrophic soilorganisms mediate PSF by influencing soil nutrient availabilitythrough decomposition of shoot and root litter, and root exu-dates (Wardle et al., 2004). Increased nutrient availability maystimulate plant growth, resulting in positive PSF effects (Chap-man et al., 2006), but the release of phytotoxins and toxic self-DNA from aged plant litter has been proposed to cause negativePSF (Bonanomi et al., 2011; Mazzoleni et al., 2015). Therefore,effects of decomposition on PSF may be expected to vary frompositive to negative.

PSF is made up of the net effect of all positive and negativeinteractions with soil communities (Bever et al., 1997); however,little is known about the proportional contribution of each ofthese individual components to the net effect (Van der Puttenet al., 2016). Moreover, root-associated and saprotrophic organ-isms may increase plant susceptibility to soil-borne pathogens byenhancing soil nutrient availability, or via predators in thedecomposer subsystem that forage on root-associated microbes

220 New Phytologist (2016) 212: 220–231 � 2016 The Authors

New Phytologist� 2016 New Phytologist Trustwww.newphytologist.com

Research

(Wardle et al., 2004; Wardle, 2006). In a theoretical study, it hasbeen shown that the potential of decomposer organisms to mod-ify PSF may be greater in systems where plant growth is stronglyaffected by mycorrhizal fungi than in systems where pathogensplay a major role (Ke et al., 2015). However, relatively few empir-ical studies have tested these theoretical predictions (Bardgettet al., 2014). Therefore, we studied the effects of decaying plantroots on the outcome of PSF effects.

We used early- and mid-successional grass and forb species, asboth the strength and direction of PSF have been shown to varyduring vegetation succession (Kardol et al., 2006, 2007; Kulma-tiski & Kardol, 2008; Jing et al., 2015). Early-succession plantspecies generally experience negative PSF, thereby speeding upplant species replacement leading to successional development,whereas later succession plant species have neutral or positivePSF that slows down succession and favors temporal stability(Kardol et al., 2006; Revilla et al., 2013). Although these previousstudies did not tease apart the relative effects of pathogenic, sym-biotic, and decomposer organisms, an increasingly positive PSFwhen succession progresses suggests that the contribution ofdecomposer organisms to PSF processes may increase towardslater successional stages via increased nutrient availability and thebuild-up of specific litter decomposer interactions (Milcu &Manning, 2011). These positive PSF effects may be furtherenhanced by increased mycorrhizal responsiveness of later succes-sion plant species (Koziol & Bever, 2015). Moreover, plantspecies induce specific changes in abiotic and biotic soil proper-ties, leading to species-specific PSF effects that may generally dif-fer between grass and forb species (Van der Putten, 2003;Bezemer et al., 2006; Kardol et al., 2007; Kulmatiski & Kardol,2008; Cameron et al., 2009).

The aim of our study was to test how decomposing roots mayinfluence PSF effects of early- and mid-successional grass andforb species. We focused on the decomposition of roots, becausethis is the main source of litter in temperate grasslands, where themajority of the plant standing biomass is situated belowground(Poorter et al., 2012), and a large part of the aboveground plantmaterial is commonly removed by grazing or mowing. We testedfour hypotheses. First, we expected that short-term effects of rootaddition (hereafter named short-term effects) would have a nega-tive effect on plant biomass production, as we expected the rootsto be a source of pathogen infection, and because decompositionof fresh roots may cause nutrient immobilization (Schmidt et al.,1997; Van der Heijden et al., 2008; Lalibert�e et al., 2015). More-over, phytotoxic compounds released from litter decompositionin early phases of decomposition may inhibit plant growth (Singhet al., 1999; An et al., 2001; Meier et al., 2009; Bonanomi et al.,2011). Our second hypothesis was that in the longer term (here-after referred to as long-term effects) root decompositionincreases plant growth, which may be a result of an increase insoil nutrient availability as a consequence of nutrient release dur-ing decomposition (Miki & Kondoh, 2002). Third, we expectedthat the short-term negative effects of root addition would bestrongest for early-successional plant species, whereas the long-term positive effects of root addition would be strongest for mid-successional plant species (Kardol et al., 2006; Koziol & Bever,

2015). As a fourth hypothesis, we predicted that long-term effectsof root addition on plant biomass production would be morepositive for forbs than grasses, because soils conditioned by forbsoften have higher soil nutrient availibility because of high litterdecomposibility (Turner et al., 1995; Kull & Aan, 1997; �Agrenet al., 2013).

To test our hypotheses, we established a two-phase glasshousePSF experiment with six early-successional plant species (thethree grasses Lolium perenne, Holcus lanatus, and Arrhenatherumelatius and the three forbs Jacobaea vulgaris, Leucanthemumvulgare, and Rumex obtusifolius) and six mid-successional plantspecies (the three grasses Anthoxanthum odoratum, Agrostiscapillaris, and Deschampsia flexuosa and the three forbsHypochaeris radicata, Plantago lanceolata, and Rumex acetosella).In the conditioning phase of the experiment, we added roots col-lected from the field to half of the pots in order to test the short-term effects of decomposing roots on plant biomass production.In the feedback phase we determined the plant biomass responseto soils conditioned by conspecific plants (designated ‘home’)with or without conspecific roots and heterospecific plants (desig-nated ‘foreign’) with or without heterospecific roots. We relatedplant biomass responses to changes in chemical and microbialbiomass properties of the soil as a result of soil conditioning byliving plant roots, with or without decaying plant roots. Thisapproach enabled us to tease apart effects of living conspecificand heterospecific plant roots and decomposing roots on PSFand to test how the effects resulting from living and decomposingroots depend upon changed biotic and abiotic soil properties.

Materials and Methods

Experimental design

The experiment included both a conditioning and a feedbackphase. The conditioning phase (12 wk) of the PSF experimentconsisted of 12 plant species (three early-successional grasses andthree early-successional forbs, and three mid-successional grassesand three mid-successional forbs)9 2 root treatments (no rootsadded vs roots added)9 6 replicates = 144 pots. The soils fromthe conditioning phase were used in the feedback phase of theexperiment in order to test the response of plants to soils condi-tioned by conspecifics (home) with or without conspecific rootsand to soils conditioned by heterospecifics (foreign) with or with-out heterospecific roots. The feedback phase (14 wk) consisted of12 plant species9 4 types of soil per plant species (home soilsconditioned with or without roots, and foreign soils conditionedwith or without roots)9 6 replicates = 288 pots. The lengths ofthe soil conditioning and feedback phases were comparable tothose in other PSF experiments (e.g. Bezemer et al., 2006; VanGrunsven et al., 2010; Jing et al., 2015).

Study area

We set up the PSF experiment in a glasshouse using soils androots from a well-established chronosequence of ex-arable fieldsin south Veluwe, the Netherlands (Kardol et al., 2005, 2006;

� 2016 The Authors

New Phytologist� 2016 New Phytologist TrustNew Phytologist (2016) 212: 220–231

www.newphytologist.com

NewPhytologist Research 221

Van de Voorde et al., 2011). The Veluwe is a nature reserve inthe center of the Netherlands, located on parental soil material ofsandy to sandy loam glacial deposits formed during the Saalienice age (Kardol et al., 2005; Van der Wal et al., 2006). We usedex-arable fields. In these fields, agricultural practices have beendiscontinued for various times, and the abandoned fields havebeen managed as semi-natural grasslands since agriculture ceased.Arable weeds and pioneer plant species such as Apera spica-venti(L.) P.Beauv., Erigeron canadensis L., Viola arvensis and Jacobaeavulgaris Gaertn. were dominant at recently abandoned fields,whereas longer term abandoned fields were dominated by latersuccessional plant species, such as Agrostis capillaris L.,Deschampsia flexuosa (L.) Trin. and Rumex acetosella L. (Kardolet al., 2005).

Preparation of experimental material

We collected background soil for sterilization from a long-termresearch site at Mossel (52.04oN, 5.45oE) (Van der Putten et al.,2000). The background soil was sterilized minimally by 25-kGygamma irradiation, which eradicates all soil life. Soil inocula werecollected from two short-term abandoned sites (early-successionalsoils) at Oud Reemst (52.02oN, 5.48oE) and Reijerskamp(52.01oN, 5.47oE), which have been abandoned since 2005, andfrom two longer term abandoned sites (mid-successional soils) atDennenkamp (52.02oN, 5.48oE) and Mosselse Veld (52.04oN,5.44oE), which were abandoned in 1982 and 1985, respectively.All soils were collected from 0 to 15 cm depth where most rootsare present, and sieved using a 5-mm sieve in order to removecoarse fragments, including stones and roots.

We collected fresh root material from multiple populations ofeach of the 12 different plant species at locations in the chronose-quence where they were abundant. We selected six plant speciesthat typically grow in early-successional grasslands, i.e. threegrasses, Lolium perenne L., Holcus lanatus L. and Arrhenatherumelatius (L.) P.Beauv. ex J. & C.Presl., and three forbs, J. vulgaris,Leucanthemum vulgare (Vaill.) Lam., and Rumex obtusifolius L.,and six plant species that are abundant in mid-successional grass-lands, that is, three grasses, Anthoxanthum odoratum L.,A. capillaris, and D. flexuosa, and three forbs, Hypochaeris radicataL., Plantago lanceolata L., and R. acetosella. The roots were care-fully rinsed to remove soil, and air-dried to constant weight toenable addition of the same amount of dry root mass to eachtreatment by accounting for variability in fresh : dry root weight.Roots were cut into 1-cm fragments, and stored at 4°C until thestart of the glasshouse experiment.

Seeds of the 12 plant species were obtained from ‘the Cruydt-hoeck’ in Assen, the Netherlands, which supplies seeds collectedfrom wild plant populations. All seeds were sterilized for 2 minusing 0.2% chloride solution, sown in a plastic box with sterilizedglass beads, and kept moist by adding demineralized water. Seedswere geminated in an incubator under a regime of 16 h, 21°Clight and 8 h, 16°C dark. One week after germination, theseedlings were placed in a climate chamber with light at 4°C,which kept all seedlings in the same state of development untilthere were sufficient amounts of seedlings of all plant species.

The use of seedlings allows us to standardize for variation in seedviability, but does not allow us to determine the effect of PSF onseed germination.

Plant–soil feedback experiment

Soil conditioning In the soil conditioning phase, for eachplant species, 12 3-l pots were filled with 3 kg of soil com-posed of 90% sterilized background soil with a 10% inoculumof field soil. Early-successional plant species received an early-successional soil inoculum, while mid-successional plant speciesreceived a mid-successional soil inoculum. For each plantspecies, half of the pots received 8 g of dry monospecific rootfragments of conspecifics that were homogenized with the soil(Fig. 1, upper right panel). The amount of roots was of thesame order of magnitude as the standing root biomass in thestudy area (Bezemer et al., 2010). We planted five seedlings ofone of the plant species in each pot. This resulted in a total of144 pots in the conditioning phase, as outlined in the ‘Experi-mental design’ section. Plants were grown in a climate-controlled glasshouse under a regime of 16 h, 21� 1°C lightand 8 h, 16� 1°C dark, and 60% relative humidity. Once aweek, we weighed each pot and added enough demineralizedwater that the soil moisture was reset at 15% (w/w). Thiswatering procedure was carried out throughout the experimen-tal period. After 12 wk, when the added roots were partiallydecomposed, aboveground biomass was harvested and soilswere used for the feedback phase of the experiment.

Feedback phase In the soil feedback phase we had 288 1-l pots,as outlined in the ‘Experimental design’ section. For each individ-ual pot from the conditioning phase, we homogeneized the soiland cut fresh roots that had grown in the conditioning phase into1-cm fragments, and homogenized these roots with soil of originto act as a source of rhizosphere microbiome. This is a commonpractice in PSF experiments (e.g. Bonanomi et al., 2005; VanGrunsven et al., 2010). From each pot in the conditioning phase,one half of the homogenized soils was used to create conditionedhome soil and the other half to create foreign soil (Fig. 1, lowerpanel). Foreign soil for one plant species consisted of a mixture ofsoils conditioned by the other five plant species (Fig. 1, lowerpanel). During this entire process, the individual replicate struc-ture initiated during the conditioning phase was maintained bycreating home and foreign soils replicate by replicate. In the feed-back phase, each pot contained 1 kg of soil with 15% moisture(w/w). Two seedlings were planted per pot, having the samemagnitude of root mass per soil volume as that in theconditioning phase. Plants were grown for 14 wk in a glasshouseunder the same growth conditions as in the conditioning phase.

Measurements

At the end of the conditioning phase, plant shoots that had beenclipped at 1 cm above the soil surface were dried at 60°C to con-stant weight and weighed to determine shoot biomass. At thesame time, we collected soil samples from each pot to determine

New Phytologist (2016) 212: 220–231 � 2016 The Authors


Research

NewPhytologist222

soil microbial and abiotic properties after the conditioning phase.Soil samples were sieved using a 4-mm sieve, and then dividedinto two subsamples: one was dried at 40°C in order to deter-mine physicochemical properties and one was freeze-dried inorder to determine phospholipid (PLFA) and neutral lipid(NLFA) fatty acid contents. After the feedback phase, plantshoots were harvested and roots were gently rinsed to remove soiland the added roots. Both shoot and root samples were thendried at 60°C until constant weight and weighed to determineshoot and root biomass.

Plant and soil total carbon (C) and N concentrations were mea-sured in ground subsamples using a Flash EA1112 elemental ana-lyzer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Tengrams of dry soil was extracted using 50 ml of 1M KCl solutionfor 2 h at 250 rpm. The extract obtained was used to determinesoil ammonium and nitrate N with a SEAL QuAAtro SegmentedFlow Analysis (SFA) system (Beun-de Ronde BV, Abcoude, theNetherlands) (Keeney & Nelson, 1982; Griffin et al., 2009). Todetermine plant phosphorus, 3 g of the plant sample was weighedand ignited at 550°C for 30 min in a muffle furnace, thenextracted with 10 ml of 2.5% potassium persulfate, and measuredwith an Auto-Analyzer (Seal QuAAtro SFA system) (Murphy &Riley, 1962). A sample of 2.5 g of dry soil was extracted in a 1 : 20(w/v) ratio with a 0.5 M NaHCO3 solution (pH 8.5), and thenshaken mechanically for 30 min following Olsen’s procedure(Olsen et al., 1954). The extract obtained was used to measure soilavailable phosphorus using the SEAL QuAAtro SFA system(Beun-de Ronde BV).

We used 3 g of freeze-dried soils to determine microbialbiomass using PLFA and NLFA analysis (Bligh & Dyer, 1959;White et al., 1979). The PLFA/NLFA extractions were carriedout according to Frosteg�ard et al. (1991), and the abundance oflipid acids is expressed in lg g�1 soil. The saturated PLFAs con-taining i15:0, a15:0, i16:0, a17:0, i17:0, 16:1x7c, cy17:0 andcy19:0 were used as bacterial indicators (Frosteg�ard & B�a�ath,1996; Zak et al., 2003). Two unsaturated PLFAs, 18:1x9cand18:2x6c, were used as fungal indicators (Zak et al., 1996,

2003). The NLFA 16:1x5c was used as a biomarker of arbuscu-lar mycorrhizal fungi (AMF; Olsson, 1999).

Statistical analyses

We conducted two-way and one-way ANOVAs using the ‘aov’ Rfunction to analyze the effects of successional stage and functionalgroup on the chemical composition of the added roots. We useda mixed effect model (‘lmer’ in the lmerTest package) with repli-cate as a random factor, and successional stage, root addition andfunctional group as fixed factors to test their effects on shootbiomass and soil properties (i.e. soil total C and N, inorganic N,Olsen P, pH, and bacterial PLFA, fungal PLFA and AMF NLFAbiomass) in the conditioning phase. We tested shoot and rootbiomass production in the feedback phase using a similar modelwith soil conditioning by living plants (i.e. home vs foreign) asan additional fixed factor. We conducted a Tukey test using the‘HSD test’ R function from the ‘agricolae’ package to perform amultiple comparison of plant biomass and soil physicochemicaland microbial properties between treatments. All data were ana-lyzed using R v.3.1.0 (R Core Team, 2014).

Results

Plant biomass and soil properties after soil conditioning

At the end of the soil conditioning phase, the shoot biomass ofboth early- and mid-succession plant species was reduced byadding roots, with a stronger reduction in the case of forbs thanof grasses (Fig. 2). Root addition resulted in a proportionaldecrease in the shoot biomass of early- and mid-successional forbsby 45% and 51%, respectively, while this was only 23% and26% for early- and mid-successional grasses. Anthoxanthumodoratum was the only species for which shoot biomass was notaffected by root addition (Supporting Information Fig. S1).

In most cases, except for mid-succession soil with grasses, rootaddition did not affect total C, N, or their ratio (Table 1).

Loliumlanatus elatius

Arrhenatherum JacobaeaobtusifoliusRumex Leucanthemum Mono-specific (12 wk)

Conditioning phase

root litter addition

Own soil(home)

Grass Forb Grass Grass GrassForb Forb Forb

Mixed soil(foreign)

Own soil(home)

Mixed soil(foreign)

Feedback phase(14 wk)

vulgarevulgarisHolcus

perenne

Fig. 1 Graphical illustration of the experimental design consisting of both conditioning (upper panel) and feedback (lower panel) phases. Here, we onlyshow the set-up of six plant species from the early-successional stage, but this also applies to the mid-successional stage. The upper panel shows thecontrol without root addition (upper left) vs plants with mono-conspecific root addition (upper right) in the conditioning phase. In the feedback phase,which is outlined in the lower panel, the grass Lolium perennewas used as an example to show how ‘home’ vs ‘foreign’ soils were obtained.





In mid-successional soils with grasses, root addition decreasedtotal C, N, and nitrate N, while it increased the C : N ratio andammonium N (Table 1). Ammonium N and Olsen P were lowerin early successional soil with forbs when roots were added(Table 1). Further, root addition increased the pH of early-successional soils with forbs and of mid-successional soils withboth forbs and grasses (Table 1).

Microbial soil properties were not affected by root addition,but were affected by plant functional group (Fig. 3). BacterialPLFA biomass was significantly higher in soils with early-successional grasses than in those with early-successional forbs. Incontrast, fungal PLFA biomass was higher in soils withmid-successional forbs than in those with mid-successional

grasses (Fig. 3). There was no difference in AMF biomass, asindicated by NLFA, between soils with grasses and those withforbs, irrespective of successional position (Fig. 3).

Plant biomass responses in the feedback phase

Early-successional grasses and forbs had higher shoot and rootbiomasses when roots were added, both in home and in foreignsoils (Fig 4a,c). In the case of early-successional forbs, plants pro-duced significantly more shoot and root biomasses in foreignthan in home soil, irrespective of root addition (Fig. 4a,c). In thecase of early-successional grasses, the effect of foreign soil onbiomass production depended on root addition. Without rootsadded, there was no difference between home and foreign soil,whereas with roots added, there was more shoot biomass in for-eign compared with home soil (Fig. 4a,c).

For the mid-successional plant species, root addition generallyhad no effect on shoot and root biomasses (Fig. 4b,d). The excep-tion was grasses in home soils, for which shoot biomass was lowerwhen roots were added (Fig. 4b). In the feedback phase, shootand root biomasses of mid-successional forbs were generallylower in home than in foreign soils, indicating negative PSF(Fig. 4b,d). Shoot biomass of mid-successional grasses showedpositive feedback (home vs foreign soils) in soil without roots,but not in soil with roots added (Fig. 4b).

Chemical properties of plant roots

The chemical composition of the field-collected root material dif-fered between grasses and forbs, as well as among plant species(Table 2). Early-successional grasses had higher root total C(F1,34 = 37.0; P < 0.001) and N concentration (F1,34 = 194;P < 0.001), but lower root total phosphorus (P) content thanearly-successional forbs (F1,34 = 8.99; P = 0.005). Root C : Nratios were lower in grasses than in forbs (F1,34 = 187; P < 0.001).Among early-successional plant species, C : N was highest inroots of Jacobaea vulgaris. Early-successional forbs had lowerC : P (F1,34 = 9.94; P = 0.003) and N : P ratios (F1,34 = 105;P < 0.001) than early-succession grasses; values were highest inroots of R. obtusifolius and H. lanatus, respectively (Table 2).

Roots of mid-successional grasses had higher total N(F1,34 = 36.5; P < 0.001), and lower total P (F1,34 = 17.0;P < 0.001) than mid-successional forbs, while total C content didnot differ (Table 2). Root C : N ratios were lower inmid-successional grasses than forbs (F1,34 = 34.2; P < 0.001) andhighest in roots of H. radicata. Mid-successional grasses had thehighest C : P (F1,34 = 11.0; P = 0.002) and N : P ratios(F1,34 = 23.7; P < 0.001). Roots of D. flexuosa had the highest C : Pvalues and A. capillaris had the highest N : P values (Table 2).

Discussion

We aimed to determine the contribution of root decompositionto the PSF effects of early- and mid-successional grasses and forbs.We found that in the conditioning phase root addition reducedthe shoot and root biomasses of both early- and mid-successional

10

8

6

2

010

Sho

ot b

iom

ass

(g p

er p

ot)

8

6

4

2

0Grass

a

bb

c

a

b b

c

Root effect (R) ***Mid-succession plants

Early-succession plants

Without roots With roots

Functional group (G) ***R × G *

Root effect (R) ***Functional group (G) ***R × G *

Forb

4

(a)

(b)

Fig. 2 Effects of root addition and plant functional group (grass vs forbs) onshoot biomass (g dry weight per pot) of (a) early- and (b) mid-successionalplant species after soil conditioning. Values shown in the figure aremean� SE (n = 6). Statistically significant effects: *, P < 0.05; ***,P < 0.001. Different lowercase letters within each successional stage referto significant differences at P < 0.05 between treatments (root addition:with vs without) and plant functional groups (grass vs forb).



Research

NewPhytologist224

plant species. Forb biomass was reduced approximately twice asmuch by root addition as grass biomass. In the feedback phase,early-successional grasses and forbs benefitted from root addition.By contrast, the shoot biomass of mid-successional grasses wasreduced by root addition, while forbs and belowground grassbiomass did not respond. Therefore, we conclude that rootdecomposition can influence plant performance, but that thestrength and direction of the effects can vary with time andbetween plant species of different successional stages.

Short-term effects of root addition

Our results are in support of our first hypothesis, which predictedthat short-term effects of root decomposition will have a negativeeffect on plant growth. A recent study has suggested that negativeeffects of decomposing roots on PSF may be mediated by self-DNA released from decomposing plant leaves, which can be toxicto conspecifics, and accumulating during 120 d of decomposition(Mazzoleni et al., 2015). Although we did not measure DNAaccumulation in our study, inhibition by self-DNA from rootscould have contributed to the short-term negative effects of rootaddition in the conditioning phase, which had a duration of 84 din the present study. However, at the end of the feedback phasewhen the added roots had experienced 182 d decomposition, theeffects of root addition were neutral to positive, indicating thatthe effects of root-derived self-DNA may not have limitedbiomass production. This could be explained by postulating thatroot litter releases are less inhibiting to self-DNA than shoot litter

(Dhillon & Miksche, 1982); however future studies will beneeded to determine the potential role of self-DNA inhibition inthe case of root decomposition.

Inhibition of plant growth may also be attributable to therelease of phytotoxic compounds in the early stages of decompo-sition (Hodge, 2004; Meier et al., 2009; Bonanomi et al., 2011).Alternatively, decomposing roots can be a source of soil-bornepathogens (Huber & Watson, 1970; Van der Putten, 2003), orunder some conditions AMF could also cause a negative effect onplant growth (Castelli & Casper, 2003; Klironomos, 2003).However, we did not find changes in the biomass of microbialgroups attributable to root addition based on PLFA and NLFApatterns (Table 1). The activity and/or biomass of root-associatedorganisms may have been reduced by drying the roots (Jasperet al., 1989); however, this is standard procedure in many PSFexperiments to align starting conditions (Reinhart et al., 2010).To further elucidate the role of root-associated pathogens in PSF,molecular sequencing-based tools in combination with isolationand inoculation trials will be needed.

Finally, the C input from fresh plant litter is known to stimu-late microbial activity (Fontaine et al., 2003), thereby triggeringthe uptake of nutrients by microbes and hence immobilizingnutrients for plant growth (Schmidt et al., 1997; Hodge et al.,2000; Van der Heijden et al., 2008). In our experiment, the roleof nutrient immobilization by the microbial community appearssmall because root addition had a limited impact on available soilnitrate and P concentrations. Therefore, even though our resultssupport previous findings that root addition often can have a

Table 1 Soil physiochemical properties (mean� SE; n = 18) and statistical results for these properties at the end of the conditioning phase

Functionalgroup

Rootaddition C (g kg�1) N (g kg�1)

C : N ratio(C : N)

NO3-N(mg kg�1)

NH4-N(mg kg�1)

Olsen P(mg kg�1) pH

Early-successional soilsGrass No 56.9� 10.5a 3.33� 0.65ab 17.4� 0.32a 0.17� 0.02a 2.71� 0.34bc 85.5� 2.18b 6.11� 0.04a

Yes 45.2� 9.67a 2.51� 0.61b 20.1� 3.52a 0.18� 0.09a 2.41� 0.27c 87.3� 2.11b 6.07� 0.03a

Forb No 54.2� 10.1a 3.16� 0.62ab 17.4� 0.30a 0.20� 0.07a 4.96� 0.38a 98.4� 8.66a 5.97� 0.05b

Yes 65.9� 10.8a 3.92� 0.68a 17.1� 0.26a 0.21� 0.09a 3.29� 0.32b 84.2� 2.94b 6.13� 0.03a

Mid-successional soilsGrass No 56.7� 8.04a 3.27� 0.49a 17.5� 0.26b 1.32� 0.40a 2.52� 0.26ab 87.3� 2.59a 6.00� 0.05b

Yes 29.7� 6.59b 1.65� 0.39b 18.5� 0.45a 0.69� 0.28b 3.04� 0.30a 86.3� 1.95a 6.09� 0.02a

Forb No 54.9� 11.7a 3.16� 0.71a 17.7� 0.32b 0.24� 0.09b 2.24� 0.18b 88.9� 2.57a 5.84� 0.05c

Yes 48.2� 7.80ab 2.77� 0.47ab 17.6� 0.21b 0.42� 0.16b 2.62� 0.28ab 87.1� 1.71a 6.05� 0.03ab

Source dfC N C : N ratio NO3-N NH4-N Olsen P pHF (P-value) F (P-value) F (P-value) F (P-value) F (P-value) F (P-value) F (P-value)

Early-successional soilsRoot (R) 1 0.00 (0.997) 0.01 (0.920) 1.32 (0.255) 0.01 (0.905) 26.8 (< 0.001) 5.05(0.028) 8.24 (0.005)Group (G) 1 2.31 (0.133) 2.83 (0.097) 2.22 (0.141) 0.38 (0.541) 67.8 (< 0.001) 3.16 (0.080) 3.20 (0.078)R9G 1 3.93 (0.052) 4.57 (0.036) 2.21 (0.142) 0.00 (0.947) 13.0 (< 0.001) 8.56 (0.005) 20.5 (< 0.001)

Mid-successional soilsRoot (R) 1 11.1 (0.001) 10.8 (0.002) 5.36 (0.024) 2.19 (0.144) 9.10 (0.004) 1.10 (0.299) 43.3 (< 0.001)Group (G) 1 2.74 (0.102) 2.75 (0.102) 3.80 (0.055) 20.4 (< 0.001) 5.48 (0.022) 0.90 (0.347) 18.1 (< 0.001)R9G 1 4.03 (0.049) 4.00 (0.049) 9.63 (0.003) 7.30 (0.009) 0.24 (0.628) 0.08 (0.776) 7.94 (0.006)

C, soil organic carbon; N, soil total nitrogen; C : N ratio, ratio of carbon to nitrogen; NO3-N, soil nitrate nitrogen; NH4-N, soil ammonium nitrogen; pH, soilpH; Olsen P, soil available phosphorus. Different lowercase letters within each column and within each successional stage refer to significant differences atP < 0.05 between treatments (root addition: no vs yes) and plant functional groups (grass vs forb).





negative impact on plant growth, the full mechanistic explanationneeds further studies aiming at teasing apart the effects of keymicrobial groups, that is, pathogens, mycorrhizal fungi anddecomposers (Van der Putten et al., 2016), as well as in-depthmeasurements of phytotoxic compounds including self-DNAand nutrients.

Longer term effects of root addition

In line with our second hypothesis, we found that longer termeffects of root addition generally had neutral to positive effects onplant biomass, thereby possibly reducing negative short-termeffects during soil conditioning. Roots of the plants that hadgrown in the conditioning phase were reintroduced into the soilsof the feedback phase, possibly causing short-term phytotoxicity(Meier et al., 2009), increased pathogen abundance (Van der Put-ten, 2003) and nutrient immobilization (Van der Heijden et al.,2008). All these effects may have reduced the feedback effects ofthe roots added before the conditioning phase. Therefore, eventhough our results may be conservative, the neutral to positiveeffects of root addition on plant growth indicate that decomposi-tion may ultimately counterbalance short-term negative PSFeffects. As such, our findings suggest that in grassland soils, wherepathogens play a role in driving PSF (Klironomos, 2002), rootlitter-mediated PSF has the potential to reduce, or even reversenegative PSF. Our results appear to be in contrast with recenttheoretical work showing that litter decomposibility is less impor-tant in driving PSF effects in cases where soil-borne pathogens

play a key role in the soil food web (Ke et al., 2015). However,our results indicate that litter-mediated effects appear to play outover longer time-scales, and this aspect may have been neglectedin many previous PSF studies (Miki et al., 2010; Van der Puttenet al., 2013).

We expected that longer term positive effects of root additionwould reduce negative short-term effects (e.g. as a result of thepresence of pathogens or parasites) via increased soil nutrientavailability (Wardle, 2002; Chapman et al., 2006; Kardol et al.,2015; Ke et al., 2015). However, we did not find a consistentincrease in soil N and P availability in response to root addition(Table 1). This may be attributable to slow decomposition ratesand nutrient release of root litter in comparison to leaf litter(Freschet et al., 2013; Hobbie, 2015) and to large differences inroot decomposition rates among plant species (Freschet et al.,2013). Still, nutrients other than the ones that we measured, suchas potassium, might have contributed to the positive effects ofroot addition on plant growth (Bezemer et al., 2006). Alternativepotential explanations involving competitive exclusion ofpathogens by other root-associated organisms such as mycorrhizalfungi (Wardle, 2006; Raaijmakers et al., 2009; Sikes et al., 2009)or relaxation of phytotoxocity (Bonanomi et al., 2006) do notprobably explain the PSF effects as measured in our study,because fresh roots produced in the conditioning phase were pre-sent in the soil at the beginning of the feedback phase as well.Further studies are needed in order to further reveal the underly-ing mechanisms of these short- and longer term effects of rootdecomposition.

Without root, grass With root, grassWith root, forbWithout root, forb

Root effects (R) NSFunctional group (G) *R × G NS

R × G NS

R NSG NS

R × G NS

R NSG NS

R × G NS

R NSG NS

R × G NS

R NSG NS

R × G NS

R NSG NS

1.0

0.8

0.6

Bac

teria

l and

fung

al b

iom

ass

(µg

PLF

A g–

1 )B

acte

rial a

nd fu

ngal

bio

mas

s(µ

g P

LFA

g–1 )

0.4

0.2

0.01.2

1.0

0.8

0.6

0.4

0.2

0.0Bacteria Fungi Arbuscular mycorrhizal fungi

(AMF)

aa

a

a

a aa

a

aa

aa

0

2

4

6

AM

F bi

omas

s (µ

g N

LFA

g–1 )

AM

F bi

omas

s (µ

g N

LFA

g–1 )

8

100

2

4

6

8

10

a

aa

a

aa

a a

a

a

a

a

(a)

(c)

(b)

(d)Fig. 3 Effects of root addition on relativebiomass of bacteria, fungi (lg PLFA g�1), andarbuscular mycorrhizal fungi (AMF;lg NLFA g�1) in soils conditioned by (a, b)early- and (c, d) mid-successional plantspecies. Values shown in the figure aremean� SE (n = 3). Statistically significanteffects: *, P < 0.05; NS, nonsignificanteffects. Different lowercase letters withineach type of microbial group and within eachsuccessional stage refer to significantdifferences at P < 0.05 between treatments(root addition: with vs without) and plantfunctional groups (grass vs forb). PLFA,phospholipid fatty acid; NLFA, neutral lipidfatty acid.



Research

NewPhytologist226

There was little evidence of species-specific effects of decom-posing roots on plant biomass production in the feedback phase,because the soil conditioning (home vs foreign)9 root additioninteraction was generally not significant, except in the case ofshoot production of mid-successional plant species. This impliesthat even though decomposition processes may be mediated byspecific decomposer communities that are specialized in breakingdown litter from plants in their immediate vicinity, referred to as‘home-field advantage’ (Gholz et al., 2000; Ayres et al., 2009;Veen et al., 2015), the feedback of this process to subsequentplant growth is apparently not species specific. By contrast, theeffect of soil conditioning by plant species on next-generationbiomass production was species specific, with plants often per-forming less well on soils conditioned by conspecifics than onsoils conditioned by heterospecifics, irrespective of root addition.This effect has been found often in PSF studies and is usuallyattributed to the accumulation of species-specific pathogens(Klironomos, 2002; Sikes et al., 2009).

Responses of early- vs mid-successional plant species

In opposition to our third hypothesis, we found no differencebetween early- and mid-succession plant species in their responsesto root addition in the conditioning phase. These results werenot expected based on previous studies by Kardol et al. (2006)and Van de Voorde et al. (2011) carried out in the same area.

Both studies by Kardol et al. (2006) and Van de Voorde et al.(2011) included more plant species and later successional speciesthan our study. As in our study the contrast between species andsuccessional stage was smaller, some pathogenic or parasitic fungimay have been shared by the early- and mid-succession plantspecies.

In contrast with our third hypothesis, we found that longerterm effects of decomposing roots increased the biomass of theearly-successional plant species in particular. In general, it hasbeen found that later successional plant species experiencemore neutral, or even positive PSF, which may favor theirreplacement of early-successional plant species (Kardol et al.,2006, 2007; Middleton & Bever, 2012). Our findings implythat such positive PSF effects in later successional stages arenot strongly mediated by root decomposition. Instead, positiveeffects of root addition were more apparent in early-successional plant species, which may have resulted from fasterlitter breakdown as a consequence of high litter quality andsoil nutrient availability (Kazakou et al., 2009), or of a moreprominent role of specialist decomposers in early successionalstages that can accelerate decomposition processes (Veen et al.,2015; Yu et al., 2015).

Our results imply that the replacement from early-successionalto later successional plant species does not depend strongly ondecomposition turning PSF from negative to positive. Instead,decreased susceptibility to pathogens or increased mycorrhizal

Early-successional stage1.0

0.8

0.6

cdcde

a

cde

e

f

b

bc

a

a

aab

ab ab

bb

b

aa

c

d

bb

c

abca

c

ab

bcabc

bc

0.4

0.2

0.02.0

1.5

1.0

0.5

0.0Grass GrassForb Forb

Roo

t bio

mas

s (g

per

pot

)S

hoot

bio

mas

s (g

per

pot

)

Soil effects (S) ***

S × R × G **

S × G **S × R NS

R × G NS

Functional group (G) ***Root effects (R) ***


S × R × G NS

S × G ***S × R NS

R × G NS

Functional group (G) ***Root effects (R) ***


S × R × G NS

S × G ***S × R NS

R × G NS

Functional group (G) NSRoot effects (R) NS

Soil effects (S) NS

S × R × G **

S × G ***S × R ***

R × G *

Functional group (G) NSRoot effects (R) NS

Mid-successional stageWithout roots, home Without roots, foreign With roots, home With roots, foreign

(a)

(c)

(b)

(d)Fig. 4 Effects of soil conditioning (with vswithout added roots) on shoot and rootbiomass (g dry weight per pot) of (a, c) early-and (b, d) mid-successional plant speciesafter the feedback phase grown in soil fromconspecifics (home) and heterospecifics(foreign). Values shown in the figure aremean � SE (n = 6). Statistically significanteffects: *, P < 0.05; **, P < 0.01; ***,P < 0.001; NS, nonsignificant effects.Different lowercase letters within eachsuccessional stage refer to significantdifferences at P < 0.05 in shoot and rootbiomass between treatments (root addition:with vs without; soil type: home vs foreign)and plant functional groups (grass vs forb).





biomass and/or activity may play a more important role in posi-tive PSF effects of later successional plant species (Kardol et al.,2006) than positive effects from root decomposition. This sug-gestion is also supported by the modeling results of Ke et al.(2015). However, in the field these PSF effects play out in com-plex communities and it remains to be tested how interactionswith neighboring species (e.g. Jing et al., 2015) and heterospecificmixtures of decomposing roots belowground modify PSF effects.Therefore, future studies need to explore the role of decomposi-tion across more realistic spatio-temporal scales relevant to fieldsituations (Van der Putten et al., 2016).

Responses of grasses vs forbs

In contrast with our fourth hypothesis, we did not find that thelonger term effects of root addition, in the feedback phase, weremore positive on forbs than grasses. Instead, we found that rootaddition had a stronger negative short-term effect, in the condi-tioning phase, on the biomass of forbs than on that of grasses. Inthe feedback phase, forb biomass was generally lower in homethan in foreign soils, while there was no difference for grassbiomass. These results indicate that forbs may experiencestronger negative PSF than grasses. Although these results are atodds with the findings of some studies comparing PSF effectsamong plant functional groups (Kardol et al., 2007; Kulmatiski& Kardol, 2008), they are consistent with the suggestion thatforbs often experience strong phytotoxicity (Bonanomi et al.,2006), which may explain, at least in part, the negative response

of forbs to root addition. Alternatively, the roots of forbs with rel-ative higher P content may have been more susceptible to soil-borne pathogens (Danhorn et al., 2004; Lalibert�e et al., 2015),thereby hosting more pathogens than grasses (Rottstock et al.,2014). In that case, root addition may have resulted in enhancedexposure to soil pathogens, provided that they survived the gentleair-drying of the roots. Consequently, root addition may havehad stronger negative effects on forb growth in the conditioningphase, and may have led to a stronger decrease in forb biomass inhome soil than in foreign soil in the feedback phase. Althoughforbs generally benefit more from AMF than grasses (Hoeksemaet al., 2010; Bunn et al., 2015), in some grass-dominated systems,like ours, the opposite pattern is observed (Wilson & Hartnett,1997; McCain et al., 2011). As a result, forbs may benefit lessfrom the presence of AMF in grassland systems, resulting instronger negative PSF effects.

Conclusions

In conclusion, we found that root addition initially reduced plantbiomass whereas in the longer term, in the feedback phase, rootaddition generally had neutral to positive effects. This suggeststhat the contribution of root decomposition to PSF varies withtime, and in the longer run may neutralize or counteract negativecomponents of PSF effects. In the short term, negative effects ofroot addition were stronger for forbs than for grasses, while thelonger term positive effect of root addition benefitted early-successional plants more than mid-successional plants.

Table 2 Chemical properties of roots collected from the field (mean� SE; n = 6) and ANOVA results for these properties

Functional group Species C (g kg�1) N (g kg�1) P (g kg�1) C : N ratio C : P ratio N : P ratio

Early-successional plant speciesG Lolium perenne 420� 4.02cd 7.45� 0.21ab 1.78� 0.10cde 56.7� 1.80e 241� 15.44c 4.23� 0.15bcd

G Holcus lanatus 423� 3.3c 8.67� 0.15a 1.75� 0.05cde 48.8� 0.51e 243� 5.92c 4.97� 0.10bc

G Arrhenatherum elatius 421� 1.05c 7.79� 0.50ab 1.97� 0.17bcd 55.1� 3.25e 221� 18.6cd 4.00� 0.15cde

F Jacobaea vulgaris 414� 1.70cde 4.20� 0.09cd 2.28� 0.04b 98.9� 2.05c 182� 2.52d 1.85� 0.05g

F Leucanthemum vulgare 402� 1.58ef 4.58� 0.15c 3.60� 0.19a 88.1� 2.67cd 113� 6.43e 1.29� 0.08g

F Rumex obtusifolius 406� 1.19de 4.86� 0.26c 1.61� 0.05de 84.7� 4.54d 254� 7.83bc 3.05� 0.21ef

Mid-successional plant speciesG Anthoxanthum odoratum 389� 7.48f 7.63� 0.14ab 1.91� 0.10bcd 51.1� 1.22e 207� 12.86cd 4.04� 0.22cde

G Agrostis capillaris 427� 2.62c 7.53� 0.45ab 1.43� 0.04e 57.8� 3.89e 299� 8.25b 5.24� 0.22b

G Deschampsia flexuosa 461� 2.48a 8.10� 0.33ab 0.61� 0.01f 57.4� 2.12e 756� 16.77a 13.3� 0.56a

F Hypochaeris radicata 406� 1.55de 3.16� 0.08d 1.78� 0.05cde 129� 3.07a 228� 6.00cd 1.78� 0.06g

F Plantago lanceolata 429� 2.34bc 7.36� 0.15b 2.06� 0.06bc 58.4� 0.88e 209� 6.40cd 3.59� 0.11de

F Rumex acetosella 442� 1.22b 3.90� 0.12cd 1.87� 0.07bcde 114� 3.77b 238� 9.90c 2.10� 0.09fg

Variable Source

Early-successional plants Mid-successional plants

df F P df F P

C Functional group 1 37.0 < 0.001 1 0.00 0.99N Functional group 1 194 < 0.001 1 36.5 < 0.001P Functional group 1 8.99 0.005 1 17.2 < 0.001C : N ratio Functional group 1 187 < 0.001 1 34.2 < 0.001C : P ratio Functional group 1 9.94 0.003 1 11.0 0.002N : P ratio Functional group 1 105 < 0.001 1 23.7 < 0.001

G, grass; F, forb; C, organic carbon; N, total nitrogen; P,total phosphorus. Different lowercase letters within each column refer to significant difference atP < 0.05 in root chemical properties between different plant species.



Research

NewPhytologist228

Our findings demonstrate that root decomposition can havea strong influence on the strength and direction of PSF effects.This implies that it will be crucial to consider root decomposi-tion processes more explicitly in future PSF studies. Moreover,our results indicate that longer term effects of root decomposi-tion do not result in more positive PSF for mid- than forearly-successional plant species and root decomposition is notlikely to contribute to the transition from early- to mid-successional stages via PSF. Future research is needed toaddress the relative importance of PSF mediated by livingroots and decomposing root (and shoot) litter pathways. It willbe crucial to focus on mechanisms underlying plant responsesto PSF mediated by living roots and litter and to tease apartthe roles of pathogens vs symbionts and decomposers, in driv-ing the strength and direction of plant–soil feedback effects,both on short and on longer time-scales.

Acknowledgements

N.Z. was subsidized by The China Scholarship Council (CSC)to visit the Netherlands Institute of Ecology (NIOO) and by theNational Natural Science Foundation of China (31270559).G.F.V. was funded by an NWO-VENI grant (no. 863.14.013).The glasshouse experiment was supported by TE funding ofNIOO. We thank Laiye Qu, Minggang Wang, Rebecca Pas andErik Reichman for their help with the plant harvests, and RutgerWilschut for his advice and help with collecting plant material inthe field. We thank Iris Chardon and Ciska Raaijmakers for per-forming the chemical analyses. We thank Gregor Disveld fortechnical support with the glasshouse experiments, and DrXiangcheng Mi for his suggestion on data analysis. This is NIOOpublication 6053.

Author contributions

N.Z., W.H.vdP. and G.F.V. contributed to the design of theexperiment, data analysis and writing of the manuscript. N.Z.was mainly responsible for establishment of the experiment,chemical analyses and plant harvest, with substantial contribu-tions from G.F.V.

References�Agren G, Hyv€onen R, Berglund S, Hobbie S. 2013. Estimating the critical N:C

from litter decomposition data and its relation to soil organic matter

stoichiometry. Soil Biology and Biochemistry 67: 312–318.An M, Pratley J, Haig T. 2001. Phytotoxicity of Vulpia residues: IV. Dynamics

of allelochemicals during decomposition of Vulpia residues and their

corresponding phytotoxicity. Journal of Chemical Ecology 27: 395–409.Augspurger CK, Kelly CK. 1984. Pathogen mortality of tropical tree seedlings:

experimental studies of the effects of dispersal distance, seedling density, and

light conditions. Oecologia 61: 211–217.Ayres E, Steltzer H, Simmons BL, Simpson RT, Steinweg JM, Wallenstein

MD, Mellor N, Parton WJ, Moore JC, Wall DH. 2009.Home–fieldadvantage accelerates leaf litter decomposition in forests. Soil Biology andBiochemistry 41: 606–610.

Bardgett RD, Mommer L, De Vries FT. 2014. Going underground: root traits as

drivers of ecosystem processes. Trends in Ecology & Evolution 29: 692–699.

Bever JD, Mangan SA, Alexander HM. 2015.Maintenance of plant species

diversity by pathogens. Annual Review of Ecology, Evolution, and Systematics 46:305–325.

Bever JD, Westover KM, Antonovics J. 1997. Incorporating the soil community

into plant population dynamics: the utility of the feedback approach. Journal ofEcology 85: 561–573.

Bezemer T, Fountain M, Barea J, Christensen S, Dekker S, Duyts H, Van Hal

R, Harvey J, Hedlund K, Maraun M. 2010. Divergent composition but

similar function of soil food webs of individual plants: plant species and

community effects. Ecology 91: 3027–3036.Bezemer TM, Lawson CS, Hedlund K, Edwards AR, Brook AJ, Igual JM,

Mortimer SR, Van der Putten WH. 2006. Plant species and functional group

effects on abiotic and microbial soil properties and plant–soil feedbackresponses in two grasslands. Journal of Ecology 94: 893–904.

Bligh EG, Dyer WJ. 1959. A rapid method of total lipid extraction and

purification. Canadian Journal of Biochemistry and Physiology 37: 911–917.Bonanomi G, Incerti G, Barile E, Capodilupo M, Antignani V, Mingo A,

Lanzotti V, Scala F, Mazzoleni S. 2011. Phytotoxicity, not nitrogen

immobilization, explains plant litter inhibitory effects: evidence from solid-state

13C NMR spectroscopy. New Phytologist 191: 1018–1030.Bonanomi G, Rietkerk M, Dekker SC, Mazzoleni S. 2005. Negative plant–soilfeedback and positive species interaction in a herbaceous plant community.

Plant Ecology 181: 269–278.Bonanomi G, Sicurezza MG, Caporaso S, Esposito A, Mazzoleni S. 2006.

Phytotoxicity dynamics of decaying plant materials. New Phytologist 169:571–578.

Bunn RA, Ramsey PW, Lekberg Y. 2015. Do native and invasive plants differ in

their interactions with arbuscular mycorrhizal fungi? A meta-analysis. Journal ofEcology 103: 1547–1556.

Cameron DD, White A, Antonovics J. 2009. Parasite–grass–forb interactionsand rock–paper–scissor dynamics: predicting the effects of the parasitic plant

Rhinanthus minor on host plant communities. Journal of Ecology 97: 1311–1319.

Castelli JP, Casper BB. 2003. Intraspecific AM fungal variation contributes to

plant-fungal feedback in a serpentine grassland. Ecology 84: 323–336.Chapman SK, Langley JA, Hart SC, Koch GW. 2006. Plants actively control

nitrogen cycling: uncorking the microbial bottleneck. New Phytologist 169: 27–34.

Danhorn T, Hentzer M, Givskov M, Parsek MR, Fuqua C. 2004. Phosphorus

limitation enhances biofilm formation of the plant pathogen Agrobacteriumtumefaciens through the PhoR-PhoB regulatory system. Journal of Bacteriology186: 4492–4501.

Dhillon SS, Miksche JP. 1982. DNA content and heterochromatin variations in

various tissues of peanut (Arachis hypogaea). American Journal of Botany 69:219–226.

Fontaine S, Mariotti A, Abbadie L. 2003. The priming effect of organic matter: a

question of microbial competition? Soil Biology and Biochemistry 35: 837–843.Freschet GT, Cornwell WK, Wardle DA, Elumeeva TG, Liu W, Jackson BG,

Onipchenko VG, Soudzilovskaia NA, Tao J, Cornelissen JH. 2013. Linking

litter decomposition of above- and below-ground organs to plant–soilfeedbacks worldwide. Journal of Ecology 101: 943–952.

Frosteg�ard�A, B�a�ath E. 1996. The use of phospholipid fatty acid analysis to

estimate bacterial and fungal biomass in soil. Biology and Fertility of Soils 22:59–65.

Frosteg�ard�A, Tunlid A, B�a�ath E. 1991.Microbial biomass measured as total

lipid phosphate in soils of different organic content. Journal of MicrobiologicalMethods 14: 151–163.

Gholz HL, Wedin DA, Smitherman SM, Harmon ME, Parton WJ. 2000. Long-

term dynamics of pine and hardwood litter in contrasting environments:

toward a global model of decomposition. Global Change Biology 6: 751–765.Griffin G, Jokela W, Ross D, Pettinelli D, Morris T, Wilf A. 2009.

Recommended soil nitrate tests. In: Sims J, Wolf A, eds. Recommended soiltesting procedures for the northeastern US, 3rd edn. Northeastern Regional

Publication 493: 27–38.Hobbie SE. 2015. Plant species effects on nutrient cycling: revisiting litter

feedbacks. Trends in Ecology & Evolution 30: 357–363.





Hodge A. 2004. The plastic plant: root responses to heterogeneous supplies of

nutrients. New Phytologist 162: 9–24.Hodge A, Stewart J, Robinson D, Griffiths B, Fitter A. 2000. Competition

between roots and soil micro-organisms for nutrients from nitrogen-rich

patches of varying complexity. Journal of Ecology 88: 150–164.Hoeksema JD, Chaudhary VB, Gehring CA, Johnson NC, Karst J, Koide RT,

Pringle A, Zabinski C, Bever JD, Moore JC. 2010. A meta-analysis of context-

dependency in plant response to inoculation with mycorrhizal fungi. EcologyLetters 13: 394–407.

Huber D, Watson R. 1970. Effect of organic amendment on soil-borne plant

pathogens. Phytopathology 60: 22–26.Jasper D, Abbott L, Robson A. 1989.Hyphae of a vesicular–arbuscularmycorrhizal fungus maintain infectivity in dry soil, except when the soil is

disturbed. New Phytologist 112: 101–107.Jing J, Bezemer TM, Van der Putten WH. 2015. Complementarity and selection

effects in early and mid-successional plant communities are differentially

affected by plant–soil feedback. Journal of Ecology 103: 641–647.Johnson N, Graham JH, Smith F. 1997. Functioning of mycorrhizal

associations along the mutualism–parasitism continuum. New Phytologist135: 575–585.

Kardol P, Bezemer T, Van der Wal A, Van der Putten WH. 2005. Successional

trajectories of soil nematode and plant communities in a chronosequence of ex-

arable lands. Biological Conservation 126: 317–327.Kardol P, Cornips NJ, Van Kempen MM, Bakx-Schotman JT, Van der Putten

WH. 2007.Microbe-mediated plant–soil feedback causes historicalcontingency effects in plant community assembly. Ecological Monographs 77:147–162.

Kardol P, Martijn Bezemer T, Van der Putten WH. 2006. Temporal variation

in plant–soil feedback controls succession. Ecology Letters 9: 1080–1088.Kardol P, Veen GF, Teste FP, Perring MP. 2015. Peeking into the black box: a

trait-based approach to predicting plant–soil feedback. New Phytologist 206:1–4.

Kazakou E, Violle C, Roumet C, Pintor C, Gimenez O, Garnier E. 2009.

Litter quality and decomposability of species from a Mediterranean succession

depend on leaf traits but not on nitrogen supply. Annals of Botany 104:1151–1161.

Ke PJ, Miki T, Ding TS. 2015. The soil microbial community predicts the

importance of plant traits in plant–soil feedback. New Phytologist 206: 329–341.

Keeney DR, Nelson DW. 1982. Nitrogen-inorganic forms. In: Page AL, Miller

RH, Keeney DR, eds.Methods of soil analysis. Part 2 – chemical andmicrobiological properties, 2nd edn. Agronomy 9: 643–698.

Klironomos JN. 2002. Feedback with soil biota contributes to plant rarity and

invasiveness in communities. Nature 417: 67–70.Klironomos JN. 2003. Variation in plant response to native and exotic arbuscular

mycorrhizal fungi. Ecology 84: 2292–2301.Koziol L, Bever JD. 2015.Mycorrhizal response trades off with plant growth rate

and increases with plant successional status. Ecology 96: 1768–1774.Kull O, Aan A. 1997. The relative share of graminoid and forb life-forms in a

natural gradient of herb layer productivity. Ecography 20: 146–154.Kulmatiski A, Kardol P. 2008. Getting plant–soil feedbacks out of thegreenhouse: experimental and conceptual approaches. In: L€uttge U, Beyschlag

W, Murata J, eds. Progress in botany. Heidelberg/Berlin, Germany: Springer,

449–472.Lalibert�e E, Lambers H, Burgess TI, Wright SJ. 2015. Phosphorus limitation,

soil-borne pathogens and the coexistence of plant species in hyperdiverse forests

and shrublands. New Phytologist 206: 507–521.Mangan SA, Schnitzer SA, Herre EA, Mack KM, Valencia MC, Sanchez EI,

Bever JD. 2010. Negative plant–soil feedback predicts tree-species relativeabundance in a tropical forest. Nature 466: 752–755.

Mazzoleni S, Bonanomi G, Incerti G, Chiusano ML, Termolino P, Mingo A,

Senatore M, Giannino F, Carten�ı F, Rietkerk M. 2015. Inhibitory and toxic

effects of extracellular self-DNA in litter: a mechanism for negative plant–soilfeedbacks? New Phytologist 205: 1195–1210.

McCain KN, Wilson GW, Blair J. 2011.Mycorrhizal suppression alters plant

productivity and forb establishment in a grass-dominated prairie restoration.

Plant Ecology 212: 1675–1685.

Meier CL, Keyserling K, Bowman WD. 2009. Fine root inputs to soil reduce

growth of a neighbouring plant via distinct mechanisms dependent on root

carbon chemistry. Journal of Ecology 97: 941–949.Middleton EL, Bever JD. 2012. Inoculation with a native soil community

advances succession in a grassland restoration. Restoration Ecology 20: 218–226.Miki T, Kondoh M. 2002. Feedbacks between nutrient cycling and vegetation

predict plant species coexistence and invasion. Ecology Letters 5: 624–633.Miki T, Ushio M, Fukui S, Kondoh M. 2010. Functional diversity of microbial

decomposers facilitates plant coexistence in a plant–microbe–soil feedbackmodel. Proceedings of the National Academy of Sciences, USA 107: 14251–14256.

Milcu A, Manning P. 2011. All size classes of soil fauna and litter quality control

the acceleration of litter decay in its home environment. Oikos 120: 1366–1370.

Murphy J, Riley JP. 1962. A modified single solution method for the

determination of phosphate in natural waters. Analytica Chimica Acta 27:31–36.

Olsen SR, Cole CV, Watanabe FS, Dean LA. 1954. Estimation of availablephosphorus in soils by extraction with sodium bicarbonate. Washington, DC,

USA: USDA: USDA Circ. 939.

Olsson PA. 1999. Signature fatty acids provide tools for determination of the

distribution and interactions of mycorrhizal fungi in soil. FEMS MicrobiologyEcology 29: 303–310.

Petermann JS, Fergus AJ, Turnbull LA, Schmid B. 2008. Janzen-Connell effects

are widespread and strong enough to maintain diversity in grasslands. Ecology89: 2399–2406.

Poorter H, Niklas KJ, Reich PB, Oleksyn J, Poot P, Mommer L. 2012. Biomass

allocation to leaves, stems and roots: meta-analyses of interspecific variation and

rnvironmental control. New Phytologist 193: 30–50.R Core Team. 2014. R: a language and environment for statistical computing.Vienna, Austria: R Foundation for Statistical Computing. [WWW document]

URL http://www.R-project.org/ [accessed 10 April 2014].

Raaijmakers JM, Paulitz TC, Steinberg C, Alabouvette C, Mo€enne-Loccoz Y.

2009. The rhizosphere: a playground and battlefield for soilborne pathogens

and beneficial microorganisms. Plant and Soil 321: 341–361.Reinhart KO, Tytgat T, Van der Putten WH, Clay K. 2010. Virulence of soil-

borne pathogens and invasion by Prunus serotina. New Phytologist 186: 484–495.

Revilla TA, Veen GC, Eppinga MB, Weissing FJ. 2013. Plant–soil feedbacksand the coexistence of competing plants. Theoretical Ecology 6: 99–113.

Rottstock T, Joshi J, Kummer V, Fischer M. 2014.Higher plant diversity

promotes higher diversity of fungal pathogens, while it decreases pathogen

infection per plant. Ecology 95: 1907–1917.Schmidt IK, Michelsen e, Jonasson S. 1997. Effects of labile soil carbon on

nutrient partitioning between an arctic graminoid and microbes. Oecologia 112:557–565.

Sikes BA, Cottenie K, Klironomos JN. 2009. Plant and fungal identity

determines pathogen protection of plant roots by arbuscular mycorrhizas.

Journal of Ecology 97: 1274–1280.Singh H, Batish DR, Kohli R. 1999. Autotoxicity: concept, organisms, and

ecological significance. Critical Reviews in Plant Sciences 18: 757–772.Terborgh J. 2012. Enemies maintain hyperdiverse tropical forests. The AmericanNaturalist 179: 303–314.

Turner C, Kneisler J, Knapp A. 1995. Comparative gas exchange and nitrogen

responses of the dominant C4 grass Andropogon gerardii and five C3 forbs to

fire and topographic position in tallgrass prairie during a wet year. InternationalJournal of Plant Sciences 156: 216–226.

Van de Voorde TF, Van der Putten WH, Bezemer TM. 2011. Intra- and

interspecific plant–soil interactions, soil legacies and priority effects during old-field succession. Journal of Ecology 99: 945–953.

Van der Heijden MG, Bardgett RD, Van Straalen NM. 2008. The unseen

majority: soil microbes as drivers of plant diversity and productivity in

terrestrial ecosystems. Ecology Letters 11: 296–310.Van der Putten WH. 2003. Plant defense belowground and spatiotemporal

processes in natural vegetation. Ecology 84: 2269–2280.Van der Putten WH, Bardgett RD, Bever JD, Bezemer TM, Casper BB,

Fukami T, Kardol P, Klironomos JN, Kulmatiski A, Schweitzer JA. 2013.



Research

NewPhytologist230

http://www.R-project.org/

Plant–soil feedbacks: the past, the present and future challenges. Journal ofEcology 101: 265–276.

Van der Putten WH, Bradford MA, Brinkman EP, van de Voorde TFJ, Veen

GF. 2016.Where, when and how plant-soil feedback matters in a changing

world. Functional Ecology. doi: 10.1111/1365-2435.12657.Van der Putten WH, Mortimer S, Hedlund K, Van Dijk C, Brown V, Lep€a J,Rodriguez-Barrueco C, Roy J, Len TD, Gormsen D. 2000. Plant species

diversity as a driver of early succession in abandoned fields: a multi-site

approach. Oecologia 124: 91–99.Van der Wal A, Van Veen JA, Smant W, Boschker HT, Bloem J, Kardol P, Van

der Putten WH, de Boer W. 2006. Fungal biomass development in a

chronosequence of land abandonment. Soil Biology and Biochemistry 38:51–60.

Van Grunsven RHA, Van der Putten WH, Bezemer TM, Veenendaal EM.

2010. Plant–soil feedback of native and range-expanding plant species isinsensitive to temperature. Oecologia 162: 1059–1069.

Veen GF, Freschet GT, Ordonez A, Wardle DA. 2015. Litter quality and

environmental controls of home-field advantage effects on litter decomposition.

Oikos 124: 187–195.Wardle DA. 2002. Communities and ecosystems: linking the aboveground andbelowground components. Princeton, NJ, USA and Oxford, UK: Princeton

University Press.

Wardle DA. 2006. The influence of biotic interactions on soil biodiversity.

Ecology Letters 9: 870–886.Wardle DA, Bardgett RD, Klironomos JN, Set€al€a H, Van der Putten WH, Wall

DH. 2004. Ecological linkages between aboveground and belowground biota.

Science 304: 1629–1633.White D, Davis W, Nickels J, King J, Bobbie R. 1979. Determination of the

sedimentary microbial biomass by extractible lipid phosphate. Oecologia 40:51–62.

Wilson G, Hartnett D. 1997. Effects of mycorrhizae on plant growth and dynamics

in experimental tall grass prairie microcosms. American Journal of Botany 84: 478.Yu Z, Huang Z, Wang M, Liu R, Zheng L, Wan X, Hu Z, Davis MR, Lin T-C.

2015. Nitrogen addition enhances home-field advantage during litter

decomposition in subtropical forest plantations. Soil Biology and Biochemistry90: 188–196.

Zak DR, Holmes WE, White DC, Peacock AD, Tilman D. 2003. Plant

diversity, soil microbial communities, and ecosystem function: are there any

links? Ecology 84: 2042–2050.Zak DR, Ringelberg DB, Pregitzer KS, Randlett DL, White DC, Curtis PS.

1996. Soil microbial communities beneath Populus grandidentata grown under

elevated atmospheric CO2. Ecological Applications 6: 257–262.

Supporting Information

Additional Supporting Information may be found online in theSupporting Information tab for this article:

Fig. S1 Effects of root addition on shoot biomass (g dry weightper pot) of both early (a) and mid (b) successional plant speciesin the conditioning phase.

Please note: Wiley Blackwell are not responsible for the contentor functionality of any supporting information supplied by theauthors. Any queries (other than missing material) should bedirected to the New Phytologist Central Office.

New Phytologist is an electronic (online-only) journal owned by the New Phytologist Trust, a not-for-profit organization dedicatedto the promotion of plant science, facilitating projects from symposia to free access for our Tansley reviews.

Regular papers, Letters, Research reviews, Rapid reports and both Modelling/Theory and Methods papers are encouraged. We are committed to rapid processing, from online submission through to publication ‘as ready’ via Early View – our average timeto decision is <28 days. There are no page or colour charges and a PDF version will be provided for each article.

The journal is available online at Wiley Online Library. Visit www.newphytologist.com to search the articles and register for tableof contents email alerts.

If you have any questions, do get in touch with Central Office ([email protected]) or, if it is more convenient,our USA Office ([email protected])

For submission instructions, subscription and all the latest information visit www.newphytologist.com





http://dx.doi.org/10.1111/1365-2435.12657

effects of root decomposition on plant soil feedback of ... · effects of root decomposition on...

Documents