1 23

Mycorrhiza ISSN 0940-6360Volume 25Number 2 Mycorrhiza (2015) 25:121-130DOI 10.1007/s00572-014-0594-3

Arbuscular mycorrhiza mediates glomalin-related soil protein production and soilenzyme activities in the rhizosphere oftrifoliate orange grown under different PlevelsQiang-Sheng Wu, Yan Li, Ying-NingZou & Xin-Hua He

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag Berlin Heidelberg. This e-offprint is

for personal use only and shall not be self-

archived in electronic repositories. If you wish

to self-archive your article, please use the

accepted manuscript version for posting on

your own website. You may further deposit

the accepted manuscript version in any

repository, provided it is only made publicly

available 12 months after official publication

or later and provided acknowledgement is

given to the original source of publication

and a link is inserted to the published article

on Springer's website. The link must be

accompanied by the following text: "The final

publication is available at link.springer.com”.

ORIGINAL PAPER

Arbuscular mycorrhiza mediates glomalin-related soil proteinproduction and soil enzyme activities in the rhizosphereof trifoliate orange grown under different P levels

Qiang-Sheng Wu & Yan Li & Ying-Ning Zou & Xin-Hua He

Received: 4 December 2013 /Accepted: 3 July 2014 /Published online: 18 July 2014# Springer-Verlag Berlin Heidelberg 2014

Abstract Glomalin-related soil protein (GRSP) is beneficial tosoil and plants and is affected by various factors. To addresswhether mycorrhizal-induced GRSP and relevant soil enzymesdepend on external P levels, a pot study evaluated effects of thearbuscular mycorrhizal fungus (AMF) Funneliformis mosseaeon GRSP production and soil enzyme activities. Three GRSPcategories, as easily-extractable GRSP (EE-GRSP), difficultly-extractable GRSP (DE-GRSP), and total (EE-GRSP+DE-GRSP) GRSP (T-GRSP), were analyzed, together with fiveenzyme activities (β-glucosidase, catalase, peroxidase,phosphatase, polyphenol oxidase) in the rhizosphere of trifoliateorange (Poncirus trifoliata) grown under 0, 3, and 30 mMKH2PO4 in a sand substrate. After 4 months, root AM coloni-zation and substrate hyphal length decreased with increasing Plevels. Shoot, root, and total biomass production was signifi-cantly increased by AM colonization, regardless of P levels, butmore profound under 0 mM P than under 30 mM KH2PO4. Ingeneral, production of these three GRSP categories under 0 or30 mM KH2PO4 was similar in non-mycorrhizosphere butdecreased in mycorrhizosphere. Mycorrhization significantlyincreased the production of EE-GRSP, DE-GRSP and T-GRSP, soil organic carbon (SOC), and activity of substrate β-glucosidase, catalase, peroxidase, and phosphatase, but de-creased polyphenol oxidase activity, irrespective of P levels.

Production of EE-GRSP, DE-GRSP, and T-GRSP significantlypositively correlated with SOC and β-glucosidase, catalase, andperoxidase activity, negatively with polyphenol oxidase activity,but not with hyphal length or phosphatase activity. These resultsindicate that AM-mediated production of GRSP and relevantsoil enzyme activities may not depend on external Pconcentrations.

Keywords Citrus . Glomalin .Mycorrhizas . Phosphorus .

Soil enzymes

Introduction

Glomalin is a protein secreted by the hyphae and spores ofarbuscular mycorrhizal (AM) fungi and which is quantifiedfrom soil as glomalin-related soil protein (GRSP) (Wright andUpadhyaya 1996; Rillig 2004; Singh et al. 2013). In general,GRSP is categorized as easily-extractable glomalin-relatedsoil protein (EE-GRSP) and total glomalin-related soil protein(T-GRSP), which are extracted at 0.11 Mpa with 20.0 mMcitrate (pH 7.0) for 30 min and 50 mM citrate (pH 8.0) formin, respectively (Wright and Upadhyaya 1996). The charac-teristics of GRSP are hydrophobicity, insolubility in water,and contents of carbon (C), nitrogen (N), P, hydrogen, oxygen,and iron (Schindler et al. 2007). GRSP plays important rolesin AM fungal effects on soil aggregation (Wright andAnderson 2000; Rillig 2004; Emran et al. 2012; Fokomet al. 2012), C storage (Rillig et al. 2001; Fokom et al.2012), cellular function such as chaperonins (Purin andRillig 2007), and stress tolerance (Hammer and Rillig 2011).The production of GRSP is affected by various factors such ashost plant species, photosynthesis and productivity, soil water-stable aggregate fractions, and soil management and physico-chemical traits (Treseder and Turner 2007; Violi et al. 2007;Wang et al. 2011; Wu et al. 2013).

Q.<S. Wu (*) :Y. Li :Y.<N. ZouCollege of Horticulture and Gardening, Yangtze University,Jingzhou, Hubei 434025, Chinae-mail: wuqiangsh@163.com

X.<H. He (*)School of Plant Biology, University of Western Australia, Crawley,WA 6009, Australiae-mail: xinhua.he@uwa.edu.au

X.<H. HeDepartment of Environmental Sciences, University of Sydney,Eveleigh, NSW 2015, Australia

Mycorrhiza (2015) 25:121–130DOI 10.1007/s00572-014-0594-3

Author's personal copy

GRSP was recently divided into fractions 1 and 2 ofBradford-reactive soil protein (BRSP) by Koide and Peoples(2013). In general, fraction 1 of BRSP corresponds to thatpreviously known as EE-GRSP that is newly produced andrelatively labile, while fraction 2 of BRSP is an older, recalcitrant,and more difficultly-extractable form of GRSP (DE-GRSP) thatremains longer in soil. DE-GRSP is extracted from the residueremaining from EE-GRSP extraction at 0.11 Mpa with 50 mMcitrate (pH 8.0) for 60 min (Koide and Peoples 2013). The totalGRSP is the sum of EE-GRSP and DE-GRSP. This new classi-fication ofDE-GRSP provides a new functional fraction betweenEE-GRSP and T-GRSP. Some studies have found highly positivecorrelations between GRSP concentrations and Olsen P concen-trations in soils of northwest and northern China (Bai et al. 2009;Guo et al. 2012), while a significantly negative correlation be-tween EE-GRSP and soil Olsen P was found in a Ultisol fromCentral-South Chile (Valarini et al. 2009). However, there are noreports about relationships between EE-GRSP and DE-GRSP inarbuscular mycorrhizal fungi (AMF) or non-AMF plants grownunder different levels of soil P.

Enzymes play important roles in soil nutrient cycling, andtheir activity represents soil microbial activity or an indicationof soil quality (Badiane et al. 2001). Among several differentsoil enzymes, catalase (CAT) catalyzes oxidation of organiccompounds, such as amines and phenols, in presence of H2O2

and is suggested to be an index of soil fertility (Garcia et al.2000). Peroxidases (PODs), released by fungi, depolymerizelignin and play a vital role in the removal of toxic substratesfrom the environment (Sinsabaugh 2010), while polyphenoloxidase (PPO) inactivates tannins (Adamczyk et al. 2009).Soil β-glucosidase (β-Gl), involved in cellulose degradation,is regarded as an early indicator of changes in organic matterstatus and soil quality (Stege et al. 2010). Phosphatases hy-drolyze organic P to release inorganic orthophosphate for cropuptake and highly correlate with the biological activity drivingP partitioning and uptake (Brooks et al. 2013).

AM fungi are also regarded as a source of soil enzymeactivities for biochemical reactions since they can increase theactivity of soil β-Gl, CAT, dehydrogenase, phosphatase, pro-tease, urease, etc. (Garcia et al. 2000; Raiesi and Ghollarata2006; Wang et al. 2006; Wu et al. 2008; Alguacil et al. 2009;Huang et al. 2009). GRSPs correlate highly with β-glucosidase and protease activities in soil (Gispert et al.2013) or the citrus rhizosphere (Xanthi-udic Ferralsol) ofChina (Wu et al. 2012). Interestingly, Raiesi and Ghollarata(2006) reported that inoculation of clover with Glomusintraradices significantly increased soil alkaline phosphataseactivity, irrespective of the substrate P levels, whereas a higheracid phosphatase activity was only observed under no-P butnot under P fertilization. These results suggest that, in additionto soil P effects on AM functioning (Fredeen and Terry 1988;Schmidt et al. 2010; Beltrano et al. 2013; Tong et al. 2013;Vasconcellos et al. 2013), the substrate P level might also

interact with AM effects on soil enzyme activities. However,few studies have evaluated the interactive effects of both AMinoculation and substrate P levels on activities of soil en-zymes. In addition, it is not clear whether soil enzyme activ-ities used to evaluate soil quality correlate with GRSPs, al-though GRSP has been regarded as a vital index of soil quality(Rillig 2004; Wu et al. 2012; Vasconcellos et al. 2013).

Citrus is one of the most important fruit trees with ~100million tons fruit production per year around the world. Thecitrus industry in southern China plays an important role inlocal economic development in mountainous regions and fruitsupply to the whole of China. In the field, citrus plants highlydepend onAM symbiosis for the uptake of water and nutrientsfrom soil (Wu et al. 2008). The objectives of the present studywere therefore to determine whether AM effects on relevantsoil enzymes and GRSP fractions are dependent on substrate Plevels, and whether GRSPs relate to soil enzyme activity in themycorrhizosphere of trifoliate orange seedlings.

Materials and methods

Experimental design

The experiment set-up was a completely randomized blockdesign with a 2×3 factorial combination of three P levels [0, 3,30 mMPKH2PO4] in acid-washed sand (from the riverside ofYangtze River near Jingzhou City) and plants with or withoutmycorrhizal inoculation. Each treatment had four replicatesfor a total of 24 pots.

Plant and fungus cultivation

Seeds of trifoliate orange [Poncirus trifoliata (L.) Raf.] weresurface-disinfected with 70 % ethanol for 15 min, rinsed withdistilled water, and then germinated in plastic boxes containingautoclaved sand. The germination conditions were 28/20 °C(day/night), 745 μM/m−2/s photosynthetic photon flux density,and 80 % relative humidity in a controlled growth chamber.Seedlings with similar size (five leaves) were transplanted intoplastic pots (20×15×18 cm) filled with acid-washed sand(<4 mm), which was mixed with 60 g of inoculum ofFunneliformis mosseae (Nicol. & Gerd.) Schüßler & Walker[BGC XZ02A] prior to transplanting. This strain of F. mosseaewas isolated from the rhizosphere of Incarvillea younghusbandiifrom Dangxiong (30° 31′ 04″ N and 90° 45′ 46″ E, a.s.l.~4,290 m), Tibet, China. Fungal spores were propagated withwhite clover (Trifolium repens) for 16 weeks and inoculum wascomposed of AM roots, spores (~23 spores/g), hyphae, andsand. Non-AM control treatments received 60 g autoclaved(121 °C, 0.11 Mpa, 2 h) mycorrhizal inoculum plus 2 ml filtrate(25 μm) of mycorrhizal inoculum to minimize differences inother microbial communities.

122 Mycorrhiza (2015) 25:121–130

Author's personal copy

Seedlings were acclimatized 2 weeks with 100 mL distilledwater daily per pot, then treated daily with 100 mL standardHoagland solution containing 0, 3, or 30 mMKH2PO4 per pot(referred to as P0, P3, and P30). These P levels are considered aslow, medium, and high level, respectively, for citrus based onprevious reports (Jifon et al. 2002; Guo et al. 2003). Seedlingswere grown for 4 months (April 1 to August 1, 2012) in aplastic greenhouse, which had an average 768 μmol/m2/sphotosynthetic photon flux density, 28/21 °C day/night tem-perature, and 85 % relative air humidity.

Determination of biomass, root mycorrhizal colonization,and hyphal length

At harvest, shoots and roots were separated and oven-dried at70 °C for 48 h. Determination of mycorrhizal colonization offresh roots was conducted according to Phillips and Hayman(1970), and expressed as the percent of AM colonized rootlength against total root length. Hyphal length in the substratewas measured using the protocol of Bethlenfalvay and Ames(1987).

Determinations of SOC and GRSP

Rhizospheric substrate, defined as sand substrate adhering toroots, was collected, mixed, air-dried, and sieved (2 mm) forfurther analyses. SOC was determined by the dichromateoxidation spectrophotometric method (Rowell 1994). EE-GRSP and DE-GRSP in the substrates were determined ac-cording to Koide and Peoples (2013). Briefly, for EE-GRSP,1.0 g substrate sample was incubated with 8 mL 20mM citrate(pH 7.0), autoclaved at 121 °C and 0.11 Mpa for 30 min, andthen centrifuged at 10,000×g for 3 min. After the collection ofEE-GRSP from the supernatants, DE-GRSP was subsequent-ly extracted from the remaining residue with 8 mL 50 mMcitrate (pH 8.0) for 60 min and centrifuged at 10,000×g for3 min. The supernatants of the EE-GRSP and DE-GRSP wereseparately assayed with bovine serum albumin as the standardaccording to Bradford (1976). Total GRSP was the sum of theEE-GRSP and DE-GRSP.

Substrate enzyme determinations

β-Gl activity (mg glucose/g DW substrate) was determinedusing a colorimetric method (Guan 1986). Briefly, 5.0 gair-dried substrate was incubated at 37 °C with 15 mL 8 %sucrose, 5 mL 0.67 mM phosphate buffer (pH 5.5), and1 mL methylbenzene for 24 h, and then filtered. Themixture containing 1 mL filtrate and 3 mL 3,5-dinitrosalicylic acid was further incubated for 5 min at100 °C and absorbance was measured at 508 nm withcontrols containing no substrate or soil.

CAT activity (mL 0.1 M KMnO4/g DW substrate) wasassayed according to a titration method (Yan 1988). Briefly,2.0 g of substrate was well mixed with 40 mL distilled waterand 5 mL 0.3 % hydrogen peroxide. After 20 min, 5 mL 3 Msulfuric acid was added and the solution filtered. Twenty-fivemilliliter filtrates were titrated with 0.1 M KMnO4 to obtain apink color solution.

Analyses of peroxidase (POD) and polyphenol oxidase(PPO) activities (mg pyrogallic acid/g DW substrate for bothenzymes) was performed accorded to Guan (1986). Briefly,1.0 g of substrate was incubated at 30 °C for 2 h with 10 mL1 % pyrogallic acid (for POD, 2 mL 0.5 % hydrogen peroxidewas added) and then 4 mL citrate-phosphate buffer (pH 4.5)and 35 mL diethyl ether were added. After 30 min extraction,absorbance of the diethyl ether phase was measured at 430 nmwith controls containing no substrate or no soil.

Phosphatase activity (mg hydroxybenzene/g DW sub-strate) was analyzed by the method described by Zhao andJiang (1986). Briefly, 1.0 g of substrate was incubated with0.4 mL methylbenzene for 15 min, then 5 mL 0.58 %disodium phenyl phosphate and 5 mL 0.1 M sodium acetatebuffer (pH 5.0) were added and incubated at 37 °C for 12 h.The reaction liquid was diluted with distilled water to 50 mLand filtered. A 0.25-mL filtrate was added to 5 mL 0.1 Msodium acetate buffer (pH 5.0), 3 mL 2.5 % K3Fe(CN)6, and3 mL 0.5 % 4-aminoantipyrine, incubated for 30 min at 30 °Cand diluted with distilled water to 25 mL. Absorbance wasmeasured at 570 nm with hydroxybenzene solution as thestandard.

Statistical analysis

Statistical analyses of data (means±SE, n=4) were performedusing the SAS software (v 8.1), and two-way variance(ANOVA) was used to compare significant differences be-tween treatments with the Duncan’s multiple range test atP<0.05. The Pearson’s correlation coefficients between vari-ables were performed using the Proc Corr’s procedure of SAS.

Results

Root AM colonization and substrate hyphal length

Substrate P levels strongly affected AM fungal colonization ofroots and substrate hyphal length (Table 1). Mycorrhizal col-onization ranged from 13.5 to 54.7% root length and substratehyphal length from 0.15 to 4.83 m/g FW soil for F. mosseae-inoculated seedlings (Table 2), with both ranking as P0>P3>P30 among P treatments. No mycorrhizal colonization orsubstrate hyphae were associated to the no-F. mosseae-inocu-lated plants.

Mycorrhiza (2015) 25:121–130 123

Author's personal copy

Plant biomass production

Shoot, root, and total (shoot+root) biomass production weresignificantly decreased in the P0 treatment compared to the P3treatment, irrespective of AM status (Table 2). When grown atP30, non-mycorrhizal seedlings had significantly lower shootand total biomass and mycorrhizal seedlings higher root bio-mass than those grown at P3. AM seedlings displayed signif-icantly greater biomass than non-AM seedlings at P0, P3, andP30, in a range of 27, 13, and 21 % for shoot biomass; 18, 10,and 17 % for root biomass; and 25, 13, and 20 % for totalbiomass in (Table 2). A significant interaction between AMfungal inoculation and P treatment occurred only for totalbiomass (Table 1).

GRSPs and SOC

EE-GRSP in the rhizosphere of non-AM seedlings was sig-nificantly increased but DE-GRSP and T-GRSP in themycorrhizosphere of AM seedlings was significantly de-creased under the P0 level as compared to P3 (Table 3). Thethree GRSP fractions were not altered in the rhizosphere ofnon-AM seedlings under the P30 treatment but significantlydecreased in the mycorrhizosphere of AM seedlings, as com-pared with the P3 treatment. Mycorrhization significantlyincreased concentrations of all GRSP fractions in the rhizo-sphere of trifoliate orange, irrespective of the P levels.Compared to non-mycorrhizal plants, AM fungal colonization

under P0, P3, and P30 increased EE-GRSP by 16, 52, and30 %; DE-GRSP by 10, 45, and 17 %; and T-GRSP by 11,46, and 18 %, respectively. There was a significant interactionbetween AM fungal inoculation and P treatment for the threeGRSP fractions (Table 1).

Concentrations of SOC were significantly higher in the P3and P30 treatments than in P0, whether plants were inoculatedor not with AMF (Table 3). Compared to the non-mycorrhizaltreatment, AM fungal inoculation increased SOC concentra-tions by 103, 42, and 42 % under P0, P3, and P30, respectively.

Rhizosphere enzyme activities

Enzyme activities in the growth substrate were affected byboth AM fungal colonization and external P levels, and my-corrhizal plants had significantly higher enzymatic activity forβ-Gl, CAT, phosphatase, and POD, but significantly lowerPPO activity (Table 4). Compared to non-mycorrhizal plants,enzymatic activity was significantly increased by AM fungalcolonization under P0, P3, and P30 by 31, 50, and 12 % for β-Gl; 10, 115, and 18 % for CAT; 11, 28, and 16 % forphosphatase; and 22, 25, and 19 % for POD, but was mark-edly decreased by 13, 23, and 26 % for PPO, respectively.Compared with the P3 treatment, the P0 treatment significantlyincreased CAT activity in the rhizosphere of non-mycorrhizalplants, significantly decreased β-Gl activity in themycorrhizosphere and phosphatase activity in both, but didnot affect β-Gl activity in the rhizosphere of non-mycorrhizalplants, CAT activity in the mycorrhizosphere, and POD andPPO activity in either (Table 4). On the other hand, comparedto the P3 treatment, the P30 treatment significantly increasedβ-Gl activity in the rhizosphere of non-mycorrhizal plants,significantly decreased CAT activity in the mycorrhizosphere,and decreased POD and phosphatase activity in both. AMfungal inoculation and P treatment had significant interactionson β-Gl, CAT, phosphatase, and PPO, but not on POD(Table 1).

Correlations among variables

Correlation analyses showed a significantly negative correla-tion of both AM colonization and hyphal length in the sub-strate with external P levels (Fig. 1), and a positive correlationof root AM colonization with hyphal length in the substrate(Table 5). EE-GRSP or DE-GRSP significantly correlatedpositively with T-GRSP, and EE-GRSP with DE-GRSP. Allthree GRSP fractions showed a significant positive correlationwith SOC (Table 4). GRSP fractions did not correlate withAM colonization, substrate hyphal length, or phosphataseactivity, but each GRSP fraction showed a significant positivecorrelation with β-Gl, CAT, and POD activity and a negativecorrelation with PPO activity (Table 4).

Table 1 Significance of the main treatment effects and their interactionsbased on two-way ANOVA on tested variables of trifoliate orange(Poncirus trifoliata) seedlings grown on sand substrate under three phos-phorus levels (0, 3, and 30 mM KH2PO4)

Variable Main effects Interaction effects(P level×AM fungus)

P level AM fungus

Mycorrhizal colonization ** ** **

Hyphal length ** ** **

Shoot biomass ** ** NS

Root biomass ** ** NS

Total biomass ** ** *

EE-GRSP ** ** **

DE-GRSP ** ** **

T-GRSP ** ** **

SOC ** ** NS

CAT ** ** **

POD ** ** NS

PPO NS ** **

β-Gl * ** **

Phosphatase ** ** **

NS not significant

*P<0.05; **P<0.01

124 Mycorrhiza (2015) 25:121–130

Author's personal copy

Tab

le3

Effectsof

Funneliformismosseae

onrhizosphericglom

alin-related

soilprotein(G

RSP

)andsoilorganiccarbon

concentrations

oftrifoliateorange

(Poncirustrifo

liata)seedlin

gsgrow

non

sand

substrateunderthreephosphorus

levels(0,3,and

30mM

KH2PO4)

Plevelinsand

substrate

Glomalin-related

soilprotein(m

g/gDW

substrate)

Soilorganiccarbon

(mg/gDW

substrate)

EE-G

RSP

DE-G

RSP

T-GRSP

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

P0

0.070±0.003a,y

0.081±0.003b,x

0.363±0.012a,y

0.401±0.019b,x

0.433±0.009a,y

0.482±0.020b,x

1.06

±0.15b,y

2.15

±0.28b,x

P3

0.060±0.005b,y

0.091±0.003a,x

0.360±0.061a,y

0.523±0.035a,x

0.420±0.057a,y

0.614±0.037a,x

2.11

±0.33a,y

3.00

±0.38a,x

P30

0.054±0.002b,y

0.070±0.003c,x

0.331±0.025a,y

0.386±0.025b,x

0.385±0.026a,y

0.456±0.023b,x

1.92

±0.31a,y

2.72

±0.30a,x

Data(m

eans±SE

,n=4)follo

wed

bydifferentlettersindicatesignificantdifferences(P<0.05)betweenPlevelsforthe

sameF.mosseae

treatm

ent(a,b)orbetweenF.mosseae

treatm

entsforthe

samePlevel

(x,y).P 0,P

3,and

P30correspond

to0,3,and30

mM

KH2PO

4in

sand

substrate,respectiv

ely

EE-G

RSP

easily-extractableGRSP,D

E-G

RSP

difficultly

-extractableGRSP,T-GRSP

total(EE-G

RSP

+DE-G

RSP)GRSP

Tab

le2

Effectsof

Funneliformismosseae

onmycorrhizal

colonizatio

n,substratehyphal

length,andbiom

assof

trifoliate

orange

(Poncirustrifo

liata)seedlin

gsgrow

non

sand

substrateunderthree

phosphorus

levels(0,3,and

30mM

KH2PO4)

Pin

sand

substrate(m

M)

Mycorrhizalcolonizatio

n(%

)Substratehyphallength

(m/g

FW)

Biomassproductio

n(g/plant

DW)

Shoot

Root

Total

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

P 00

54.7±2.7a

04.83

±0.13a

0.93

±0.03c,y

1.18

±0.03b,x

0.33

±0.01b,y

0.39

±0.02c,x

1.26

±0.03c,y

1.57

±0.02b,x

P 30

28.0±4.2b

00.39

±0.04b

1.20

±0.04a,y

1.36

±0.04a,x

0.39

±0.03a,y

0.43

±0.01b,x

1.59

±0.05a,y

1.79

±0.05a,x

P 30

013.5±2.6c

00.15

±0.05c

1.09

±0.06b,y

1.32

±0.04a,x

0.42

±0.01a,y

0.49

±0.01a,x

1.51

±0.06b,y

1.81

±0.03a,x

Data(m

eans±SE

,n=4)

follo

wed

bydifferentlettersindicatesignificantdifferences

(P<0.05)betweenPlevelsforthesameF.mosseae

treatm

ent(a,b,c)or

betweenF.mosseae

treatm

entsforthesameP

level(x,y).P

0,P

3,and

P 30correspond

to0,3,and30

mM

KH2PO

4in

sand

substrate,respectiv

ely

Mycorrhiza (2015) 25:121–130 125

Author's personal copy

Discussion

AM colonization is usually negatively correlated with soil Plevels (Lekberg and Koide 2005; Schreiner and Linderman2005; Gabriel-Neumann et al. 2011), and soil hyphal length ismore pronounced in soils under P limiting conditions (Leifheitet al. 2013). In the present study on trifoliate orange, externalP levels used in the substrate also negatively correlated withboth AM colonization and hyphal length (Fig. 1 and Table 5).These might be due to the fact that spore germination andhyphal growth of F. mosseae can be affected by P levels in thegrowth zone (Hepper 1983).

Results presented here showed shoot, root, and total bio-mass to be generally depressed under both P0 and P30 levels innon-mycorrhizal trifoliate orange seedlings as compared to theP3 treatment. In contrast, inoculation with F. mosseae signif-icantly increased shoot, root, and total biomass of seedlings,irrespective of the applied P level. Plant biomass increase bythe AM fungus was higher under P0 than under P30, indicatinga greater impact of mycorrhization at the lower P level.Substrate P levels can profoundly affect plant responses toAM, with high P supply diminishing beneficial effects of AMon plant biomass (e.g., Schroeder and Janos 2005). This mightbe due to a high P-induced decrease in AM fungal hyphalgrowth and spore production, thereby restricting beneficialAM effects (Schmidt et al. 2010). However, the effect of Psupply on AM-plant biomass interactions may depend on theplant and AM fungal species studied (Valentine et al. 2001).For example, Kwapata and Hall (1985) reported that underhigh P supply, mycorrhizal symbiosis did not affect the bio-mass of cowpea, while Peng et al. (1993) found that it de-pressed the growth of G. intraradices-colonized Volkamerlemon.

The EE-GRSP and DE-GRSP concentrations found in thesand substrate used to culture trifoliate orange were, respec-tively, less than 0.10 and 0.55 mg/g, and significantly lowerthan that reported for different soils (Bedini et al. 2009; Koideand Peoples 2013). These low GRSP concentrations mightresult from pot leaching through daily nutrient solution sup-ply, similar to the loss of GRSP observed in a soilless substratewatered four times per day (Nichols 2010). An explanationmight be that GRSP could tightly be bound to certain soils butnot to the surface of sand (Wright and Upadhyaya 1998),resulting in a lower accumulation in the sand substrate used.Although GRSP is insoluble, the movement of GRSP in asand substrate can be strongly affected by watering and by itswater holding capacity (Gispert et al. 2013). In this context,the mycorrhizal citrus seedlings might have produced moreGRSPs than detected but leaching with the nutrition solutionmay have resulted in a relatively high decrease of GRSPs inthe sand substrate.

A highly positive correlation was displayed between EE-GRSP or DE-GRSP and the T-GRSP in the rhizosphericT

able

4Effectsof

Funneliformismosseae

onactiv

ities

ofβ-glucosidase

(β-G

l),catalase

(CAT),peroxidase

(POD),phosphatase,

andpolyphenol

oxidase(PPO)in

rhizosphereof

trifoliate

orange

(Poncirustrifo

liata)seedlin

gsgrow

non

sand

substrateunderthreephosphorus

levels(0,3,and

30mM

KH2PO4)

Pin

sand

substrate(m

M)

β-G

l(mgglucose/g

DW

substrate)

CAT(0.1

MKMnO

4mL/g

DW

substrate)

POD(m

gpyrogallic

acid/g

DW

substrate)

Phosphatase(m

ghydroxybenzene/g

DW

substrate)

PPO(m

gpyrogallic

acid/g

DW

substrate)

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

−AMF

+AMF

P0

9.17

±0.26b,y

11.99±0.10b,x

0.39

±0.02a,y

0.43

±0.02a,x

9.53

±1.01a,y

11.60±0.37a,x

1.78

±0.07b,y

1.97

±0.09b,x

79.44±2.22ab,x

68.88±5.29a,y

P3

9.22

±0.65b,y

13.81±1.46a,x

0.20

±0.04b,y

0.43

±0.02a,x

9.71

±0.35a,y

12.10±0.76a,x

2.05

±0.10a,y

2.62

±0.17a,x

81.82±3.81a,x

63.19±5.85ab,y

P30

10.79±0.47a,y

12.10±0.79a,x

0.17

±0.01b,y

0.20

±0.02b,x

8.39

±0.34b,y

10.00±0.59b,x

1.59

±0.10c,y

1.85

±0.09b,x

77.01±1.62b,x

56.91±4.53b,y

Data(m

eans±SE,n

=4)follo

wed

bydifferentlettersindicatesignificantdifferences(P<0.05)betweenPlevelsforthe

sameF.mosseae

treatm

ent(a,b)orbetweenF.mosseae

treatm

entsforthe

samePlevel

(x,y).P 0,P

3,and

P30correspond

to0,3,and30

mM

KH2PO4in

sand

substrate,respectiv

ely

126 Mycorrhiza (2015) 25:121–130

Author's personal copy

substrate of trifoliate orange seedlings, which is in accordancewith the result of Koide and Peoples (2013). The concentra-tions of these GRSP fractions might also vary due to move-ment in and out with time (Steinberg and Rillig 2003), whichneeds to be further studied. In addition, both no-P or high Psupply significantly decreased EE-GRSP, DE-GRSP, and T-GRSP in the rhizosphere of AM seedlings, but generally not inthe rhizosphere of non-AM seedlings (expect for an increaseof EE-GRSP). These results indicate that mycorrhizal plantswere more sensitive to variation in P levels than non-mycorrhizal plants under sand culture.

In the present study, inoculationwithF.mosseae significantlyincreased both EE-GRSP and DE-GRSP, irrespective of sub-strate P levels, indicating that the release of GRSPs byAM fungimight not be dependent on P supply. Bedini et al. (2009)similarly reported significant increases of EE-GRSP or T-GRSP in Medicago sativa inoculated with F. mosseae andG. intraradices. Since GRSP is produced by AM fungi as acomponent of hyphae and spore walls (Driver et al. 2005),GRSP concentrations are often positively related to root coloni-zation and soil hyphae (Curaqueo et al. 2010; Guo et al. 2012;Wu et al. 2012). However, apart from a negative correlationbetween AM colonization and DE-GRSP, present results indi-cated that substrate hyphal length andAMcolonizationwere notsignificantly correlated with these three GRSP fractions. Suchresults are in agreement with previous reports (Rillig et al. 2002;Bedini et al. 2009). This may be because GRSP determination

by the Bradford assay also includes proteins of non-AM fungalorigin and other compounds (e.g., polyphenolic substances,lipids, and humic materials), besides proteins of AM fungalorigin (Purin and Rillig 2007; Gillespie et al. 2011), which arewashed away by watering so that a correlation between GRSP,hyphal length, and root colonization would be lost.

AM fungal inoculation of trifoliate orange seedlings sig-nificantly increased rhizospheric SOC concentrations, irre-spective of substrate P levels, which is in accordance withBedini et al. (2009). However, no significant correlation wasobserved between SOC and AM colonization or substratehyphal length. This might be due to the fact that organic matteris difficult to retain in sand or sandy soils (Qian et al. 2012).On the other hand, SOC significantly positively correlatedwith amounts of EE-GRSP, DE-GRSP, and T-GRSP, whichis similar to the result for Hapludalf soils (Koide and Peoples2013). These results suggest the potential contribution ofGRSPs to soil storage of C (Emran et al. 2012). GRSPs aredistinct components of SOC (Lovelock et al. 2004), and theamount of C in GRSP has for example been reported torepresent 4–5 % of total soil C in an old (4.1×106 years)tropical forest Oxisol soil (Rillig et al. 2001).

Soil enzymes are known to play substantial roles in main-taining soil health and its environment (Das and Varma 2011).A significantly higher activity ofβ-Gl, CAT, phosphatase, andPOD, but a lower PPO activity, was observed in themycorrhizosphere of AM citrus than in the rhizosphere of

y = 43.08 - 1.03 x, r2 = 0.66, P < 0.01

Substrate P level (mM)

0 5 10 15 20 25 30 35R

oot A

M c

olon

izat

ion

(%)

0

10

20

30

40

50

60

70

y = 2.86 - 0.098 x, r2 = 0.38, P < 0.05

Substrate P level (mM)

0 5 10 15 20 25 30 35

Subs

trat

e hy

phal

leng

th (m

/g s

oil F

W)

-1

0

1

2

3

4

5

6(a) (b)

Fig. 1 Relationships betweensubstrate P level and AMcolonization of trifoliate orangeroots (a) or substrate hyphallength (b) (n=12)

Table 5 Pearson correlation coefficients among the variables in the rhizosphere of trifoliate orange (Poncirus trifoliata) seedlings grown on sandsubstrate under three phosphorus levels (0, 3, and 30 mM KH2PO4) and with or without Funneliformis mosseae

Variable Hyphal length EE-GRSP DE-GRSP T-GRSP SOC CAT POD PPO β-Gl Phosphatase

Root colonization 0.93** −0.32 −0.58* −0.56 −0.25 0.09 −0.02 0.34 −0.24 0.00

Hyphal length 1.00 −0.03 −0.32 −0.29 0.03 0.33 0.25 0.07 0.08 −0.31EE-GRSP 1.00 0.77** 0.83** 0.44* 0.80** 0.83** −0.71** 0.68** 0.20

DE-GRSP 1.00 0.99** 0.56** 0.60** 0.72** −0.60** 0.68** −0.02T-GRSP 1.00 0.56** 0.66** 0.76** −0.64** 0.70** 0.02

n=12 between AM status and other variables, and n=24 between GRSPs and soil enzymes

*P<0.05; **P<0.01

Mycorrhiza (2015) 25:121–130 127

Author's personal copy

non-mycorrhizal seedlings, irrespectively of substrate P levels(Table 4). As mentioned above, no significant correlationswere found between root AM colonization or substrate hyphallength and the five tested soil enzymes in the present study. Ingeneral, these soil enzymes originate from microbes and plantresidues (Zhang et al. 2012). The presence of AM fungi couldinduce changes in the composition and quantity of root exu-dates (Xiao et al. 2009), or AM fungi might possess extracel-lular enzymes, thereby resulting in release of β-Gl, CAT,phosphatase, and POD which is not dependent on substrateP level. As a result, mycorrhizal colonization could conferbetter soil fertility and health in the rhizosphere of trifoliateorange.

In this study, EE-GRSP, DE-GRSP, or T-GRSP positivelycorrelated with β-Gl, CAT, and POD activity, but negatively(P<0.01) with PPO activity. Gispert et al. (2013) also reportedthat T-GRSP in 0.25–2.00 and 2.00–5.60 mm aggregatespositively correlated with β-Gl and protease in shallowLithic Xerorthent soils. A positive correlation was also foundbetween EE-GRSP or T-GRSP and β-Gl in citrus rhizospherein a Xanthi-udic Ferralsol (Wu et al. 2012). Soil enzymeactivities reflect the intensity and direction of soil biochemicalprocesses (Bai et al. 2009) which may mediate GRSP produc-tion, seeing the highly significant correlation of GRSPwithβ-Gl, CAT, POD, and PPO in trifoliate orange rhizosphere soil .However, the three GRSP fractions were not correlated withphosphatase activity, which is in disagreement with thefindings by Gispert et al. (2013) of a significant correlationbetween T-GRSP and phosphatase in 0.25–2.00 and 2.00–5.60 mm soil aggregates. Either substrate types or P levelsmight alter GRSP functions since GRSP levels can be influ-enced by soil fertility and minerals, such as available P or N(Rillig et al. 2001; Lovelock et al. 2004). In short, results fromthe present study on trifoliate orange growing in a sandsubstrate suggest that AM-induced production of GRSP andrelevant soil enzyme activities could be useful indicators ofsoil fertility to evaluate soil quality, although they varied withexternal P levels supplied to the substrate.

Acknowledgments This study was jointly supported by the NationalNatural Science Foundation of China (31372017), the Key Project ofNatural Science Foundation of Hubei Province (2012FFA001), the KeyProject of the Chinese Ministry of Education (211107), and the ExcellentYoung Teacher Research Support Program of Yangtze University(cyq201324). We sincerely thank anonymous reviewers for their valuablecomments on the manuscript.

References

Adamczyk B, Kitunen V, Smolander A (2009) Polyphenol oxidase,tannase and proteolytic activity in relation to tannin concentrationin the soil organic horizon under silver birch and Norway spruce.Soil Biol Biochem 41:2085–2093

Alguacil MM, Caravaca F, Azcón R, Roldan A (2009) Changes inbiological activity of a degraded Mediterranean soil after usingmicrobially-treated dry olive cake as a biosolid amendment andarbuscular mycorrhizal fungi. Eur J Soil Biol 44:347–354

Badiane NNY, Chotte JL, Pate E, Dasse D, Rouland C (2001) Use of soilenzyme activities to monitor soil quality in natural and improvedfallows in semi-arid tropical regions. Appl Soil Ecol 18:229–238

Bai CM, He XL, Tang HL, Shan BQ, Zhao LL (2009) Spatial distributionof arbuscular mycorrhizal fungi, glomalin and soil enzymes underthe canopy of Astragalus adsurgens Pall. in the Mu Us sandland,China. Soil Biol Biochem 41:941–947

Bedini S, Pellegrino E, Avio L, Pellegrini S, Bazzoffi P, Argese E,Giovannetti M (2009) Changes in soil aggregation and glomalin-related soil protein content as affected by the arbuscular mycorrhizalfungal species Glomus mosseae and Glomus intraradices. Soil BiolBiochem 41:1491–1496

Beltrano J, RusittiM, AgrangoMC, RoncoM (2013) Effects of arbuscularmycorrhiza inoculation on plant growth, biological and physiologi-cal parameters and mineral nutrition in pepper grown under differentsalinity and p levels. J Soil Sci Plant Nutr 13:123–141

Bethlenfalvay GJ, Ames RN (1987) Comparison of two methods forquantifying extraradical mycelium of vesicular-arbuscular mycor-rhizal fungi. Soil Sci Soc Am J 51:834–837

BradfordMM (1976) A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dyebinding. Anal Biochem 72:248–252

Brooks DD, Twieg BD, Grayston SJ, Jones MD (2013) Physical extent,frequency, and intensity of phosphatase activity varies on soil pro-files across a Douglas-fir chronosequence. Soil Biol Biochem 64:1–8

Curaqueo G, Acevedo E, Cornejo P, Seguel A, Rubio R, Borie F (2010)Tillage effect on soil organic matter, mycorrhizal hyphae and aggre-gates in a mediterranean agroecosystem. R C Suelo Nutr Veg 10:12–21

Das SK, Varma A (2011) Role of enzymes in maintaining soil health. In:Shukla G, Varma A (eds) Soil enzymology. Springer, Berlin, pp 25–42

Driver JD, HolbenWE, RilligMC (2005) Characterization of glomalin asa hyphal wall component of arbuscular mycorrhizal fungi. Soil BiolBiochem 37:101–105

Emran M, Gispert M, Pardini G (2012) Patterns of soil organic carbon,glomalin and structural stability in abandoned Mediterranean ter-raced lands. Eur J Soil Sci 63:637–649

FokomR, Adamou S, TeugwaMC, Begoude Boyogueno AD, NanaWL,NgonkeuMEL, Tchameni NS, Nwaga D, Tsala NdzomoG, AmvamZollo PH (2012) Glomalin related soil protein, carbon, nitrogen andsoil aggregate stability as affected by land use variation in the humidforest zone of south Cameroon. Soil Tillage Res 120:69–75

Fredeen AL, Terry N (1988) Influence of vesicular-arbuscular mycorrhizalinfection and soil phosphorus level on growth and carbon metabolismof soybean. Can J Bot 66:2311–2316

Gabriel-Neumann E, Neumann G, Leggewie G, George E (2011)Constitutive overexpression of the sucrose transporter SoSuT1 inpotato plants increased arbuscular mycorrhiza fungal root coloniza-tion under high, but not under low, soil phosphorus availability. JPlant Physiol 168:911–919

Garcia C, Hernandez T, Roldan A, Albaladejo J, Castillo V (2000)Organic amendment and mycorrhizal inoculation as a practice inafforestation of soils with Pinus halepensis Miller: effect on theirmicrobial activity. Soil Biol Biochem 32:1173–1181

Gillespie AW, Farrell RE, Walley FL, Ross ARS, Leinweber P, EckhardtKU, Regier TZ, Blyth RIR (2011) Gloamlin-related soil proteincontains non-mycorrhizal-related heat-stable proteins, lipids andhumic materials. Soil Biol Biochem 43:766–777

Gispert M, EmranM, Pardini G, Doni S, Ceccanti B (2013) The impact ofland management and abandonment on soil enzymatic activity,glomalin content and aggregate stability. Geoderma 202–203:51–61

128 Mycorrhiza (2015) 25:121–130

Author's personal copy

Guan SY (1986) Soil enzymes and its research methods (in Chinese).Agricultural Press, Beijing

Guo YP, Chen PZ, Zhang LC, Zhang SL (2003) Phosphorus deficiencystress aggravate photoinhibition of photosynthesis and function ofxanthophyll cycle in citrus leaves. Plant Nutr Fertil Sci 9:359–363(in Chinese with English abstract)

Guo HJ, He XL, Li YP (2012) Spatial distributuion of arbuscular mycor-rhiza and glomalin in the rhizosphere ofCaragana korshinskiiKom.in the Otindag sandy land, China. Afr J Microbiol Res 6:5745–5753

Hammer EC, Rillig MC (2011) The influence of different stresses onglomalin levels in an arbuscular mycorrhizal fungus–salinity in-creases glomalin content. PLoS ONE 6(12):e28426. doi: 10.1371/journal.pone.0028426

Hepper CM (1983) Effect of phosphate on germination and growth ofvesicular-arbuscular mycorrhizal fungi. Trans Br Mycol Soc 80:487–490

Huang HL, Zhang SZ, Wu NY, Luo L, Christie P (2009) Influence ofGlomus etunicatum/Zea mays mycorrhiza on atrazine degradation,soil phosphatase and dehydrogenase activities, and soil microbialcommunity structure. Soil Biol Biochem 41:726–734

Jifon JL, Graham JH, Drouillard DL, Syvertsen JP (2002) Growth de-pression of mycorrhizal Citrus seedlings grown at high phosphorussupply is mitigated by elevated CO2. New Phytol 153:133–142

Koide RT, Peoples MS (2013) Behavior of Bradford-reactive substancesis consistent with predictions for glomalin. Appl Soil Ecol 63:8–14

KwapataMB, Hall AE (1985) Effects of moisture regime and phosphoruson mycorrhizal infection, nutrient uptake, and growth of cowpeas(Vigna unguiculata (L.) Walp.). Field Crop Res 12:241–250

Leifheit EF, Veresoglou SD, Lehmann A, Kathryn Morris E, Rillig MC(2013) Multiple factors influence the role of arbuscular my-corrhizal fungi in soil aggregation—a meta-analysis. PlantSoil 374:523–537

Lekberg Y, Koide RT (2005) Arbuscular mycorrhizal fungi, rhizobia,available soil P and nodulation of groundnut (Arachis hypogaea) inZimbabwe. Agric Ecosyst Environ 110:143–148

Lovelock CE, Wright SF, Clarck DA, Ruess RW (2004) Soil stocks ofglomalin produced by arbuscular mycorrhizal fungi across a tropicalrain forest landscape. J Ecol 92:278–287

Nichols KA (2010) Glomalin production and accumulation in soilless potcultures. Can J Soil Sci 90:567–570

Peng SB, Eissenstat DM, Graham JH, Williams K, Hodge NC (1993)Growth depression in mycorrhizal citrus at high-phosphorus supply:Analysis of carbon costs. Plant Physiol 101:1063–1071

Phillips JM, Hayman DS (1970) Improved procedures for clearing rootsand staining parasitic and vesicular-arbuscular mycorrhizal fungi forrapid assessment of infection. Trans Br Mycol Soc 55:158–161

Purin S, Rillig MC (2007) The arbuscular mycorrhizal fungal proteinglomalin: limitations, progress, and a new hypothesis for its func-tion. Pedobiologia 51:123–130

Qian K, Wang L, Yin N (2012) Effects of AMF on soil enzyme activityand carbon sequestration capacity in reclaimed mine soil. Int J MinSci Technol 22:553–557

Raiesi F, GhollarataM (2006) Interactions between phosphorus availabil-ity and anAM fungus (Glomus intraradices) and their effects on soilmicrobial respiration, biomass and enzyme activities in a calcareoussoil. Pedobiologia 50:413–425

Rillig MC (2004) Arbuscular mycorrhizae, glomalin and soil quality. CanJ Soil Sci 84:355–363

Rillig MC, Wright SF, Nichols KA, Schmidt WF, Torn MS (2001) Largecontribution of arbuscular mycorrhizal fungi to soil carbon pools intropical forest soils. Plant Soil 233:167–177

Rillig MC, Wright SF, Eviner VT (2002) The role of arbuscular mycor-rhizal fungi and glomalin in soil aggregation: comparing effects offive plant species. Plant Soil 238:325–333

Rowell DL (1994) Soil science: methods and applications. LongmanGroup U.K. Ltd, London

Schindler FV, Mercer ER, Rice JA (2007) Chemical characteristics ofglomalin-related soil protein (GRSP) extracted from soils of varyingorganic matter content. Soil Biol Biochem 39:320–329

Schmidt B, Mononkos M, Sumalan R, Biro B (2010) Suppression ofarbuscular mycorrhiza development by high concentrations of phos-phorus at Tagetes patula L. Res J Agric Sci 42:156–162

Schreiner RP, Linderman RG (2005)Mycorrhizal colonization in drylandvineyards of the Willamette Valley, Oregon. Small Fruits Rev 4:41–55

Schroeder MS, Janos DP (2005) Plant growth, phosphorus nutrition, androot morphological responses to arbuscular mycorrhizas, phos-phorus fertilization, and intraspecific density. Mycorrhiza 15:203–216

Singh PK, Singh M, Tripathi BN (2013) Glomalin: an arbuscular mycor-rhizal fungal soil protein. Protoplasma 250:663–669

Sinsabaugh RL (2010) Phenol oxidase, peroxidase and organic matterdynamics of soil. Soil Biol Biochem 42:391–404

Stege PW, Messina GA, Bianchi G, Olsina RA, Raba J (2010)Determination of β-glucosidase activity in soils with a bioanalyticalsensor modified with multiwalled carbon nanotubes. Anal BioanalChem 397:1347–1353

Steinberg PD, Rillig MC (2003) Differential decomposition of arbuscularmycorrhizal fungal hyphae and glomalin. Soil Biol Biochem 35:191–194

Tong Y, Gabriel-Neumann E, Ngwene B, Krumbein A, Baldermann S,Schreiner M, George E (2013) Effects of single and mixed inocula-tion with two arbuscular mycorrhizal fungi in two different levels ofphosphorus supply on β-carotene concentrations in sweet potato(Ipomoea batatas L.) tubers. Plant Soil 372:361–374

Treseder KK, Turner KM (2007) Glomalin in ecosystems. Soil Sci SocAm J 71:1257–1266

Valarini PJ, Curaqueo C, Seguel A, Manzano K, Rubio R, Cornejo P,Borie F (2009) Effect of compost application on some properties of avolcanic soil from Central South Chile. Chil J Agric Res 69:416–425

Valentine AJ, Osborne BA, Mitchell DT (2001) Interactions betweenphosphorus supply and total nutrient availability on mycorrhizalcolonization, growth and photosynthesis of cucumber. Sci Hortic88:177–189

Vasconcellos FLF, Bonfim JA, Baretta D, Cardoso EJBN (2013)Arbuscular mycorrhizal fungi and glomalin-related soil protein aspotential indicators of soil quality in a recuperation gradient of theAtlantic forest in Brazil. Land Degrad Dev. doi: 10.1002/ldr.2228

Violi HA, TresederKK MJA, Wright SF, Lovatt CJ (2007) Densitydependence and interspecific interactions between arbuscular my-corrhizal fungi mediated plant growth, glomalin production, andsporulation. Can J Bot 85:63–75

Wang FY, Lin XG, Yin R, Wu LH (2006) Effects of arbuscular mycor-rhizal inoculation on the growth of Elsholtzia splendens and Zeamays and the activities of phosphatase and urease in a multi-metal-contaminated soil under unsterilized conditions. Appl Soil Ecol 31:110–119

Wang P, Liu JH, Xia RX, Wu QS, Wang MY, Dong T (2011) Arbuscularmycorrhizal development, glomalin-related soil protein (GRSP)content, and rhizospheric phosphatase activity in citrus orchardsunder different types of soil management. J Plant Nutr Soil Sci174:65–72

Wright SF, Anderson RL (2000) Aggregate stability and glomalin inalternative crop rotations for the central Great Plains. Biol FertilSoils 31:249–253

Wright SF, Upadhyaya A (1996) Extraction of an abundant and unusualprotein from soil and comparison with hyphal protein of arbuscularmycorrhizal fungi. Soil Sci 161:575–586

Wright SF, Upadhyaya A (1998) A survey of soils for aggregate stabilityand glomalin, a glycoprotein produced by hyphae of arbuscularmycorrhizal fungi. Plant Soil 198:97–107

Mycorrhiza (2015) 25:121–130 129

Author's personal copy

Wu QS, Xia RX, Zou YN (2008) Improved soil structure and citrusgrowth after inoculation with three arbuscular mycorrhizal fungiunder drought stress. Eur J Soil Biol 44:122–128

Wu QS, He XH, Zou YN, He KP, Sun YH, Cao MQ (2012) Spatialdistribution of glomalin-related soil protein and its relationships withroot mycorrhization, soil aggregates, carbohydrates, activity of pro-tease and β-glucosidase in the rhizosphere of Citrus unshiu. SoilBiol Biochem 45:181–183

Wu QS, He XH, Cao MQ, Zou YN, Wang S, Li Y (2013) Relationshipsbetween glomalin-related soil protein in water-stable aggregate frac-tions and aggregate stability in citrus rhizosphere. Int J Agric Biol15:603–606

Xiao M, Gao YZ, Ling WT, Cheng ZX, Zeng YC (2009) Effectsof arbuscular mycorrhizal fungi on enzymes activity in soilscontaminated by phenonthrene and pyrene. Chin Environ Sci39:668–672

Yan CR (1988) Research methods in soil fertility (in Chinese).Agricultural Press, Beijing

Zhang W, Zhang GS, Liu GX, Dong ZB, Chen T, Zhang MX, Dyson P,An LZ (2012) Bacterial diversity and distribution in the southeastedge of the Tengger Desert and their correlation with soil enzymeactivities. J Environ Sci 24:2004–2011

Zhao LP, Jiang Y (1986) Determination of soil phosphatase activity. ChinJ Soil Sci 17:138–141 (in Chinese)

130 Mycorrhiza (2015) 25:121–130

Author's personal copy