at SciVerse ScienceDirect

Soil Biology & Biochemistry 56 (2013) 13e20

Contents lists available

Soil Biology & Biochemistry

journal homepage: www.elsevier .com/locate/soi lb io

Diversity of adsorption affinity and catalytic activity of fungal phosphatasesadsorbed on some tropical soils

Brice Kedi a,b, Josiane Abadie a, Joseph Sei b, Hervé Quiquampoix a, Siobhán Staunton a,*

a INRA, UMR Eco&Sols, INRA/IRD/Cirad/Montpellier SupAgro, 2 place Viala, W34060 Montpellier, Franceb LCMI, UFR SSMT, Université de Cocody, 22 BP 582 Abidjan 22, Côte d’Ivoire, France

a r t i c l e i n f o

Article history:Received 13 September 2011Received in revised form2 February 2012Accepted 5 February 2012Available online 18 February 2012

Keywords:AdsorptionEnzymatic activityAcid phosphatasesEctomycorrhizal fungiFerralsolOxisolVertisolSuillus collinitusHebeloma cylindrosporum

* Corresponding author. Tel.: þ33 (0) 499 61 23 31E-mail address: staunton@montpellier.inra.fr (S. St

0038-0717/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.soilbio.2012.02.006

a b s t r a c t

Extracellular phosphatases from ectomycorrhizal fungi may contribute to plant nutrition, particularly inhighly weathered tropical soils where available phosphorus is limited. However, the expression of theircatalytic activity may be influenced by many factors including adsorption on organo-mineral surfaces ofsoils. We have investigated the pH-dependent adsorption and the resulting change in catalytic activity oftrace amounts of extracellular fungal acid phosphatases in contact with various tropical soils. Different sizefractions and the effect of chemical extraction of soil organo-mineral coatings were studied for acidphosphatases secreted by two Hebeloma cylindrosporum and one Suillus collinitus strains. The threeenzymes differed in their average affinity for soil and the pH dependence. However the difference betweenenzymes was often greater than that between soils for a given enzyme. Only Suillus phosphatase hadmarkedly different affinities for vertisol and ferralsol. In general, activity was largely conserved in theadsorbed state. Activity of Suillus phosphatases varied more than Hebeloma phosphatases between soilsand soil physical and chemical fractionation. There was no relationship between the affinity of an enzymefor a surface and the resulting change in activity after adsorption. Suillus phosphatases showed more pHdependence of both affinity for soil and loss of activity after adsorption than the Hebeloma enzymes. Theseresults highlight the variability of the interaction of fungal phosphatases with soil.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Extracellular enzymes are known to play important roles in thebiogeochemistry in soils of many essential elements, includingphosphorus. Organic P may constitute a large proportion of soil P inmany soils (30e65%, but up to >90%) and may thus play animportant role in P nutrition, particularly in highly weatheredtropical soils (Harrison, 1987; Vance et al., 2003; Turner andEngelbrecht, 2011). However it must be hydrolysed to inorganic Pbefore it can be available to biological systems. The availability ofsoil organic P depends therefore not only on the size of stocks butalso on the catalytic activity of phosphatases in soil. Phosphatasesmay be produced and released into soil by various micro-organisms, including ectomycorrhizal fungi (Quiquampoix andMousain, 2005; Cairney, 2011). Catalytic activity of extracellularphosphatases is one of the mechanisms whereby ectomycorrhizalfungi contribute to the nutrition of higher plants (Cairney, 2011).

Extracellular enzymes in soil are likely to be adsorbed on organo-mineral surfaces. Adsorption may favour conformational changes

; fax: 33 (0) 499 61 30 88.aunton).

All rights reserved.

that modify catalytic activity or may favour orientation of the activesite towards the surface thereby restricting access to substrate ortowards the solution thus enhancing substrate accessability(Leprince and Quiquampoix, 1996). Various mechanisms have beenproposed to explain the extent of adsorption of enzymes on organo-mineral surfaces, including electrostatic forces and hydrophobicinteractions, van der Waals forces, covalent bonds and intercalationof enzymes in interlayer spaces of swelling clays (George et al.,2007a; Huang et al., 2005; Quiquampoix, 2000; Quiquampoixet al., 2002; Violante et al., 1995). Some authors state that theextent of the interaction between enzymes and mineral surfacesdepends strongly on the nature of the surface. For example, it hasbeen reported that in general more catalytic activity is lost in contactwith 2:1 type minerals than 1:1 minerals (Nannipieri et al., 1996)and that less activity is lost when organic matter coats surfaces thanwhen enzymes are adsorbed on clays or aluminium oxides (Raoet al., 2000). A greater loss of activity of fungal phytases has beenobserved in contact with phyllosilicates than with iron oxides(Giaveno et al., 2010). Both adsorption and changes in activitydepend strongly upon pH. A shift in optimal pH towards alkaline pHis often observed in the presence of mineral surfaces (Leprince andQuiquampoix, 1996; McLaren and Estermann, 1957; Quiquampoix,

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e2014

1987a). A strong pH dependence of native phosphatase activity isobserved in soils, but the optimal pH is only loosely related to soil pH(Turner, 2010).

Many of the studies of the effect of mineral and organo-mineralsurfaces on catalytic activity of enzymes have been carried outusing pure minerals or synthetic complexes (Rao and Gianfreda,2000; Rao et al., 1996; Rosas et al., 2008) chosen to simulatenatural soil. Such model systems may be very different to thecomplex heterogeneous complexes in soils containing clay minerals,inorganic and organic particles and coatings (Rowell, 1994). In manycases the enzyme studied would not be found naturally in soil.Extracellular enzymes might be expected to perform better thanintracellular enzymes in contact with organo-mineral surfaces. Wehave studied tropical soils since the role of extracellular phospha-tases may be more crucial to phosphorus nutrition in these soils,than in less weathered or better fertilised temperate soils. We haveused ectomycorrhizal fungi as sources of phosphatase despite theongoing debate as to the importance of ectomycorrhizal fungi intropical soils. While endomycorrhizal fungi are generally assumed todominate in tropical forests, ectomycorrhizae are found in manytropical soils (Mayor and Henkel, 2006, and references within).

The objective of this study was to investigate the pH-dependentadsorption and the resulting change in catalytic activity of extra-cellular fungal acid phosphatases added to various fractions ofcontrasting natural tropical soils. In particular we have studied theeffect of dominant mineralogy, particle size fractionation and thechemical removal of organic and mineral coatings.

2. Materials and methods

2.1. Preparation of acid phosphatases

Three strains of ectomycorrhizal fungi were used to produceextracellular phosphatases: one strain of Suillus collinitus, (laboratorycode J3.15.35) referred to as Sc and two strains of Hebeloma cylin-drosporum (laboratory codes D3.25.2 and D2) referred to as Hc1 andHc2. All strains were selected from the collection maintained by theEco&Sols (Ecologie Fonctionnelle et Biogéochimie des Sols et desAgroécosystèmes, INRA-IRD-Cirad-SupAgro, Montpellier, France)unit and have been used in previous studies. In particular, they werefound to secrete contrasting amounts of phosphatases and tocontribute differently to phosphorus nutrition of pine seedlings(Matumoto-Pintro, 1996). They were grown on sterile agar gel (16 g/L) in Petri dishes at 25 �C in the dark for 25 days. Fungal plugs weretaken and grown under sterile conditions in 25mL liquid culture at25 �C in the dark for 30 days. The sterilised (autoclaved at 120 �C for20min) nutrient solution contained NaCl (0.1 mM), KNO3 (4 mM),KCl (1 mM), NH4Cl (2 mM), MgSO4 (1 mM), CaCl2 (1 mM),thiamineeHCl (0.3 mM), ferric citrate (10 g/L), glucose (10 g/L),

Table 1Origin and some properties of the selected soils.

Soil Samplingdepth (cm)

0e2 mm(g kg�1)

2e50 mm(g kg�1)

50e200 mm(g kg�1)

Corg (after H2O2)(g kg�1)

Fe-C(g kg

V 0e10 563 184 112 31.2 (23.5) 15.7

F1 0e10 411 403 117 45.7 (14.3) 45.0

F2 0e10 768 153 36 28.2 (11.3) 65.3

F3 0e10 495 339 140 29.5 (16.2) 71.8

F4 0e10 800 190 10 34.6 (31.7) 81.2

Soil Survey Staff (a1999); FAOeUNESCOeIRIC (b1998). Ferralsol or oxisol using the USDA

KH2PO4 (3.4 mM), and trace elements as recommended by Morizetand Mingeau (1976). At the end of the culture period the thalliwere removed, the solutions filtered using a Vileda-Mopmembrane,pooled, divided into smaller containers and frozen until required.Protein assays using the Bradford (1976) method indicated that thetotal protein concentrations were <15 mg/L (below the detectionlimit).

2.2. Soil samples

Five clay-textured tropical soils were chosen from the collec-tion maintained by the Eco&Sols unit, all of which have beenstudied previously and had mineralogies typical of many tropicalsoils. All had been sampled from a depth of 0e10 cm. Some of theirproperties and references to publications are given in Table 1. Thesoils were gently crushed by hand using a ceramic mortar andpestle and sieved <200 mm. The clay-sized fraction, <2 mm wasobtained by sedimentation, and sand-sized fraction, 50e200 mmby wet sieving after sonification. The clay-sized fraction wasstored refrigerated as a suspension 10 g/L until required. The sand-sized fraction was dried before storage. Subsamples of the sievedsoils were also chemically treated to remove mineral oxides(citrate-bicarbonate-dithionite, CBD, (Mehra and Jackson, 1960))or organic matter (30% H2O2, (Lavkulich and Wiens, 1970)).Chemically treated samples were washed in CaCl2 (1 M) to removechemical residues and then rinsed in water to remove excess salt.The amount of Fe removed by the CBD treatment was measured byanalysing the metal content of the extraction solution by AtomicAbsorption Spectroscopy. The carbon content of the soils beforeand after H2O2 treatment was measured by dry combustion usingFisons-EA-1108 CHNS-O Element Analyzer�. The specific surfacearea of samples was measured using the BET method (Micro-meritics-ASAP 2020�).

2.3. Measurement of enzymatic activity

Three procedures were used to distinguish activity in solution,extent of adsorption and catalytic activity in the adsorbed state, asdescribed by Quiquampoix (1987a,b):

� Procedure A e enzymatic activity in solution, in the absence ofsoil material

� ProcedureBe enzymatic activity in suspension containing 1 g/Lsoil material

� Procedure Ce enzymatic activity in the supernatant solution ofthe above suspension.

The contact period between soil and enzyme in suspension priorto addition of substrate was 1 h. Soil and solution were separated

BD�1)

pH(H2O)

Classification Localisation Reference

6.25 Smectitic LepticHapluderta

(EutricVertisol)b

Martinique,FrenchWest Indies

(Chevallier et al., 2000)

5.93 Andic Dystrustepta

FerralsolcMadagascar (Razafimbelo et al., 2006a)

4.82 Typic Haplorthoxa

(Orthic Ferralsol)b,cCongo (Barthès et al., 2008;

Kouakoua et al., 1997)5.33 Typic Hapludoxa

(Orthic Ferralsol)b,cBrazil (Razafimbelo et al., 2006b)

7.1 Haplorthoxa

(Rhodic Ferralsol)b,cBrazil (Kouakoua et al., 1999)

classification, Krasilnikov et al., 2009.

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e20 15

(prior to additionof substrate for ProcedureC and after the enzymaticreaction for Procedure B) by centrifugation at 23 000g for 15 min.

Enzymatic activity was measured in solutions or suspensionsbuffered to the required pH using sodium acetate buffer (300 mM).The substrate was p-nitrophenyl phosphate (pNPP), with an initialconcentration of 6.7 mM, which was greatly in excess of productformed. After 20min at 25 �C the reactionwas stopped by addition ofglycine buffer (glycine (133 mM); NaOH (125mM); NaCl (67 mM);Na2CO3 (83mM) at pH 10.5). The concentration of the product wasmeasured at 405 nm in a microplate spectrophotometer (DIALAB-ELx808). Allmeasurementswere carriedout in triplicate. Appropriateblanks and controls were measured. Suspensions were centrifugedimmediately after addition of stop buffer to avoid colouration due tothe release of humic substances from soil in alkaline solution.

2.4. Data interpretation and treatment

The reaction velocities were calculated for each procedure, VA,VB and VC at each pH and expressed relative to the maximum valuein solution at the optimum pH, for ease of comparison betweenenzymes. A comparison of VA and VC indicates the affinity of theenzyme for the organo-mineral surface. When VC tends towards 0,adsorption was complete. When VA, and VC were significantlydifferent, namely adsorption occurred, the comparison of VA, and VBindicates the extent to which activity was retained in the adsorbedstate. The non parametric statistical KolmogoroveSmirnov (KS)tests were applied to compare series of V with respect to pH. Fora given soil the data of VA, VB and VC were compared and either VBor VC for different soils or soil treatments were compared using theKS test. When series of data was found to be significantly differentthe following parameters were also calculated. The fraction ofenzyme free in solution after contact with soil, F was calculated as

F ¼ VC=VA (1)

The activity in the adsorbed state, relative to that of the sameamount of enzyme in solution at the same pH, R, was calculated(Leprince and Quiquampoix, 1996; Quiquampoix, 1987a) as

R ¼ ðVB � VCÞ=ðVA � VCÞ (2)

When VA and VC are close (little adsorption) it is numericallyimpossible to calculate R with certainty.

3. Results

The biomass production of thallus was 0.3e0.5 g per thallus andthe maximum catalytic activity was in the range 10e20 nKatal/g

Fig. 1. Effect of soil clays on the pH dependence of catalytic activity (V) of extracellular acid pcatalytic activity in solution VA (C), in suspension of soil clay VB (-) and in the supernatantriplicates, but are not always visible, given the symbol size.

fresh mass thallus depending on the fungal strain. The optimal pHvalues were in the range 4.5e5.5 as expected for acid phosphatases.The specific surface area of the size fraction and sieved soils beforeand after physical fractionation and chemical treatments to removeorganic and mineral coatings on mineral surfaces are given inTable SI-1, Supporting Information.

3.1. Adsorption and activity of phosphatases in the presence of soilclay as a function of pH

3.1.1. Reaction rate, V, for procedures A, B and CFig. 1 shows the catalytic activity of each of the three phos-

phatases as a function of pH in solution (VA), in suspension of theclay-sized fractions of the vertisol or one of the ferralsols, F1 (VB) orin the supernatant solution of those suspensions, after separationof the soil and solution (VC). Each enzyme followed very similartrends in contact with either soil. Large contrasts were observedbetween enzymes. The results of the KS tests comparing pairs ofcurves for each soil clay and each enzyme are given in Table SI-2,Supporting Information. Curves VA, and VC were significantlydifferent and VA and VB were not, for both Sc and Hc1 in thepresence of both soils, indicating adsorption with complete pres-ervation of activity in the adsorbed state. The behaviour of theenzyme obtained from the other Hebeloma strain, Hc2 was mark-edly different. For both soil clays, the curves VA, VB and VC were notsignificantly different, thus Fw 1 and R thus could not be calculatedwith any certainty.

Three other ferralsols, differing in their texture, organic C contentand extractable Fe content were compared. Data and statisticalanalyses are given as Supporting Information (Figure SI-1 andTable SI-3). For eachenzyme the trends of affinity and relative activityas functions of pH were similar for each of the ferralsols. The mostnotable exception was observed for Hc2 for which there wasdetectable adsorption on F3 and F4, unlike F1, with loss of activity inthe adsorbed state.

3.1.2. Fraction not adsorbed and relative activity retained in theadsorbed state, F and R

Fig. 2 shows the values of the parameters F and R calculated fromthe data in Fig. 1 for both soils and for the enzymes Sc and Hc1. Thecontrasting behaviour of Suillus phosphatase in the presence of thetwo soils is more apparent than in Fig. 1. In the presence of thevertisol the free fraction of Sc remained constant at about 25% overthe pH range, whereas this fraction increased three-fold from about25% to 75% in the presence of the ferralsol. The pH dependence ofHc1 in contact with either soil was less marked. There was a smalldecrease in F in contact with the vertisol with increasing pH, but no

hosphatases from three ectomycorrhizal fungi, Suillus (Sc) and Hebeloma (Hc1 and Hc2):t solution of soil clay suspension VC (:). Bars show experimental variability between

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e2016

change in contact with the ferralsol, and values of R were constantat about 80% on both soils.

Each of the four ferralsols showed the same pH dependence,namely an increase in the free fraction, hence decrease in affinity,with increasing pH for Suillus phosphatases and no pH dependencefor the Hebeloma phosphatases (Figure SI-1). There was littledifference in affinity between soils for the Sc and Hc1 enzymesdespite differences in texture, organic carbon and iron oxidecontents. However for Hc2 the average value of F was 90% for F1,about 80% for F2 and F3 and about 50% for F4, but these values didnot follow the same trend as any of the measured soil properties.The trends of R showed some differences between the ferralsols.While R for Sc in contact with F1 was fairly constant at about 100%throughout the pH range, the values of R fell sharply at pH above 4.5or 5 with increasing pH for the other ferralsols. It should be notedthat the small proportion of adsorbed phosphatases means thattheir contribution to activity in suspensionwas small. For Hc1 therewas no variation of R with pH for F4, as observed for F1, whereas Rincreased with increasing pH for F2 and F3 from about 50% to 100%.Therewas no coherent trend of variation of R as a function of pH forHc2 on any of the ferralsols. The average values of R for Hc2 wereabout 25% for F4, much smaller than for the other enzymeesoilcombinations.

3.2. Affinity of phosphatases for soils and soil fractions

Fig. 3 shows the effect of soil particle size fraction on the freefraction of phosphatases of Sc and Hc1 in contact with the vertisolor the ferralsol F1 as a function of pH. The pH trends were alwaysthe same for both particle size fractions with any soileenzymecombination. For each soileenzyme combination, the free fractionwas always larger and hence affinity smaller for the sand fractionthan the clay fraction at every pH value, although for Hc1 in contactwith the vertisol the difference was not significant (Table SI-4,Supporting Information).

Fig. 4 shows the effect of chemical removal of organic or mineralcoatings from sieved soils (V and F1) on the F values of phospha-tases Sc and Hc1 as a function of pH. Removal of mineral oxidesfrom the vertisol had no effect on the affinity of Sc at the most acidpH, but led to a marked increase in affinity (decrease in free frac-tion) as pHwas increased above 5. For the ferralsol there was a shiftto lower values of F (increased affinity) throughout the range of pHstudied. Removal of organic matter led to a fairly constant shift in F(�20%) for the vertisol, but led to complete adsorption on the fer-ralsol for pH 3e5 and at larger pH adsorption decreased markedly,

Fig. 2. Non-adsorbed enzyme fraction of phosphatases in contact with soil (F) and relative activClosed symbols represent clay fractions of vertisol and open symbols that of the ferralsol, F1. B

but remained greater than for the untreated soil. The effect ofchemical treatments on the affinity of Hc1 was very different fromthat observed for Sc. Neither treatment had any significant effectfor the vertisol whereas both treatments led to a decrease in F ofabout 20% at all pH. The statistical tests (Table SI-4, SupportingInformation) confirm that the chemical treatments had no effect onthe adsorption of Hc1 on the vertisol and that in all other casessignificant differences were observed between treated anduntreated soils for both enzymes.

3.3. Relative catalytic activity of adsorbed phosphatases

Fig. 5 shows the trends in relative activity in the adsorbed phaseof Sc and Hc1 in contact with clay and sand-sized fractions of thevertisol and the ferralsol, F1 (with statistical analysis of the curvesof VB and VC in Table SI-4). There was no significant difference in thepH profile of activity in solution and in suspension for Hc1 on eithersize fraction of either soil. The relative activity of Hc1 afteradsorption, R, for both clay-sized and sand-sized fractions of boththe vertisol and the ferralsol, F1 was 100%. The same was true of Scin contact with both size fractions of the vertisol. However for Scthe values of R on the two size fractions of the vertisol decreasedwith increasing pH, from about 100% at pH 3 to 30% at pH 6 for theclay-sized fraction and with values of R always about 20% smalleron the sand-sized fraction.

The values of relative activity after adsorption as a function ofpH for Sc and Hc1 in contact with soil before or after chemicaltreatments are shown in Fig. 6 (with statistical analysis inTable SI-4, Supporting Information). Neither of the chemicaltreatments had a significant effect on activity in suspension.Chemical treatments had no effect on R, which remained at 100%for Hc1 throughout the pH range and for Sc at the most acid pH forboth soils. At pH above about 4.5, chemical treatments led to lowervalues of R on the ferralsol, but removal of organic matter protectedactivity, with a smaller decline in R.

4. Discussion

The most striking feature of the data presented in this study isthat both surface affinity and relative activity in the adsorbed stateshow greater contrast between enzymes than between soils or soilfractions for a given enzyme. This diversity would contribute to therobustness of natural ecosystems to change, to heterogeneousconditions in time and space and to the multitude of functionsrequired in a soil (Turner, 2008). The comparison also points to the

ity of the adsorbed enzyme (R) as functions of pH calculated from data presented in Fig.1.ars show experimental variability between triplicates, but are not always visible.

Fig. 3. Effect of soil particle size on adsorption of AcPase Sc and AcPase Hc1 as a function of pH: non-adsorbed fraction (F) with soil clay (closed symbol) and sand-sized fraction(open symbol). Bars show experimental variability between triplicates, but are not always visible.

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e20 17

caution required in generalising observations made for a singleenzyme, particularly if that enzyme was not exposed to soil ormineral surfaces in its natural environment.

4.1. Affinity of enzymes for soil surfaces

This is the first study to compare the affinity of different enzymeson contrasting soils over a range of pH. In most cases little change inaffinity with pH was observed, in contrast with the strong pHdependence reported in some cases by Leprince and Quiquampoix(1996) and Quiquampoix and Burns (2007) for mineral surfacesand by George et al. (2007b) for soils. S. collinitus phosphatasespresent an exception since affinity decreased (increasing F) withincreasing pH in contact with ferralsol, particularly after removal oforganic surface coatings. This pH dependence is consistent withincreased repulsion between like-charged surfaces. It would be ex-pected to be more marked at high surface coverages where sterichindrance would also contribute to the decline (Helassa et al., 2009;Leprince and Quiquampoix, 1996). The free fraction was often in therange 20e40%. The average adsorption is smaller and the contrastbetween soils less than that previously reported for fungal phos-phatases on mineral surfaces (Quiquampoix and Burns, 2007).

In agreement with the observation of George et al. (2007b) forfungal phytases we observe a greater affinity for the clay-sized

Fig. 4. Effect of chemical treatments of soils on adsorption of AcPase Sc and AcPase Hc1 as a ftreatment (open symbol) and H2O2 treatment (shaded symbol). Bars show experimental va

fractions than the sand fractionof a given soil. Given the trace amountof phosphatases in each system, this cannot be attributed to theadsorption capacity of a surface.When different soils were comparedthere was no relationship between affinity and clay content. Incontrast to studies of the adsorption capacity of surfaces in thepresence of large amounts of protein (George et al., 2007b; Helassaet al., 2009; Nannipieri et al., 1996) there was no relation betweenadsorption affinity and specific surface area (comparison of soils or ofsoil after chemical treatments that modified specific surface area).Only in the case of particle size was the difference in affinityconsistently in line with changed surface area for each soileenzymecombination. The greater affinity of the enzymes for the clay-sizedfraction of each soil agrees with the observation of Marx et al.(2005) who found that soil phosphatases were accumulated in theclay-sized fraction of a soil in preference to coarser size fractions.However in that study, other enzymes were not accumulated in theclay-sized fraction so the observation may be due in part to theaccumulation of natural substrate in certain size fractions, and notsolely to accumulation of secreted enzymes.

Despite differences in organic matter content and CBD-extractable Fe in the ferralsols, in which the principal mineralwas kaolinite, each of the enzymes showed very similar affinitiesfor each of the four ferralsols studied. This suggests that the effectof the organo-mineral coatings on affinity was small. Nevertheless

unction of pH: non-adsorbed fraction (F) with untreated soil (closed symbol), after CBDriability between triplicates, but are not always visible.

Fig. 5. Effect of soil particle size on relative activity of adsorbed enzyme (R) of AcPase Sc and AcPase Hc1 as a function of pH. Closed symbols represent soil clay and open symbolssand-sized fraction. Bars show experimental variability between triplicates, but are not always visible, given the symbol size.

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e2018

removal of organic matter from soil tended to enhance the affinityof the enzymes for the surfaces. The effect was greater for theferralsol, particularly at the more acid pH and greater for the Suillusthan the Hebeloma enzymes. The contrast between enzymes illus-trates that the effect does not simply arise from a mechanicalcreation of new adsorption sites by dispersion or the liberation ofsites previously occupied by organic molecules. It is interesting tonote that Sc also showed a very strong affinity for pure mineral StAustell kaolinite (data not shown) with almost complete adsorp-tion for pH in the range 3e4.5. The smaller effect on the vertisolcould reflect the recalcitrance of organic matter in this soil, withonly a small proportion of organic matter being removed (25%, seeMikutta et al., 2005 for a review of chemical removal of organicmatter from soils). There are limited published data on the effect oforganic matter on either the adsorption maximum or the affinity ofsmall amounts of enzyme on mineral surfaces. Previous resultsfrom this group (Quiquampoix, 1987b; Quiquampoix et al., 1995)led us to expect a protective effect of organic matter, with chemicalremoval of organic matter leading to enhanced affinity andadsorption capacity. Huang et al. (2005) report very small increasesin affinity of potato acid phosphatases when soil organic matterwas oxidised from either the fine or coarse clay fraction of a red soil(Ultisol with a dominant kaolin clay fraction). Yan et al. (2010)

Fig. 6. Effect of chemical treatments of soils on relative catalytic activity of adsorbed enzuntreated soil, open symbols after CBD treatment and shaded symbols H2O2 treatment. Ba

report little effect of organic matter destruction on the Langmuiraffinity parameter for b-glucosidase on soil clay.

4.2. Relative activity of adsorbed phosphatase

Changes in activity after adsorption may result from changes inconformation and from preferred orientation of the active site.When changes in activity between the solution and adsorbed phaseare reported for soils and soil minerals, in most cases there is nochange or a reduction in activity (George et al., 2007b; Huang et al.,2009; Kelleher et al., 2004; Lammirato et al., 2010; Leprince andQuiquampoix, 1996; Shindo et al., 2002; Yan et al., 2010). Anotable exception is the enhanced activity reported for an acidphosphatases adsorbed on allophane (Rosas et al., 2008). We haveobservedmarkedly different pH dependence of R: constant at about100%, initially 100 then a decrease after a threshold pH, or increasewith increasing pH. A comparison of R values made at a single pH,such as the optimum pH, would not have shown such contrastingpatterns. Leprince and Quiquampoix (1996) observed that F and Rof fungal phosphatases in contact with montmorillonite followedsimilar trends with pH, which could have been interpreted asa relationship between affinity and strength of interaction. Incontrast, this study did not show any shared pattern between F and

yme (R) of AcPase Sc and AcPase Hc1 as a function of pH. Closed symbols representrs show experimental variability between triplicates, but are not always visible.

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e20 19

R for different enzymeemineral pairs. It is striking that in nearlyevery case the value of R at the optimal pHwas about 100%, the onlyclear exception being for Sc on the vertisol. This would suggest thatthese fungal enzymes are particularly well adapted to the presenceof soil minerals. There are advantages to an organism if the extra-cellular enzymes it produces are strongly adsorbed, since adsorp-tion which may confer protection against microbial degradationand limit mobility so that phosphate hydrolysed would be closeenough to be rapidly absorbed. These advantages would be out-weighed if catalytic efficiency was lost.

4.3. Effect of particle size and organo-mineral coatings on adsorbedphosphatase activity

Particle size did not modify the pH dependence of R despitemineralogical difference in sand and clay fractions. In all caseswhen R w100% for whole soil or soil clay, the same was observedfor the sand fraction. In the case of Sc in contact with vertisol sandor clay, a decrease in R was observed with increasing pH, but thevalue of R for sand fraction was always lower by about 20%. Thiscould reflect the greater surface saturation by the enzyme on thesand fraction since the surface area is much smaller. Other studieshave observed a similar texture effect on activity loss afteradsorption on soil clay, albeit when comparing clay and fine clay(0.2e2 mm and <0.2 mm) and at much greater surface coverages,approaching surface saturation (Huang et al., 2005; Yan et al., 2010).

Neither the comparison of ferralsols, with similar dominantmineralogy, nor the effect of chemical treatments to remove organo-mineral coatings point to a simple mechanism to determine therelative activity of adsorbed enzymes. The different pH trends of Robserved for the four ferralsols did not follow the same patterns aseither organic carbon or CBD-extractible Fe and so cannot be used toelucidate the effect of organo-mineral coatings. Neither chemicaltreatment to remove organo-mineral coatings had any effect on thevalue of R for Hc1 in contact with either soil, which remained at 100%throughout the pH range. Both treatments led to the value of R forSuillus enzyme in contact with soil F1 to fall markedly at pH above 5,in contrast to the untreated soil where R had remained stable.Removal of organic matter from the vertisol led to some protectionsof the activity of S, with a greater proportion of activity beingmaintained at near-neutral pH. This protective effect of organicmatter has been reported both by this group and others (Giavenoet al., 2010; Quiquampoix, 1987b; Quiquampoix et al., 1995; Raoand Gianfreda, 2000). But other studies suggest an inhibitory effectof organic matter, both in solution and in solid state (Huang et al.,2005; Rosas et al., 2008; Vuorinen and Saharinen, 1996).

However the protective effect is clearly not a general effect. Itdepends on both the soil and the enzyme. It must also depend onthe nature of the organic coating removed and the extent of theremoval. Interestingly, there is no link between the change inaffinity and the resulting variation in R. Increased affinity, partic-ularly on ferralsol, was not accompanied by stronger interactionand loss of activity. Chemical removal of soil organic matter had noeffect on the activity of Hc1 phosphatase which maintained itsactivity after adsorption on either soil, whereas contrasting effectsfor Sc phosphatase were observed, particularly at near-neutral pHon the two soils.

5. Conclusions

Both the affinity of fungal phosphatases for these tropical soilsand subsequent decrease in catalytic activity after adsorption showconsiderable diversity. In some cases affinity and relative activity inthe adsorbed state varied with pH and so observations made ata single pH value, as is often the case in other studies, may not give

a complete representation of interactions. Affinity varied greatlybetween enzymes, even between enzymes obtained from the twostrains of H. cylindrosporum. Only Suillus phosphatases showedmarkedly different affinities for ferralsol and vertisol. In general,adsorbed phosphatases retained a large proportion of their cata-lytic activity over the whole pH range studied. Only Suillus phos-phatase adsorbed on the vertisol showed a marked decrease in Rwith increasing pH. Enzymes would thus be efficient after secretioninto soil. There was no relationship between the affinity of anenzyme for a surface and the resulting change in activity.

Acknowledgements

The authors thank Catherine Pernot who maintains the fungalcollection, Bernard Barthes who helped to select soil samples andBenoît Cloutier-Hurteau for advice on statistical analysis. BKacknowledges a grant from the Côte d’Ivoire Ministry of HigherEducation.

Appendix. Supplementary material

Supplementary data related to this article can be found online atdoi:10.1016/j.soilbio.2012.02.006.

References

Barthès, B.G., Kouakoua, E., Larré-Larrouy, M.-C., Razafimbelo, T.M., de Luca, E.F.,Azontonde, A., Neves, C.S.V.J., de Freitas, P.L., Feller, C.L., 2008. Texture andsesquioxide effects on water-stable aggregates and organic matter in sometropical soils. Geoderma 143, 14e25.

Bradford, M.M., 1976. Rapid and sensitive method for quantitation of microgramquantities of protein utilizing principle of protein-dye binding. AnalyticalBiochemistry 72, 248e254.

Cairney, J., 2011. Ectomycorrhizal fungi: the symbiotic route to the root for phos-phorus in forest soils. Plant and Soil 344, 51e71.

Chevallier, T., Voltz, M., Blanchart, E., Chotte, J.L., Eschenbrenner, V., Mahieu, M.,Albrecht, A., 2000. Spatial and temporal changes of soil C after establishment ofa pasture on a long-term cultivated vertisol (Martinique). Geoderma 94, 43e58.

FAO-ISRIC-ISSS, 1998. World Reference Base for Soil Resources, Soil ResourcesReport N� 84, UN Food and Agriculture Organization, Rome.

George, T.S., Quiquampoix, H., Simpson, R.J., Richardson, A.E., 2007a. Interactionsbetween phytases and soil constituents: implications for the hydrolysis ofinositol phosphates. In: Turner, B.L., Richardson, A.E.,Mullaney, E.J. (Eds.), InositolPhosphates: Linking Agriculture and the Environment. CABI, Wallingford, UK,pp. 221e241.

George, T.S., Simpson, R.J., Gregory, P.J., Richardson, A.E., 2007b. Differential inter-action of Aspergillus niger and Peniophora lycii phytases with soil particlesaffects the hydrolysis of inositol phosphates. Soil Biology and Biochemistry 39,793e803.

Giaveno, C., Celi, L., Richardson, A.E., Simpson, R.J., Barberis, E., 2010. Interaction ofphytases with minerals and availability of substrate affect the hydrolysis ofinositol phosphates. Soil Biology and Biochemistry 42, 491e498.

Harrison, A.F., 1987. Soil Organic Phosphorus. A Review of World Literature. CABInternational, Wallingford, Oxon, UK, 257 pp.

Helassa, N., Quiquampoix, H., Noinville, S., Szponarski, W., Staunton, S., 2009.Adsorption and desorption of monomeric Bt (Bacillus thuringiensis) Cry1Aatoxin on montmorillonite and kaolinite. Soil Biology and Biochemistry 41,498e504.

Huang, Q., Liang, W., Cai, P., 2005. Adsorption, desorption and activities of acidphosphatase on various colloidal particles from an Ultisol. Colloids and SurfacesB: Biointerfaces 45, 209e214.

Huang, Q., Zhu, J., Qiao, X., Cai, P., Rong, X., Liang, W., Chen, W., 2009. Conformation,activity and proteolytic stability of acid phosphatase on clay minerals and soilcolloids from an Alfisol. Colloids and Surfaces B: Biointerfaces 74, 279e283.

Kelleher, B.P., Willeford, K.O., Simpson, A.J., Simpson, M.J., Stout, R., Rafferty, A.,Kingery, W.L., 2004. Acid phosphatase interactions with organo-mineralcomplexes: influence on catalytic activity. Biogeochemistry 71, 285e297.

Kouakoua, E., Larre-Larrouy, M.C., Barthes, B., de Freitas, P.L., Neves, C., Sala, G.H.,Feller, C., 1999. Relationship between aggregate stability and total and hot-water-extractable organic matter in clayey ferrallitic soils (Congo, Brazil).Canadian Journal of Soil Science 79, 561e569.

Kouakoua, E., Sala, G.-H., Barthès, B., Larre-Larrouy, M.-C., Albrecht, A., Feller, C.,1997. La matière organique soluble à l’eau chaude et la stabilité de l’agrégation.Aspects methodologiques et application à des sols ferrallitiques du Congo.European Journal of Soil Science 48, 239e247.

Krasilnikov, P., Ibá�nez Martí, J.J., Arnold, R., Shoba, S., 2009. A Handbook of SoilTerminology, Correlation and Classification. Earthscan, London.

B. Kedi et al. / Soil Biology & Biochemistry 56 (2013) 13e2020

Lammirato, C., Miltner, A., Wick, L.Y., Kästner, M., 2010. Hydrolysis of cellobiose byb-glucosidase in the presence of soil minerals e interactions at solid-liquidinterfaces and effects on enzyme activity levels. Soil Biology and Biochem-istry 42, 2203e2210.

Lavkulich, L.M., Wiens, J.H., 1970. Comparison of organic matter destruction byhydrogen peroxide and sodium hypochlorite and its effects on selected mineralconstituents. Soil Science Society of America Journal 34, 755e758.

Leprince, F., Quiquampoix, H., 1996. Extracellular enzyme activity in soil: effect ofpH and ionic strength on the interaction with montmorillonite of two acidphosphatases secreted by the ectomycorrhizal fungus Hebeloma cylin-drosporum. European Journal of Soil Science 47, 511e522.

Marx, M.C., Kandeler, E., Wood, M., Wermbter, N., Jarvis, S.C., 2005. Exploring theenzymatic landscape: distribution and kinetics of hydrolytic enzymes in soilparticle-size fractions. Soil Biology and Biochemistry 37, 35e48.

Matumoto-Pintro, P.T., 1996. Rôle des phosphatases dans l’utilisation du phosphoreorganique par les champignon ectomycorhiziens et leurs associations avec lepin laricio de Corse. Ph.D. thesis. University of Montpellier, France, 197 pp.

Mayor, J.R., Henkel, T.W., 2006. Do ectomycorrhizas alter leaf-litter decompositionin monodominant tropical forests of Guyana? New Phytologist 169, 579e588.

McLaren, A.D., Estermann, E.F., 1957. Influence of pH on the activity of chymotrypsinat a solid-liquid interface. Archives of Biochemistry and Biophysics 68, 157e160.

Mehra, O.P., Jackson, M.L., 1960. Iron oxide removal from soils and clays bya dithionite-citrate system buffered with sodium bicarbonate. Clays and ClayMinerals 7, 317e327.

Mikutta, R., Kleber, M., Kaiser, K., Jahn, R., 2005. Review: Organic matter removalfrom soils using hydrogen peroxide, sodium hypochlorite, and disodium per-oxodisulfate. Soil Science Society of America Journal 69, 120e135.

Morizet, J., Mingeau, M., 1976. Effect of environment on water-uptake, as studied onbeheaded exudating tomato. 1. Role of nutrients. Annales Agronomiques 27,183e205.

Nannipieri, P., Sequi, P., Fusi, P., 1996. Humus and enzyme activity. In: Alessandro, P.(Ed.), Humic Substances in Terrestrial Ecosystems. Elsevier Science B.V.,Amsterdam, pp. 293e328.

Quiquampoix, H., 1987a. A stepwise approach to the understanding of extracellularenzyme activity in soil. I. Effect of electrostatic interactions on the conformationof a b-D-glucosidase on different mineral surfaces. Biochimie 69, 753e763.

Quiquampoix, H., 1987b. A stepwise approach to the understanding of extracellularenzyme activity in soil. II. Competitive effects on the adsorption of a b-D-glucosidase in mixed mineral or organo-mineral systems. Biochimie 69,765e771.

Quiquampoix, H., 2000. In: Bollag, J.M., Stotzky, G. (Eds.), Mechanisms of ProteinAdsorption on Surfaces and Consequences for Extracellular Enzyme Activity inSoil. Marcel Dekker, New York, pp. 171e206.

Quiquampoix, H., Burns, R.G., 2007. Interactions between proteins and soil mineralsurfaces: environmental and health consequences. Elements 3, 401e406.

Quiquampoix, H., Mousain, D., 2005. Enzymatic hydrolysis of organic phosphorus.In: Turner, B.L., Frossard, E., Baldwin, D.S. (Eds.), Organic Phosphorus in theEnvironment. CABI Publishing, Wallingford UK, pp. 89e112.

Quiquampoix, H., Abadie, J., Baron, M.H., Leprince, F., Matumoto Pintro, P.T.,Ratcliffe, R.G., Staunton, S., 1995. Mechanisms and consequences of proteinadsorption on soil mineral surfaces. ACS Symposium Series 602, 321e333.

Quiquampoix, H., Servagent-Noinville, S., Baron, M.H., 2002. Enzyme adsorption onsoil mineral surfaces and consequences for the catalytic activity. In: Burns, R.G.,Dick, R.P. (Eds.), Enzymes in the Environment: Activity, Ecology, and Applica-tions. Marcel Dekker, Inc., New York, pp. 285e306.

Rao, M.A., Gianfreda, L., 2000. Properties of acid phosphatase-tannic acid complexesformed in the presence of Fe and Mn. Soil Biology and Biochemistry 32,1921e1926.

Rao, M.A., Gianfreda, L., Palmiero, F., Violante, A., 1996. Interactions of acid phos-phatase with clays, organic molecules and organo-mineral complexes. SoilScience 161, 751e760.

Rao, M.A., Violante, A., Gianfreda, L., 2000. Interaction of acid phosphatase withclays, organic molecules and organo-mineral complexes: kinetics and stability.Soil Biology and Biochemistry 32, 1007e1014.

Razafimbelo, T., Albrecht, A., Basile, I., Borschneck, D., Bourgeon, G., Feller, C., Ferrer, H.,Michellon, R., Moussa, N., Muller, B., Oliver, R., Razanamparany, C., Seguy, L.,Swarc,M., 2006a. Effet dedifférents systèmesde culture à couverture végétales surle stockage du carbone dans un sol argileux des Hautes Terres de Madagascar.Etude et Gestion des Sols 13, 113e127.

Razafimbelo, T., Barthès, B., Larré-Larrouy, M.-C., Luca, E.F.D., Laurent, J.-Y.,Cerri, C.C., Feller, C., 2006b. Effect of sugarcane residue management (mulchingversus burning) on organic matter in a clayey Oxisol from southern Brazil.Agriculture, Ecosystems & Environment 115, 285e289.

Rosas, A., de la Luz Mora, M., Jara, A.A., López, R., Rao, M.A., Gianfreda, L., 2008.Catalytic behaviour of acid phosphatase immobilized on natural supports in thepresence of manganese or molybdenum. Geoderma 145, 77e83.

Rowell, D.L., 1994. Soil Science: Methods and Applications. Longman Group Limited,Longman Scientific & Technical, Harlow, UK, 350 pp.

Shindo, H., Watanabe, D., Onaga, T., Urakawa, M., Nakahara, O., Huang, Q.Y., 2002.Adsorption, activity, and kinetics of acid phosphatase as influenced by selectedoxides and clay minerals. Soil Science and Plant Nutrition 48, 763e767.

Soil Survey Staff, 1999. Soil Taxonomy: A Basic System of Soil Classification forMaking and Interpreting Soil Surveys, second ed. USDA Handbook 436 USGovernment Printing Office, Washington, DC.

Turner, B.L., 2008. Resource partitioning for soil phosphorus: a hypothesis. Journalof Ecology 96, 698e702.

Turner, B.L., 2010. Variation in pH optima of hydrolytic enzyme activities in tropicalrain forest soils. Applied & Environmental Microbiology 76, 6485e6493.

Turner, B.L., Engelbrecht, B.M.J., 2011. Soil organic phosphorus in lowland tropicalrain forests. Biogeochemistry 103, 297e315.

Vance, C.P., Uhde-Stone, C., Allan, D.L., 2003. Phosphorus acquisition and use:critical adaptations by plants for securing a nonrenewable resource. NewPhytologist 157, 423e447.

Violante, A., DeCristofaro, A., Rao, M.A., Gianfreda, L., 1995. Physicochemical prop-erties of protein-smectite and protein-Al(OH)x-smectite complexes. ClayMinerals 30, 325e336.

Vuorinen, A.H., Saharinen, M.H., 1996. Effects of soil organic matter extracted fromsoil on acid phosphomonoesterase. Soil Biology and Biochemistry 28,1477e1481.

Yan, J., Pan, G., Li, L., Quan, G., Ding, C., Luo, A., 2010. Adsorption, immobilization,and activity of b-glucosidase on different soil colloids. Journal of Colloid andInterface Science 348, 565e570.