Soil enzyme activities as affected by anthropogenic alterations: intensive agricultural practices and organic pollution

Science of the Total Environment 341 (2005) 265 – 279 www.elsevier.com/locate/scitotenv

Soil enzyme activities as affected by anthropogenic alterations: intensive agricultural practices and organic pollution Liliana Gianfredaa,*, Maria Antonietta Raoa, Anna Piotrowskaa,1, Giuseppe Palumbob, Claudio Colombob a

Dipartimento di Scienze del Suolo, della Pianta e dell’Ambiente Sezione Scienze Chimico-Agrarie, Universita` di Napoli Federico II Via Universita` 100, 80055 Portici (Napoli), Italy b Dipartimento di Scienze Animali, Vegetali e dell’Ambiente, Campobasso, Italy Received 9 January 2004; accepted 7 October 2004

Abstract The activity of a range of enzymes related to the cycling of the main biologically important nutrients C, N, P and S was investigated in cultivated and non-cultivated soils from various parts of Europe. Two agricultural sites from North Italy under continuous corn (Zea mays L.) with and without organic fertilization were compared. Two other agricultural sites from South Italy under hazel (Corylus avellana L.) never flooded or repeatedly flooded over by uncontrolled urban and industrial wastes were investigated. The non-cultivated soils were from Middle and South Europe with different pollution history such as nopollution and pollution with organic contaminants, which is phenanthrene and other polycyclic aromatic hydrocarbons (PAHs). Agricultural soils showed significant differences in some of physical–chemical properties (i.e. organic C, total and labile phosphate contents, available Ca and Mg) between the two sites studied. Enzyme activities of hazel sites periodically flooded by wastes were mainly higher than in the hazel sites never flooded. Sites under many years of continuous corn showed dehydrogenase, invertase, arylsulphatase and h-glucosidase activities generally lower than the soils under hazel either flooded or not by wastes. As compared to agricultural soils, non-cultivated soils heavily or moderately polluted by organic contaminants displayed much lower values or complete absence of enzymatic activities. Dissimilar, contradictory correlations between soil enzyme activities and the majority of soil properties were observed separately in the two groups of soils. When the whole set of enzyme activities and soil properties were considered, all significant correlations found separately for the groups of soils were lost. The overall results seem to confirm that no direct

* Corresponding author. Tel.: +39 081 2539179, lab: +39 081 2539166; fax: +39 081 2539186. E-mail address: [email protected] (L. Gianfreda). 1 Permanent address: Department of Biochemistry, Faculty of Agriculture, University of Technology, Bydgoszcz, Poland. 0048-9697/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2004.10.005

266

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

cause–effect relationships can be derived between the changes of a soil in response to a given factor and both the variations of the activity and the behaviour of the enzymes in soil. D 2004 Elsevier B.V. All rights reserved. Keywords: Enzyme activities; Agricultural soils; Non-cultivated soils; Anthropogenic alterations; Organic pollution

1. Introduction Soil is a living dynamic, non-renewable, resource and its conditions influence food production, environmental efficiency and global balance (Dick, 1997; Doran and Parkin, 1994; Doran and Zeiss, 2000). The quality of soil depends in part on its natural composition, and also on the changes caused by human use and management (Pierce and Larson, 1993). Human factors influencing the environment of the soil can be divided into two categories: those resulting in soil pollution and those devoted to improve the productivity of soil (Gianfreda and Bollag, 1996). Unusual management of the soil, such as intensive cultivation without crop rotation (Reeves, 1997), or accidental/deliberate contamination by municipal and industrial wastes (Edwards, 2002), are major causes of land degradation and reduced soil productivity. The determination of the quality-related properties of soil (which are sensitive to changes caused by management practices and environmental stress) may help to monitor the changes in its sustainability and environmental quality. This is especially true for the agricultural management and recovery of soil, and to assist into the establishment of policies for the use of land. Soil enzymes activities have been suggested as suitable indicators of soil quality because: (a) they are a measure of the soil microbial activity and therefore they are strictly related to the nutrient cycles and transformations; (b) they rapidly may respond to the changes caused by both natural and anthropogenic factors; (c) they are easy to measure (Gianfreda and Bollag, 1996; Drijber et al., 2000; Calderon et al., 2000; Colombo et al., 2002; Nannipieri et al., 2002). Moreover, as claimed by several authors (Dick and Tabatabai, 1993; Dick, 1997, van Beelen and Doelman, 1997, Trasar-Cepeda et al., 2000), soil enzymes activities may be considered early and sensitive indicators to measure

the degree of soil degradation in both natural and agro-ecosystems, being thus well suited to measure the impact of pollution on the quality of soil. Among soil pollutants, major environmental concerns are heavy metals and polycyclic aromatic hydrocarbons (PAHs) (Smreczek et al., 1999). The presence of heavy metals in the soil may influence the biochemical processes by affecting both microbial proliferation and enzyme activities. Heavy metals may inhibit enzyme activities by masking catalytically active groups, having denaturing effects on the conformation of proteins, or competing with the metal ions involved in the formation of enzyme–substrate complexes (Gianfreda and Bollag, 1996 and references therein). As multi-ring by-products of the incomplete combustion of several organic pollutants, PAHs are considered hazardous soil contaminants. In soil, they may have different fates. The sorption of organic contaminants by natural soil colloids (organic and inorganic) often limits their bioavailability as substrates, thus affecting their microbial degradation rate in the soil (Grosser et al., 2000). When bio-available, PAHs may strongly affect the biological and biochemical activity of soil. In the present study we have measured the activity of a range of enzymes, related to the cycling of the main biologically important nutrients (C, N, P, S) in soils of two typologies: agricultural and non-cultivated soils. The selected soils were characterized by different origins, properties and history. We studied four soils from two large agricultural sites located in North and South Italy and characterized by different management practices (continuous monoculture cultivation and accidental overflooding with polluted wastewater), and four non-cultivated sites polluted by organic and inorganic pollutants located in different parts of Europe. Attempts were made to find some cause–effect relationships, if any, between soil properties, depending on their management and/or pollution, and enzyme activity levels.

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

2. Materials and methods 2.1. Sampling sites 2.1.1. Agricultural soils Two large sites were studied, characterized by a typical Mediterranean humid, climate with short summers and mild winters (an average annual rainfall of 900 mm and a temperature of 13–17 8C) from North and South Italy, with different types of soil management and alteration degree. (1) The first sites are located in the middle of the Cremonese lowland, crossed by the Po River in North Italy, and formed from sedimentary limestone materials. The area was subjected to natural and anthropogenic hydrologic alterations. Natural alterations are linked to the dramatic changes often occurring in the water level of the Po River. Anthropogenic causes include the construction and the modification of banks as well as the construction of pig farms and animal buildings near the Po River. Both these alterations can result in increasing levels of inundation, and, in turn, in dramatic effects on soil such as soil loss by water erosion or soil biological degradation. Furthermore, soils of this area are often affected by point sources of pollution (e.g. heavy metals and treated wastes from dairy farm). The soils are alluvial soils often with big problems of hydromorphism connected to the low depth of the vadose zone located between 2 and 4 m. The area has a long history (more than 20 years) of continuous corn (Zea mays L.) production for cattle and pig husbandry. No organic fertilization was performed. Neighbouring continuous corn soils, highly productive and amended with organic fertilization, were selected as control soils. (2) The second sites are volcanic areas from the Palma Campania Plain, South Italy. The area is a fertile volcanic valley located on the base of a calcareous formation of the Campanian Apennin Mountains (South Western Italy), at 50 m above sea level. It is mainly an intensive agricultural farmland, cultivated with hazel (Corylus avellana L.), strongly overcrowded

267

(N500,000 inhabitants), with diffuse activities from small industries. The area is close to Sarno Mountains and is surrounded by Somma-Vesuvius and Sarno Mountains. Both are affected by an active water erosion of volcanic incoherent soils by surface run-off with a consequent detachment of surface soils. Sudden, fast and extremely dangerous mud flows on the area often result. To reduce intensive water erosive events, a network of channels (namely dRegi LagniT built in the early 19th century) crosses the area and collects the rainfall water of Sarno Mountains into a basin of 1095 km2. The Regi Lagni channels are also affected by non-point sources of pollution from dairy farm, urban effluents; and wastes from small manufacturing industries. In the last 20 years, frequent intense rains caused a repeated overflooding of the area by uncontrolled urban and industrial wastes (alluvial sediments, civil and industrial sludge) from the open drains. The wastes contained oils and industrial fats, heavy metals and other organic toxic substances deriving from agricultural or industrial activities. They were not decontaminated. Hazel sites never subjected to any flooding event were the corresponding control soils. 2.1.2. Non-cultivated soils Four soils, three from the Middle Europe and one from the South Europe, having a different pollution history were studied. Namely: (1) An aged, heavily polluted soil, at an altitude of 180 m and characterized by a relatively coldclimate (average annual temperature of 10 8C and rainfall of 800 mm). The soil (thereafter called 50-year-petroleum polluted soil) was exposed for a long time (N50 years) to petroleum contamination. According to the information provided by the site’s owner, the site is still heavily contaminated. (2) Two moderately polluted soils, characterized by a 3-year exposition to different levels (~32 and 9 mg kg 1 soil) of PAHs (thereafter called, respectively, 3-year-high [PAH] and 3-year-low [PAH] polluted soils). The soils were subjected to an accidental pollution event that caused a spread

268

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

distribution of PAHs on its surface. The soils were sampled after 3 years from the pollution event. (3) A non-polluted soil (thereafter called non-polluted soil) from South Europe, with an insignificant presence of pollutants. 2.2. Soil sampling and processing Five samples were collected in a 1 ha area, following a W scheme, with a sample taken at each angle or tip of the W. At each sampling point, four representative sub-samples were taken at random from a depth of 5 to 15 cm. The 0–5 cm soil layer was discarded to reduce spatial variability and also possible point contamination. The four sub-samples were taken with a soil auger (F 4 cm), mixed, pooled in the field, giving five independent composite samples, and transferred in sealed plastic bags to the laboratory. The five composite samples were sieved to b2 mm, homogenized and stored at 4 8C until the analysis. The samples of the agricultural soils were collected twice after a 12-month interval, i.e. in March 2000, and March 2001. The sampling of the noncultivated soils was done in March 2001 and no further sampling was performed. 2.3. Chemical properties determination Chemical analyses were done according to the Methods of Soil Analysis (1996) on air-dried and sieved (b2 mm) soil samples. Particle size distribution analysis was carried out by the pipette method; pH was measured in 1:2.5 soil/water suspension; organic matter was determined by dichromate oxidation; total N with Kjeldahl method; labile phosphate with bicarbonate extraction; exchangeable bases (K and Na) and heavy metals were determined after acid digestion with HF/HNO3 (Liu et al., 2002). Overall content of PAHs and phenanthrene amounts of noncultivated polluted soils was determined by standard gas-chromatography procedures as described in Saccomandi and Gianfreda (2001). 2.4. Enzyme activity assays Enzyme activities were determined on fresh moist sieved (b2 mm) soils within 15–20 days from the collection of the samples. Arylsulphatase (ARYL), h-

glucosidase (GLU) and phosphatase (PHO) were determined as described by Tabatabai and Bremner (1970), Eivazi and Tabatabai (1990) and Sannino and Gianfreda (2001), respectively. Specific substrates ( pnitrophenylsulphate, p-nitrophenyl-h-d-glucoside, pnitrophenylphosphate) and buffers were used for each enzyme. A pH close to the natural soil pH was usually utilized. Urease (UR) activity was measured according to Kandeler and Gerber (1988). For the determination of invertase (INV) activity, the method used by Sannino and Gianfreda (2001) was adopted, using saccharose as substrate. The determination of the dehydrogenase (DH) activity was based on the use of soluble tetrazolium salt as an artificial acceptor (Trevors, 1984) and the activity of the o-diphenol oxidase (DPO) was determined using a mixture of catechol and proline as the substrate as described by Perucci et al. (2000). With non-cultivated soils strong interferences were observed in the determination of some enzyme activities. In particular, the colorimetric determination of invertase, h-glucosidase and phenoloxidase reaction products were strongly affected by the presence of soils (i.e. blanks lacking the substrate and containing only soil gave strongly coloured mixtures). Therefore, the activity of these three enzymes was not determined in these soils. The activity of fluorescein diacetate hydrolase (FDAH), representative of a large arrays of hydrolytic enzymes, was then determined. The fluorescein diacetate hydrolase (FDAH) activity was assessed as described by Adam and Duncan (2001). A unit of enzyme activity was defined as the Ag of substrate hydrolysed at 30 8C h 1 by 1 g of dried soil. Control tests with autoclaved soils were carried out to evaluate the spontaneous or abiotic transformation of substrates. All chemical determinations and enzymatic activities were determined in triplicate. 2.5. Statistical analysis The results are expressed as arithmetic means of 53 data at each time point. The standard deviations of the means and correlations among the means of all variables were performed by Statistica Package for Windows (Version 5.1 Edition 98).

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

3. Results 3.1. Physical–chemical properties of agricultural and non-cultivated soils Tables 1 and 2 summarize the physical and chemical properties of investigated soils and the amounts of both organic and inorganic pollutants. 3.1.1. Agricultural soils No significant differences were measured for the physical–chemical properties of soils at the two sampling times. Therefore, the results discussed in the following refer to mean values calculated on the triplicates of 52 data sets. Soils from the two different cultivation sites showed a differentiated range of physical–chemical properties (Table 1). The hazel cultivated soils from South Italy were sandy–loamy, quite deep, and well drained with high field capacity. Soil characteristic formations in the Regi Lagni areas were strongly influenced by co-alluvial and alluvial processes from the erosive phenomena. Volcanic material occurring on the mountains was co-alluviated to the valley, forming thick, fertile soils with low weathering and moderate andic properties. The continuous corn soils

269

from North Italy were alluvial and hydromorphic with shallow ground water and showed sandy–loamy characteristics. According to pair-wise analysis statistically significant differences in some of these properties were observed between the two sites as well as within the hazel site between never and periodically flooded soil and in the continuous corn site between the fertilized and never-fertilized soils. No a common behaviour or trend was observed. The agricultural soils from both sites were subacidic to neutral soils with pH ranging from 6.3 to a maximum value of 7.1 measured in the continuouscorn fertilized soils of North Italy. The different management of the two soils resulted mainly in significantly different organic C contents. Both continuous-corn soils from North Italy showed amounts of organic C, characteristic of cultivated soils, with higher values for the fertilized soils. By contrast, the hazel soils from South Italy, particularly those periodically flooded, exhibited very high C levels, probably because of the overflooding of urban carbon rich sediments from the open drains into the soil. Since no significant differences in N content among soils were measured, very different C/N ratios derived (Table 1).

Table 1 Physico-chemical variables measured in agricultural soilsa Soil properties

Hazel sites periodically flooded

Hazel sites never flooded

Continuous corn-fertilized sites

Continuous corn never fertilized sites

pH (H2O) Organic carbon (g kg 1) Total nitrogen (g kg 1) C/N ratio Total P (mg kg 1) Labile P (mg kg 1) Available Ca (cmol(+) kg 1) Available Mg (cmol(+) kg 1) Available Na (cmol(+) kg 1) Available K (cmol(+) kg 1) Total Cu (mg kg 1) Total Zn (mg kg 1) Total Cr (mg kg 1) Total Ni (mg kg 1) Total Fe (g kg 1) Clay (g kg 1) Silt (g kg 1) Sand (g kg 1)

6.9 24.2 2.7 9.1 2612 47 18.3 1.8 0.5 1.8 139 134 8 24 36 101 284 615

6.3 19.5 1.9 10.6 3463 22 11.5 0.8 0.4 1.4 148 96 55 24 46 71 200 729

7.1 10.3 1.5 6.7 1317 21 15.2 1.4 0.08 0.2 265 77 57 18 25 167 99 734

7.0 8.7 1.8 5.0 797 30 14.7 2.9 0.11 0.4 18 62 38 20 22 205 167 627

a

(F0.2) (F0.3) (F0.5) (F0.9) (F285) (F4) (F2.4) (F0.1) (F0.2 (F0.1) (F10) (F12) (F1) (F6) (F1) (F2) (F5) (F51)

(F0.2) (F0.3) (F0.4) (F1.5) (F95) (F3) (F1.6) (F0.1) (F0.1) (F0.3) (F21) (F7) (F12) (F7) (F2) (F4) (F18) (F29)

(F0.1) (F0.2) (F0.4) (F0.3) (F57) (F2) (F0.8) (F0.3) (F0.01) (F0.1) (F23) (F8) (F16) (F6) (F3) (F3) (F2) (F41)

Values represent means calculated on 52 soil samples (three replicates) (n=52)Fstandard deviation.

(F0.3) (F0.5) (F0.2) (F0.4) (F16) (F3) (F1.8) (F0.4) (F0.01) (F0.1) (F2) (F2) (F2) (F6) (F0.7) (F3) (F2) (F19)

270

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

Total P content ranged from 797 to 3463.0 mg kg 1 and was the highest in the hazel soils never flooded from South Italy. By contrast, greater labile phosphate levels (POls) were measured in the soils periodically flooded than in the corresponding never flooded ones. Phosphorus fertilization was possibly performed on these soils to reduce the decline of soil productivity due to the sediments overflooding. High P contents could have resulted. A consequent inhibitory effect on the phosphatase activity of this soil could be also expected (see below). Available Ca and Mg were significantly higher in hazel soils periodically flooded than in hazel sites never flooded from South Italy. An opposite situation for Mg occurred in the soils from North Italy. Available Na content in investigated soils ranged widely from 0.08 cmol(+) kg 1 in the continuous corn-fertilized soils from North Italy to 0.5 cmol(+) kg 1 in the hazel soils periodically flooded from South Italy. The highest amount of available K was found in South Italy soils, whereas in North Italy soils its level was fourfold lower. The levels of heavy metals were not elevated and all below the maximum permitted concentrations for agricultural soils (Commission of the European Communities, 1986).

3.1.2. Non-cultivated soils Non-cultivated soils showed a sub- to moderatealkaline character as indicated by the pH values (Table 2). The lowest value of pH was shown by the 50-year-petroleum polluted soil (Table 2). The amounts of clay, silt and sand fractions (Table 2) indicate the 50-year-petroleum polluted soil and the non-polluted soil as being sandy–clay loam soils, and both 3-year-high and low [PAH] polluted soils being typically loamy sand soils. Moderate to high total organic C values were measured for the three soils polluted by organic contaminants. These values, however, could be considered not representative of intrinsic organic C contents of the soils, being the soils contaminated by high levels of organic contaminants. Low amounts of N measured in both soils, and higher C/N ratios than those normally found in unpolluted soils were obtained (Asseng et al., 2001). The amounts of heavy metals were not elevated and within the range of permitted concentrations (Commission of the European Communities, 1986). A very high amount of Cu (300 mg kg 1 soil) was present in the non-polluted soil. The three soils from Middle Europe resulted heavily polluted by high concentrations of PAH

Table 2 Physico-chemical variables measured in non-cultivated soilsa Soil properties

50-year-petroleum polluted soil

3-year-high [PAH]b polluted soil

3-year-low [PAH]c polluted soil

Non-polluted soil

pH (H2O) Organic carbon (g kg 1) Total nitrogen (g kg 1) C/N ratio Labile-P (mg kg 1) Available K (mg kg 1) Total Cu (mg kg 1) Total Zn (mg kg 1) Total Cr (mg kg 1) Total Ni (mg kg 1) Total Fe (g kg 1) Clay (g kg 1) Silt (g kg 1) Sand (g kg 1) PAHs (mg kg 1) PHenanthrene (mg kg 1)

6.6 (F0.2) 11 (F1) 0.40 (F0.09) 28 (F3) Trace nd 135 (F9) 80 (F9) 14 (F3) 39 (F8) 5 (F2) 247 (F14) 150 (F9) 603 (F57) 100 (F6.4) 15 (F2.6)

8.1 8.5 0.68 12.5 14 220 48 120 84 55 39 69 77 853 32 5

8.2 8.1 1.4 5.8 21 485 51 315 67 65 33 69 86 845 9 0.5

7. 8 (F0.6) 7.8 (F1.2) 2.3 (F0.5) 3.4 (F0.45) 35 (F4) 340 (F18) 300 (F21) 118 (F9) 70 (F6) 76 (F8) 40 (F5) 225 (F19) 245 (F24) 470 (F45) nd nd

(F1.1) (F2) (F0.09) (F1.2) (F1) (F13) (F5) (F7) (F5) (F6) (F6) (F5) (F6) (F67) (F3.2) (F0.7)

nd=not detected. a Values represent means calculated on 52 soil samples (n=52)Fstandard deviation. b [PAH]=32 mg kg 1 soil. c [PAH]=9 mg kg 1 soil.

(F0.6) (F0.9) (F0.08) (F0.4) (F0.8) (F18) (F8) (F17) (F6) (F8) (F2) (F5) (F7) (F43) (F0.87) (F0.06)

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

with phenanthrene relatively the most abundant (Table 2). In the 50-year-petroleum polluted soil alkanes, BTEX and phenols were also detected (data not shown). No relevant levels of pollutants were measured in the non-polluted soil, which could be thus considered as a non-cultivated, control soil. 3.2. Enzyme activities in the soils 3.2.1. Agricultural soils Enzyme activities were within the range of values usually found in soil (Nannipieri et al., 2002). They varied widely with the sampling site and soil properties (Fig. 1). Enzyme activities of hazel sites periodically flooded were generally higher than in hazel sites never flooded at all sampling times in South Italy. Only PHO and UR showed an opposite trend (Fig. 1). In the North Italy site, the overall trend of a number of enzymes was similar to the South Italy one. Differences among North Italy values were, however, much smaller and often not statistically significant (Fig. 1). Our study showed different trends for the set of enzyme activities investigated and for the same enzyme at different sampling times. While arylsulphatase activity values increased by about 2.5–3.0 times in the hazel sites periodically flooded compared to the sites never flooded, the enzyme activities were similar in the continuous cornfertilized soils and the continuous corn never fertilized soils of North Italy (Fig. 1). DH reached high levels in both North Italy and South Italy soils, irrespective of soil alteration (Fig. 1). Moreover, the enzyme activity differed between never flooded and periodically flooded South Italy soils, being the latter value significantly higher than the former. An opposite behaviour occurred for North Italy soils, where continuous corn never fertilized soils showed higher DH levels than the fertilized ones. In South Italy samples, the GLU activity of flooded soils increased after 1 year, whereas that of controls remained level or increased very slightly (Fig. 1). The activity was significantly higher (from 40% to 97%) in flooded than in control soils at all sampling times (Fig. 1). By contrast, small differences were detected between the two investigated North Italy soils (Fig. 1).

271

INV activity did not differ between flooded and never flooded South Italy soils (Fig. 1). A statistically significant difference was only observed at the second sampling time between the continuous corn-fertilized soil and the never fertilized one, with a lower content in the fertilized samples (Fig. 1). The activity of UR was lower for hazel sites periodically flooded than for the never flooded at the second sampling. The activity fell to about 50% in March 2001, whereas only a 10% decrease was measured at the March 2000 sampling (Fig. 1). At North Italy site, some differences between fertilized and never fertilized samples, although not statistically significant, were detected at the two sampling times (Fig. 1). The overall levels of PHO detected in South Italy soils were higher than those in North Italy soils (Fig. 1). As reported above, the values observed in the soil periodically flooded with wastes, lower than in the never flooded soil, could be explained by an inhibitory effect of labile phosphate contents. The o-diphenoloxidase activity was usually higher in altered than in control soils at both the sites (Fig. 1). 3.2.2. Non-cultivated soils In the Middle Europe soils polluted by organic contaminants (i.e. 50-year-petroleum polluted soil, 3year-high [PAH] and 3-year-low [PAH] polluted soils) it was not possible to determine the activities of hglucosidase, invertase and phenoloxidase. Indeed, a consistent interference, probably by the organics therein present, occurred with the colorimetric procedures adopted to determine h-glucosidase and invertase-reaction products. The presence of phenols in soils also interfered with the analytical assay of the phenoloxidase activity. To have further information on the overall enzymatic activity of those soils, the activity of fluorescein diacetate hydrolase (FDAH), a measurement of the hydrolytic activity of a soil, was determined. As compared to agricultural soils, non-cultivated polluted soils in general showed lower values of enzymatic activities (Fig. 2). In the non-polluted soil, however, the enzyme activity showed moderate to high range levels, and comparable to those measured in agricultural soils (Nannipieri et al., 2002). A relatively low dehydrogenase activity was, however, measured in this soil, possibly because of the possible interference

272

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

Fig. 1. Enzymatic activities of agricultural soils, sampled in March 2000 and 2001. Error bars represent standard error of the means calculated on five soil samples (three replicates) at each time point.

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

273

Fig. 2. Enzymatic activities of non-cultivated soils, sampled in March 2001. Error bars represent standard error of the means calculated on five soil samples (three replicates) at each time point.

by its high copper content (Table 2) on the analytic assay used in the determination of DH reaction product (Chander and Brookes, 1991). The presence of organic contaminants was reflected in very low enzymatic activities (Fig. 2). The 50-year-petroleum polluted soil, heavily and for long-term polluted by PAHs, did not show dehydrogenase activity, at all, and had a very low level of urease. Urease and dehydrogenase were totally absent, or at not detectable levels, in both the 3-year-high and -low PAH polluted soils. Phosphatase activity was the greatest (over 1500 Agram PNP g 1 h 1), also if compared to agricultural soils. 3.3. Relationship between enzyme activities and physical–chemical properties In order to find some cause–effect relationships, if any, between soil properties, depending on their

managements and/or pollution, and enzyme activity levels, the enzymatic activities were correlated between themselves and physical and chemical soil properties, and levels of heavy metals and organic pollutants. The Pearson’s correlation coefficient was used to quantify the strength of enzyme activities relationships vs. chemical properties. In this study, the lower limits of significance for correlation between variables in agricultural and non-cultivated soils were rz0.590 and z0.520 and pV0.05, respectively (Tables 3–6). 3.3.1. Agricultural soils Arylsulphatase activity was highly and positively (r=0.883**) correlated with soil pH, whereas a negative relationship was detected between phosphatase activity and pH (r= 0.726*) (Table 3). Differently to what was expected, statistical analysis showed a significant positive correlation only

274

Table 3 Correlation matrix between enzymatic activities and physico-chemical variables (agricultural soils) Enzymes Physical–chemical variablesa pH H2O

O.C.

Ntot

Ptot

Plab

Clay

Silt

Sand

Available forms Ca

0.883** 0.726* 0.621 0.022 0.065 0.496 0.096

0.347 0.767* 0.892** 0.682 0.670 0.659 0.755*

0.049 0.349 0.702 0.773* 0.819* 0.469 0.804*

0.701 0.931*** 0.828* 0.349 0.291 0.650 0.515

0.345 0.590* 0.429 0.712* 0.765* 0.215 0.514

0.623 0.936*** 0.819* 0.393 0.347 0.657 0.580

0.212 0.435 0.793* 0.648 0.720* 0.554 0.578

0.424 0.381 0.078 0.413 0.438 0.120 0.040

0.704 0.246 0.069 0.614 0.665 0.001 0.511

Total forms Na

0.555 0.851** 0.467 0.099 0.013 0.436 0.412

K

0.372 0.719* 0.896** 0.661 0.664 0.650 0.693

Cu 0.443 0.729* 0.901** 0.610 0.614 0.654 0.635

Zn

0.078 0.398 0.013 0.113 0.050 0.098 0.425

Cr

0.121 0.630 0.817* 0.766* 0.799* 0.620 0.857**

Ni

0.325 0.010 0.430 0.693 0.767* 0.239 0.510

Fe

0.688 0.636 0.873** 0.333 0.446 0.704 0.420

0.783* 0.899** 0.824* 0.268 0.240 0.672 0.451

n =20. ARYL=Arylsulphatase, PHO=Phosphatase, UR=Urease, DH=Dehydrogenase, GLU=Glucosidase, DPO=Phenoloxidase, INV=Invertase. a The units of each variables are those given in Table 1 and Fig. 1. * pb0.05. ** pb0.01. *** pb0.001.

Table 4 Correlation matrix between enzymatic activities and physico-chemical variables (non-cultivated soils) Physical-chemical variablesa pH H2O a

ARYL PHO UR DH FDAH

0.217 0.798*** 0.060 0.027 0.497

O.C.

0.216 0.625* 0.441 0.454 0.168

Ntot 0.844*** 0.025 0.886*** 0.940** 0.465

Pols 0.576* 0.325 0.771*** 0.737** 0.194

Clay

0.635* 0.670** 0.635* 0.520* 0.650**

Silt

0.842*** 0.346 0.959*** 0.867*** 0.608*

Sand

0.747** 0.471 0.850*** 0.707** 0.633*

Kavail 0.560* 0.193 0.454 0.614* 0.259

Phen

0.423 0.446 0.555* 0.558* 0.061

PAH

0.422 0.457 0.523* 0.589* 0.051

n=20. ARYL=Arylsulphatase, PHO=Phosphatase, UR=Urease, DH=Dehydrogenase, FDAH=Fluorescein diacetate hydrolase. a The units of each variables are those given in Table 2 and Fig. 2. * pb0.05. ** pb0.01. *** pb0.001.

Total forms Cu

Zn

0.833*** 0.311 0.967*** 0.878*** 0.567*

0.370 0.000 0.032 0.243 0.312

Cr 0.041 0.778*** 0.243 0.311 0.456

Ni 0.702** 0.189 0.779*** 0.867*** 0.321

Fe 0.142 0.832*** 0.221 0.147 .432

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

ARYLa PHO UR DH GlU DPO INV

Mg

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

between organic C and phosphatase and invertase activities. A positive correlation was, also, determined between total N amount and the three C-carbon cycleinvolved enzymes: dehydrogenase (r=0.733*), hglucosidase (r=0.819*) and invertase (r=0.804*) (Table 3). A high significant and positive correlation coefficient between the phosphatase activity and total P content was obtained. By contrast, when correlation was made with labile phosphate (POls) a significant negative coefficient (r= 0.590*) was found. It is also noteworthy to mention that phosphatase significantly but negatively correlated with the clay content (r= 0.936***). Urease was the most sensitive to the presence of heavy metals. Indeed significant negative correlations were found between urease activity and Zn (r= 0.817*), Ni (r= 0.873**) and Fe (r= 0.824*). Significant, but positive correlations were also observed between Zn and GLU, DH and INV activities. When enzyme activities were correlated to each other, significant correlations were found between the dehydrogenase, h-glucosidase and invertase activities (Table 5), thus confirming some of the above results and in agreement with several findings reported in the literature (Gianfreda and Bollag, 1996, Nannipieri et al., 2002 and references therein). An interesting correlation was found between dehydrogenase and phenoloxidase. Both enzymes may give indications on the oxidative potential of a soil. Hydrolytic enzymes such as urease and arylsulphatase also were highly Table 5 Correlation coefficients between enzyme activities in agricultural soils ARYLa a

INV DPO GLU DU UR PHO

0.458 0.209 0.353 0.146 0.640* 0.204

PHO 0.087 0.619* 0.174 0.361 0.474

UR

DH

GLU

DPO

0.531 0.556 0.524 0.492

0.665* 0.678* 0.790**

0.872*** 0.231

0.151

n=20. ARYL=Arylsulphatase, PHO=Phosphatase, UR=Urease, DH= Dehydrogenase, GLU=Glucosidase, DPO=Phenoloxidase, INV= Invertase. a The units of each variables are those given in Fig. 1. * pb0.05. ** pb0.01. *** pb0.001.

275

Table 6 Correlation coefficients between enzyme activities in non-cultivated soils a

FDAH DH UR PHO

ARYLa

PHO

UR

DH

0.807*** 0.845*** 0.815*** 0.529*

0.759** 0.129 0.120

0.472 0.929***

0.455

n=20. ARYL=Arylsulphatase, PHO=Phosphatase, UR=Urease, DH= Dehydrogenase, FDAH=Fluorescein diacetate hydrolase. a The units of each variables are those given in Fig. 2. * pb0.05. ** pb0.01. *** pb0.001.

correlated to each other as well as with dehydrogenase activity (Table 5). 3.3.2. Non-cultivated soils In non-cultivated polluted soils, urease and dehydrogenase activities proved to be more sensitive to several chemical properties and pollutant levels. They were negatively correlated with the contents of PAH and phenanthrene, the main pollutants present in soil (Table 4). Soil Ntot and Pols contents were significantly and positively correlated to arylsulphatase, urease and dehydrogenase activity. A negative, although not significant, correlation was found between phosphatase and the labile phosphorous content POls. Organic C content was correlated significantly only with phosphatase activity (r=0.625*). Total Cu content was significantly and positively correlated with almost all enzyme activities except for the phosphatase activity (r values ranging from 0.567* to 0.967***). Strong positive coefficients of correlation were found for the Ni content and arylsulphatase, urease and dehydrogenase activities. In these soils, quite all enzyme activities significantly and positively correlated to each other and they were significantly and positively correlated with clay and silt contents and negatively with sand (Tables 4 and 6). Attempts were also made to correlate the whole set of enzyme activities, common to the two groups of soils, and the soil properties. No significant correlations were obtained and in addition several of those derived separately for the two groups of soils were lost.

276

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

4. Discussion The activities of enzymes related to the cycling of the main biologically important nutrients C, N, P and S were investigated in soils as affected by anthropogenic alterations. Dehydrogenase, h-glucosidase and invertase are involved in the C-cycle. Soil DH reflects the soil oxidative power. Being present in all microorganisms, it may give a measure of the total viable microbial cells. INV and GLU catalyze hydrolytic processes in the organic matter breakdown. Urease, phosphatase and arylsulphatase play an important role in the mineralization of nitrogen, phosphorous and sulphur compounds. They may give indications on the soil potential to perform specific biochemical reactions, and are important in contributing to soil fertility. The o-diphenoloxidase activity is related to the capacity of a soil to perform the oxidation of phenolic compounds, including potentially recalcitrant organic ones. Its involvement in the formation of humic substance in soil has been many times proposed. Our findings demonstrated that the enzyme activities in soil are affected by the presence and nature of the plant cover. A long-term intensive monoculture usually supplies lower amounts and diversity of organic matter than crop rotation, thus suppressing microbial activities ad consequently decreasing enzymatic ones (Klose and Tabatabai, 2000). Furthermore, the total metabolic activity of plant roots and the soil microorganisms, developing under these conditions, lead to the production, accumulation, and action of only a few classes of enzymes. Soils taken from North Italy with a long history of continuous corn monoculture, without proper amendment with organic matter, showed low organic matter contents and low dehydrogenase, invertase, arylsulphatase and h-glucosidase as compared with continuous corn-fertilized soils and other investigated agricultural soils. Furthermore, the high total Cu content observed in this soil may still be explained by monoculture practices, which rely on large chemical inputs particularly rich in this element. Enzyme activities of soils are usually correlated either with their organic C and/or total N contents (Deng and Tabatabai, 1997; Acosta-Martinez and Tabatabai, 2000; Dodor and Tabatabai, 2002; Taylor et al., 2002). Indeed, organic C is the main constituent of the soil organic matter, and as such it may represent

a source of enzyme production but also a substrate for enzyme degradation. In our investigations, only phosphatase and invertase activities were significantly correlated with organic C content in agricultural soils, whereas a negative correlation was observed with urease. In non-cultivated polluted soils a positive significant correlation was found for phosphatase. Considering that in some of examined soils (e.g. hazel soils periodically flooded, and the three soils polluted by petroleum and high and low levels of PAHs) the measured organic C contents can be mostly attributed to the presence of organic wastes, rather than to organic biomass, these results could indicate that the high organic C content does not necessarily reflect corresponding increases of enzymatic activities. Soil phosphatase activities were positively correlated to total phosphorous, thus confirming the involvement of these enzymes in the P cycle. However, they negatively correlated to the available form of phosphate POls, thus confirming that a phosphatase activity is usually inversely proportional to labile P content (Tadano and Sakai, 1991, Gianfreda and Bollag, 1996 and references herein). The phosphate ions, products of phosphatase hydrolysis, are competitive inhibitors of this enzyme (Gianfreda and Bollag, 1996 and references herein). Soil enzymes activities are usually significantly correlated to soil pH. Positive, negative or no correlations have been reported (Kang and Freeman, 1999; Acosta-Martinez and Tabatabai, 2000; Canet et al., 2000). Accordingly, arylsulphatase of agricultural soils was significantly and positively correlated to pH, whereas a different behaviour (i.e. a negative, statistically significant correlation) was exhibited by phosphatase. Although phosphatase activity was determined at the natural soil pH, the endogenous soil phosphatase might show an optimal pH different from that used in the enzymatic assay. In addition, phosphatase activity significantly correlated to clay (r= 0.936***) and organic carbon (r=0.767*). These results could indicate that the enzyme is prevalently present as an extracellular enzyme bound to inorganic and organic soil colloids. In this case, a different behaviour in terms of enzyme activity levels and activity vs. pH dependence is expected (Nannipieri and Gianfreda, 1998). Among the various elements of inorganic origin responsible for soil pollution, heavy metals are by far

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

the most important. Once metals enter the soil, they remain there for long periods of time without being destroyed by the soil microorganisms, whereas molecules of organic origin can be microbially degraded (Blum, 1989). Furthermore, the effect of heavy metals on enzyme activities may vary considerably among the elements, enzymes and soils. Indeed, it is connected both with physical and chemical properties of the soil, especially soil organic matter and clay content, as well as to the kind of enzyme and metal involved. Some authors have proposed enzyme activities, especially dehydrogenase, urease and phosphatase, as indicators of soil contamination with heavy metals (Tyler, 1976, Al-Khafaji and Tabatabai, 1979, Rogers and Li, 1985, Welp, 1999). Accordingly, urease activity of agricultural soils, here investigated, was negatively influenced by Zn, Ni, and Fe. The situation of non-cultivated polluted soils is much more complex being the soils contaminated with high levels of PAHs and with heavy metals. Few studies have been performed of the influence of PAHs on the soil enzymatic activity so far. MaliszewskaKordybuch and Smreczek (2003) demonstrated that contamination with PAHs and alkanes inhibited the dehydrogenase activity in several soils. Among all biological indexes tested by the authors (dehydrogenase activity, phosphatase activity, total bacteria number and intensity of respiration) the dehydrogenase activity appeared to be the most sensitive parameter to PAH contamination (Maliszewska-Kordybuch and Smreczek, 2003). A high applicability of the dehydrogenase activity for soil ecotoxicological testing, as well as a sensitive index for measuring the toxicity of heavy metals and polycyclic aromatic hydrocarbons on microbial communities present in soils was also suggested by several authors (Rossel et al., 1997, Ihra et al., 2003). In our investigations, dehydrogenase activity was not detected in the soil, polluted for a long time (over 50 years) with petroleum and very high levels of PAHs (100 mg kg 1). Significant and negative correlation coefficients between phenanthrene and PAHs contents and dehydrogenase and urease activities were also obtained, thus suggesting the two enzymes being the most sensitive to this kind of pollutions. Furthermore, the higher the amount of PAHs, the lower the arylsulphatase and FDA hydrolase activities. The use of these two enzymes for

277

quantification of soil degradation caused by PAHs could be also suggested. According to Elliott (1997) pollution indicators should have not only sensitivity to the presence of pollutants but also ability to reflect different levels of pollutants. The presence of highly significant intercorrelations between the five enzyme activities and between them and inorganic soil colloids (clay, silt and sand) may suggest that in these soils a consistent fraction of enzymes was prevalently present as extracellular enzymes associated to inorganic soil colloids. The absence of any plant cover as well as of organic amendment may have repressed the synthesis and production of new enzyme molecules. Lower enzyme activities and different responses to natural and anthropogenic factors may then result.

5. Conclusive remarks One of the main limitations in the interpretation of soil properties assessments including enzyme activities is that it is not possible to generalize the results obtained with a particular group of soils to another group of soils differing for their properties and characteristics. For this reason, the investigations here reported were deliberately performed on completely different soil types with the aim to find, if any, some relationships applicable to a more general level. Although restricted to a limited number of soils, some of our results seem to be encouraging. For instance, dehydrogenase and urease activities showed significant and negative relationships with phenanthrene and PAHs content, thus suggesting that these enzymes could be proposed as sensitive indicators in soil polluted with these compounds. The overall of results have unfortunately demonstrated that no direct correlations or relationships can be derived between the soil status and both the activity levels and behaviours of the enzymes. For instance, correlations between soil enzyme activities and heavy metals were completely different in the two groups of soils. Furthermore, all significant correlations found separately for the groups of soils were lost when the whole set of enzyme activities and soil properties were considered. In our opinion, the shortcomings of soil enzyme activities to be used as soil quality indicators may rely on the fact that investigations are usually performed

278

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279

under laboratory conditions and not in situ, and as such they are affected by the methodologies used for evaluating and assaying their enzyme activity levels. The available methodologies do not discriminate neither among the various components contributing to the overall enzymatic activity of soil nor between enzyme and enzyme-like activities. Moreover, they often are not applicable to soils that are so different from each other that it is difficult to handle them in the same way. The effects of interference observed with non-cultivated soils and analytical procedures used to detect the product of some enzyme action (invertase, h-glucosidase and phenoloxidase) seem to confirm this assumption. In conclusion, the use of soil enzyme activities as a sensitive warning of soil functioning is still problematic. Indeed, it is particularly difficult to explain a change of soil enzymatic activity in response to a certain factor or to establish the cause–effect relationships between the applied disturb and the soil enzyme activity variation.

Acknowledgments This research was supported by Ministero dell’Universita´ e della Ricerca Scientifica e Tecnologica, Italy, Programmi di Interesse Nazionale PRIN 2000–2001 and PRIN 2002–2003. The authors are grateful to Dr. Anna Maria Woods for her technical help in editing the paper in respect to the English language and style. DiSSPA Contribution no. 82.

References Acosta-Martinez V, Tabatabai MA. Enzyme activities in a limed agricultural soil. Biol Fertil Soils 2000;31;85 – 91. Adam G, Duncan H. Development of a sensitive and rapid method for the measurement of total microbial activity using fluorescein diacetate (FDA) in a range of soils. Soil Biol Biochem 2001; 33;943 – 51. Al-Khafaji AA, Tabatabai MA. Effect of trace elements on arylsulphatase activity in soils. Soil Sci 1979;127;129 – 33. Asseng S, Turner NC, Keating BA. Analysis of water- and nitrogenuse efficiency of wheat in a Mediterranean climate. Plant Soil 2001;233;127 – 43. Blum WEH. Soil Pollution By Heavy Metals: Causes, Processes, Impact And Need For Future Actions. Proceeding of the 3rd

meeting. Steering Committe for the conservation and management of the environment and natural habitants (CDPE), Strasbourg; 1989. Calderon JF, Jackson LE, Scow KM, Rolston DE. Microbial responses to simulated tillage in cultivated and uncultivated soils. Soil Biol Biochem 2000;32;1547 – 59. Canet R, Albiach R, Pomares F. Indexes Of Biological Activity As Tools For Diagnosing Soil Fertility In Organic Farming. In: Garcia C, Herna´ndez T, editors. Research And Perspectives Of Soil Enzymology In Spain. Murcia, Spain7 Cebas-CSIC; 2000. p. 27 – 39. Chander K, Brookes PC. Is the dehydrogenase assay invalid as a method to estimate microbial activity in copper-contaminated soils? Soil Biol Biochem 1991;23;909 – 15. Colombo C, Palumbo G, Sannino F, Gianfreda L. Chemical and biochemical indicators of managed agricultural soils. 17th World Congress of Soil Science, Bangkok, Thailand, vol. 1740; 2002. p. 1 – 9. Commission of the European Communities. Council directive on the protection of the environment and in particular the soil, when sewage sludge is used in agriculture. Off J Eur Communities 1986;L81;6 – 12 Annex 1A. Deng SP, Tabatabai MA. Effect of tillage and residue management on enzyme activities in soils: III Phosphatases and arylsulfatase. Biol Fertil Soils 1997;24;141 – 6. Dick RP. Soil enzyme activities as integrative indicators of soil health. In: Pankhurst CE, Doube BM, Gupta VVSR, editors. Biological Indicators of Soil Health. Wallingford, USA7 CAB International; 1997. p. 121 – 56. Dick WA, Tabatabai MA. Significance and potential uses of soil enzymes. In: Metting FB, editor. Soil Microbial Ecology: Application in Agricultural and Environmental Management. New York7 Marcel Dekker; 1993. p. 95 – 125. Dodor DE, Tabatabai MA. Effects of cropping system and microbial biomass on arylamidase activity in soils. Biol Fertil Soils 2002;35;253 – 61. Doran JW, Parkin TB. Defining and assessing soil quality. In: Doran JW, editor. Defining Soil Quality for Sustainable Environment. SSSA Special Publication. Madison, WI7 Soil Science Society of America, Inc. and American Society of Agronomy, Inc.; 1994. p. 3 – 23. Doran JW, Zeiss MR. Soil health and sustainability: managing the biotic component of soil quality. Appl Soil Ecol 2000;15;3 – 11. Drijber RA, Doran JW, Parkhurst AM, Lyon DJ. Changes in soil microbial community structure with tillage under long-term wheat–fallow management. Soil Biol Biochem 2000;32; 1419 – 30. Edwards CA. Assessing the effects of environmental pollutants on soil organisms, communities, processes and ecosystems. Eur J Soil Biol 2002;38;225 – 31. Eivazi F, Tabatabai MA. Factors affecting glucosidase and galactosidase activities in soils. Soil Biol Biochem 1990;22; 891 – 7. Elliott ET. Rationale for developing bioindicators of soil health. In: Pankhurst CE, Doube BM, Gupta VVSR, editors. Biological Indicators of Soil Healthy. Wellington7 CAB International; 1997. p. 49 – 78.

L. Gianfreda et al. / Science of the Total Environment 341 (2005) 265–279 Gianfreda L, Bollag JM. Influence of natural and anthropogenic factors on enzyme activity in soil. In: Stotzky G, Bollag JM, editors. Soil Biochemistry, vol. 9. New York7 Marcel Dekker; 1996. p. 123 – 94. Grosser RJ, Friedrich M, Ward DM, Inskeep WP. Effect of model sorptive phases on phenanthrene biodegradation: different enrichment conditions influence bioavailability and selection of phenanthrene-degrading isolates. Appl Environ Microbiol 2000;66;2695 – 702. Ihra N, Slet J, Petersell V. Effect of heavy metals and PAH on soil assessed via dehydrogenase assay. Environ Intern 2003;28; 779 – 82. Kandeler E, Gerber H. Short-term assay of soil urease activity using colorimetric determination of ammonium. Biol Fertil Soils 1988;6;68 – 72. Kang H, Freeman C. Phosphatase and arylsulphatase activities in wetland soils: annual variation and controlling factors. Soil Biol Biochem 1999;31;449 – 54. Klose S, Tabatabai MA. Urease activity of microbial biomass in soil. Biol Fertil Soils 2000;6;68 – 72. Liu F, Colombo C, Adamo P, He JZ, Violante A. Trace elements in manganese–iron nodules from a Chinese Alfisol. Soil Sci Soc Am J 2002;66;661 – 70. Maliszewska-Kordybuch B, Smreczek B. Habitat function of agricultural soils as affected by heavy metals and polycyclic aromatic hydrocarbons contamination. Environ Intern 2003;28; 719 – 28. Methods of Soil Analysis. Part 3 Chemical Methods. Sparks D.L., editor. SSA Book Series, vol. 5. Madison, WI: Soil Science Society of America, Inc. and American Society of Agronomy, Inc.; 1996. Nannipieri P, Gianfreda L. Kinetics of enzyme reactions in soil environments. In: Huang PM, Senesi N, Buffle J, editors. Environmental Particles—Structure and Surface Reactions of Soil Particles. Madison, WI7 J. Wiley and Sons; 1998. p. 449 – 79. Nannipieri P, Kandeler E, Ruggiero P. Enzyme activities and microbiological and biochemical processes in soil. In: Burns RP, Dick RP, editors. Enzymes in the Environment Activity, Ecology and Applications. New York7 Marcel Dekker; 2002. p. 1 – 33. Perucci P, Casucci C, Dumontet S. An improved method to evaluate the o-diphenol oxidase activity of soil. Soil Biol Biochem 2000;32;1927 – 33. Pierce FJ, Larson WE. Developing criteria to evaluate sustainable land management. In: Kimble JM, editor. Proceeding of the VIII International Soil Management Workshop Utilization of

279

soil survey information for sustainable land use Sacramento, CA; 1993. p. 7 – 14. Reeves DW. The role of soil organic matter in maintaining soil quality in continuous cropping systems. Soil Tillage Res 1997;43;131 – 67. Rogers JR, Li SW. Effect of metals and other inorganic ions on soil microbial activity: Soil dehydrogenase assay as a simple toxicity test. Bull Environ Contam Toxicol 1985;34;856 – 8. Rossel, D, Tarradellas, J, Bitton, G, Morel, JL, Use of enzymes in soil ecotoxycology: a case for dehydrogenase and hydrolytic enzymes. In: Tarradellas, J, Bitton, G, Rossel, D, editors. Soil ecotoxycology. Boca Raton, Fl: Lewis Publ., CRC Press; 1997. p. 179–206. Saccomandi F, Gianfreda L. Can intrinsic microbial population restore an bagedQ heavily NAPL-polluted site? Prospect and Limits of Natural Attenuation at tar Oil Contaminated SitesDodreacht Hordbomd7 DECHEMA e.V;2001 p. 486 – 92. Sannino F, Gianfreda L. Pesticide influence on soil enzymatic activities. Chemosphere 2001;22;1 – 9. Smreczek B, Maliszewska-Kordybuch B, Martyniuk S. Effect of PAHs and heavy metals on activity of soil microflora. In: Baveye P, Block J-C, Goncharuk VV, editors. Bioavailability Of Organic Xenobiotics in the Environment. Dodreacht Hordbomd7 Kluwer Academic Publishing; 1999. p. 377 – 80. Tabatabai MA, Bremner JM. Arylsulphatase activity of soils. Soil Sci Soc Am Proc 1970;34;427 – 9. Tadano T, Sakai H. Secretion of acid phosphatase by roots of several crop species under phosphorus-deficient conditions. Soil Sci Plant Nutr 1991;37;129 – 40. Taylor JP, Wilson B, Mills MS, Burns RG. Comparison of microbial numbers and enzymatic activities in surface soils and subsoils using various techniques. Soil Biol Biochem 2002;34;387 – 401. Trasar-Cepeda C, Leiro´s MC, Seoane S, Gil-Sotres F. Limitation of soil enzymes as indicators of soil pollution. Soil Biol Biochem 2000;32;1867 – 75. Trevors JT. Deydrogenase in soil: a comparison between the INT and TTC assay. Soil Biol Biochem 1984;16;673 – 4. Tyler G. Heavy metal pollution, phosphatase activity, and mineralization of organic phosphorus in forest soils. Soil Biol Biochem 1976;8;327 – 32. van Beelen PV, Doelman P. Significance and application of microbial toxicity tests in assessing ecotoxicological risks of contaminants in soil and sediments. Chemosphere 1997;43; 455 – 99. Welp G. Inhibitory effects of the total and water-soluble concentrations of nine different metals on the dehydrogenase activity of a loess soil. Biol Fertil Soils 1999;30;132 – 9.