Ptaquiloside in Pteridium aquilinum subsp. aquilinum and corresponding soils from the South of Italy: Influence of physical and chemical features of soils on its occurrence

Science of the Total Environment 496 (2014) 365–372

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Ptaquiloside in Pteridium aquilinum subsp. aquilinum and corresponding soils from the South of Italy: Influence of physical and chemical features of soils on its occurrence Claudio Zaccone a,⁎, Ivana Cavoski b, Roberta Costi c, Giorgia Sarais d, Pierluigi Caboni d, Andreina Traversa a, Teodoro M. Miano e a

Department of the Sciences of Agriculture, Food and Environment, University of Foggia, via Napoli 25, 71122 Foggia, Italy Mediterranean Agronomic Institute of Bari-CIHEAM, via Ceglie 9, 70010 Valenzano, Italy c Dipartimento di Chimica e Tecnologie del Farmaco, Università di Roma “La Sapienza”, p.le Aldo Moro 5, 00185 Rome, Italy d Department of Life and Environmental Sciences, University of Cagliari, via Ospedale 72, 09124 Cagliari, Italy e Department of Soil, Plant and Food Sciences, University of Bari “Aldo Moro”, via Amendola 165/A, 70126 Bari, Italy b

H I G H L I G H T S • • • •

Ptaquiloside concentration in Pteridium aquilinum ranged from 2 to 780 mg kg− 1. Ptaquiloside production was affected by nutrient availability (mainly P) in soil. Ptaquiloside concentration in soil samples was always undetectable (b 0.015 mg kg− 1). The degradation of ptaquiloside by indigenous soil microbial community is suggested.

a r t i c l e

i n f o

Article history: Received 9 April 2014 Received in revised form 24 June 2014 Accepted 14 July 2014 Available online xxxx Editor: Charlotte Poschenrieder Keywords: Bracken fern DT50 Kinetic models Soil properties Southern Europe

a b s t r a c t The bracken fern Pteridium aquilinum (L.) Kuhn, one of the most common plant species on Earth, produces a wide range of secondary metabolites including the norsesquiterpene glucoside ptaquiloside (PTA). Several studies are present in literature about eco-toxicological aspects related to PTA, whereas results about the effect of growth conditions and soil properties on the production and mobility of PTA are sometimes conflicting and further investigations are needed. The aim of the present work is to investigate the occurrence and possible fate of PTA in soils showing different physical and chemical features, and collected in several areas of the South of Italy. The PTA content was determined in both soil and fern samples by GC–MS; both the extraction protocol and recovery were previously tested through incubation studies. Soils samples were also characterized from the physical and chemical points of view in order to correlate the possible influence of soil parameters on PTA production and occurrence. PTA concentration in P. aquilinum fern seemed to be significantly affected by the availability of nutrients (mainly P) and soil pH. At the same time, PTA concentration in soil samples was always undetectable, independent of the PTA concentration in the corresponding Pteridium samples and pedo-climatic conditions. This seems to suggest the degradation of the PTA by indigenous soil microbial community, whereas incubation studies underlined a certain affinity of PTA for both organic colloids and clay/silt particles. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Pteridium aquilinum (L.) Kuhn (Pteridaceae) is considered to be one of the five most successful and widely distributed organisms of the plant kingdom (Taylor, 1990), being present in all continents except Antarctica; the only limitations to its distribution are extreme

⁎ Corresponding author. Tel.: +39 0881 598119. E-mail address: [email protected] (C. Zaccone).

http://dx.doi.org/10.1016/j.scitotenv.2014.07.046 0048-9697/© 2014 Elsevier B.V. All rights reserved.

temperatures and lack of humidity (Smith and Seawright, 1995). In Italy, where it occurs only as P. aquilinum subsp. aquilinum, it is widely present from the sea level until 2100 m a.s.l., islands included (Pignatti, 1982; Conti et al., 2005), often forming extensive populations. Bracken ferns have been linked with significant health concerns for both grazing domestic animals and human populations (Fletcher et al., 2011; Gil da Costa et al., 2012). The major bracken carcinogen is a water-soluble norsesquiterpene glucoside named ptaquiloside (PTA) (Hirono, 1987; Ojika et al., 1987). Domestic animals grazing on lands rich in bracken fern (e.g., horses, pigs, sheep and cattle) can be affected

366

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

by several acute and chronic syndromes being attributed to ingestion of bracken fern (Bonadies et al., 2011). Besides animals, also humans could be affected by bracken, both directly and indirectly. PTA, as well as other compounds of bracken, may interact with human tissues alone and/or in combination with infectious agents such viruses, particularly papillomaviruses (Campo, 1997), while epidemiological evidence suggests that PTA causes cancer in humans (Smith and Seawright, 1995). Furthermore, P. aquilinum is consumed as a human food in Japan and some other countries (Yamada et al., 1998), whereas milk and meat from bracken-eating cows is postulated to be one of the links between the carcinogen effects of bracken and gastric cancer in some geographic areas worldwide (Alonso-Amelot et al., 1996, 1998; Bonadies et al., 2011; Fletcher et al., 2011). Finally, water contamination may also occur as PTA in fronds can be leached by rain and subsequently contaminate groundwater (Rasmussen et al., 2003a, 2005; Jensen et al., 2008). PTA has been generally found in all bracken parts analyzed so far (including fronds, rhizomes, roots and spores) (Rasmussen et al., 2013), with particularly high concentrations detected in the crosiers (i.e., young shoots) (Smith et al., 1994). Great variations in PTA content can also occur during the year, between bracken strains and even between different bracken stands located near each other (Gil da Costa et al., 2012). For example, in Venezuela, brackens were found to have low PTA contents at higher rather than at lower altitudes on mountain slopes, while an opposite pattern was observed in a Costa Rican study (Alonso-Amelot et al., 2000). PTA concentrations ranging between 0 and 1.03 mg g− 1 have been reported for Indian ferns (Pathania et al., 2012), up to 5.8 mg g− 1 have been found in British bracken fronds (Smith et al., 1994), and around 37 mg g−1 have been found in bracken samples from New Zealand (Engel et al., 2007). PTA is transferred to soil mainly by leaching from bracken litter and/or living plants, and it has been detected at concentrations around 8.5 μg g− 1 (Rasmussen et al., 2003a). Although several studies are already present in literature about eco-toxicological aspects related to PTA, the fate of PTA in the soil system is not completely understood. A rapid chemical hydrolysis is known to occur under either acidic (pH b 4) or alkaline (pH N 7) conditions (Agnew and Lauren, 1991), whereas PTA seems to be chemically stable at soil pH 4–7 (Rasmussen et al., 2005). Saito et al. (1989) reported that at pH 5.5 and 7.0, ca. 60% and 90% of the initial PTA concentration were still to be found after 7 days. In a recent study, Rasmussen et al. (2005) found that PTA was still present in several soils after 72 days, and concluded that pH, carbon and clay contents can affect its degradation patterns in soils. In the present work, we investigate the occurrence of PTA in both P. aquilinum fronds and corresponding topsoil samples in several sites from the South of Italy, showing different pedological and climatic conditions. The PTA occurrence in bracken ferns was also correlated with main soil physical and chemical features as well as climatic parameters, in order to better understand if and how they affect PTA production. To the best of our knowledge, this seems to be the first study carried out in the whole South of Europe.

2. Materials and methods 2.1. Study sites and sampling Five sites located in the South of Italy were selected in this study: 1) Canale del Conte, San Giovanni Rotondo (S1a,b); 2) Colle Calcarulo, Monte Sant'Angelo (S2); 3) Tenerano, Umbra forest, Vico del Gargano (S3a,b); 4) Piano Canale, Monte Sant'Angelo (S4); and 5) Magnano forest, San Severino Lucano (S5) (Fig. 1). The site selection was carried out taking into account differences in climatic and vegetational conditions (Table 1 and Supporting information Table S1). All selected sites were characterized by extensive (“wild”) livestock farming.

Soil and P. aquilinum samples were collected in October 2010; soils were sampled, where possible, at two depths, i.e., 0–10 and 10–20 cm. 2.2. Soil and fern sample characterization Soils were air-dried, 2 mm-sieved and stored at room temperature in the dark prior to further analyses. Soil analyses were carried out according to Sparks (1996) and references therein. Soil texture was determined by the pipette method, and the textural class assigned according to the United States Department of Agriculture (USDA) classification system. Soil pH was measured on a soil suspension using both water (1:2.5, w/v) and KCl (1:2.5, w/v) by a Philips pH-meter equipped with a Hanna Instruments HI 1230 probe, while the electrical conductivity (EC) was determined on a water extract (1:2, w/v) using a XS cond 510 conductimeter. The organic C (Corg) content was determined both in soil and in fern samples using the Walkley–Black and the Springer–Klee method, respectively, while total N (Ntot) was determined using the Kjeldahl method. Available P (Pava) was determined according to the Olsen method. Exchangeable bases (Ca, Mg, K, Na) were extracted from soil samples using a solution of BaCl2 (100 g l−1) and triethanolamine (22.5 ml l−1) at pH 8.2, and their concentrations measured by inductively couple plasma optical emission spectrometry (ICP-OES) (Thermo Electron ICAP 6000). The total concentration of Cd, Cr, Cu, Ni, Pb, and Zn was determined by wet digestion (1 ml H2O2, 1 ml HCl, and 5 ml HNO3) using the microwave-assisted digestion method in a microwave digestor system (Milestone, model Ethos 900). Samples were then cooled, diluted with ultrapure water in a 50-ml volumetric flask, filtered through Whatman 42 filters, and finally measured by ICP-OES. 2.3. Standards The amount of PTA in the samples was evaluated by converting the PTA into bromo-pterosine (Br-Pt) and adding appropriate amounts of Br-pterosine-d2 in the samples as an internal standard. Br-Pt and Br-pterosine-d2 were synthesized according to Miele et al. (2008). Mather solutions of Br-Pt and deuterated standard of 1 mg ml−1 were prepared in methanol and stored at −20 °C in order to keep them stable for months. Working standards were prepared in diethyl ether daily. 2.4. Soil incubation studies: extraction protocol and recovery In order to test the extraction protocol to be used and recovery obtained, incubation studies were carried out. Two top soils (0–20 cm depth), collected in Turi (L) and Valenzano (SCL) (Apulia region, South of Italy), were analyzed according to the methods reported in section “Soil and fern sample characterization”. In addition, cation exchange capacity (CEC) was determined by BaCl2 and triethanolamine solution. The soil water content was determined on three replicates of approximately 20 g of soil by using a pressure plate extractor system. Samples were left in cells under a pressure of −33 kPa until there was no further change in weight. The water content was then determined by the difference in weight from the oven-dried sample. Five-hundred grams of L and SCL soils were spiked with 12.5 g lyophilized bracken material to reach a PTA concentration of 0.02 mg kg− 1. Then, soils were thoroughly mixed with a spatula and 1 g soil aliquots were used to check the uniform distribution of PTA. Soils were transferred into incubation flasks which were lightly closed in order to maintain aerobic conditions. Degradation studies were performed in standardized laboratory conditions, in the dark at 20 ± 2 °C and at 40% water holding capacity (WHC). The incubation flasks were weighed, capped, and transferred to incubators at the required temperatures. Untreated soil samples were incubated under the same conditions. In order to study PTA decline with time, three replicates of each sample were taken on 7 dates (0, 1, 3, 5, 7, 10, 15 days after the incubation), following the OECD guidelines (OECD, 2002).

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

367

S1a S1b

S2

S3a

S1

S3

S4 S2 S3b

S5

S5 S4

Fig. 1. Location and view of the study sites (details are reported in Table 1 and in Supporting information Table S1). It is important to note the presence of cattle raised in the wild.

2.5. PTA extraction and conversion Fresh plant material was blended with deionized water for 5 min (sample/solution ratio, 1:6 w/w). Then, samples were transferred to the 50 ml high-clarity polypropylene conical centrifuge tubes where 5 g of soils was weighed and 10 ml of deionized water was

added before extraction. PTA was extracted from plant and soil samples by sonication at ambient temperature for 30 min. The extract was separated by ultracentrifugation at 14,000 rpm for 15 min at ambient temperature. PTA is converted into Br-Pt (Bonadies et al., 2004). Two grams of NaBr was added to a known amount of 2 ml bracken or soil extracts; 10 μl of a solution containing 10 ng of deuterated internal standard was

Table 1 Location, description and main climate parameters of the study sites. Phytoclimatic indices are reported in Supporting information Table S1. Site

Coordinates (UTM ED50)

Elevation (m a.s.l.)

Mean monthly Tmin (°C) (avg ± st.dev.)

Mean monthly Tmax (°C) (avg ± st.dev.)

Mean monthly T (°C) (avg ± st.dev.)

Annual precipitations (mm)

Site description

S1a,b

567640 E 4622035 N

700

10.0 ± 5.7

17.4 ± 7.2

13.7 ± 6.4

822.9

S2

575937 E 4617389 N

720

9.3 ± 5.8

15.9 ± 7.1

12.6 ± 6.5

787.2

S3a

582440 E 4630905 N

790

7.8 ± 5.5

15.4 ± 7.4

11.6 ± 6.5

958.5

S3b

582251 E 4630966 N 580343 E 4621267 N

790

7.8 ± 5.5

15.4 ± 7.4

11.6 ± 6.5

958.5

740

8.6 ± 5.8

15.3 ± 7.2

12.0 ± 6.5

847.7

665

8.5 ± 5.7

14.7 ± 5.9

11.6 ± 6.3

980.6

Population of Pteridium aquilinum subsp. aquilinum, with some Urtica dioica L. (b10%), dominating a small karstic valley (probably a previously cultivated field, now abandoned) surrounded by Quercus cerris L., Q. pubescens Willd. s.l., Ostrya carpinifolia Scop. and Fraxinus ornus L. Small population of Pteridium aquilinum subsp. aquilinum, with some Rubus ulmifolius Schott, naturalized in a short valley previously occupied by a vineyard now abandoned. Population of Pteridium aquilinum subsp. aquilinum with Rubus ulmifolius Schott present as groundcover of a plantation forest consisting of Pinus nigra J.F. Arnold with Acer pseudoplatanus L. Population of Pteridium aquilinum subsp. aquilinum as groundcover of a plantation forest consisting of Quercus cerris L. and Q. rubra L. Vast and dense population of Pteridium aquilinum subsp. aquilinum (other plant species are almost absent) occupying a valley surrounded mainly by Quercus ilex L. Population of Pteridium aquilinum subsp. aquilinum present as groundcover of a wood consisting manly of Quercus cerris L. and Fagus sylvatica L.

S4

S5

596271 E 4434100 N

368

Table 2 Main physical and chemical features of soil samples collected at different depths in the study sites. Values are reported as the average of three replicates.

g kg−1 g kg−1 g kg−1

dS m−1 g kg−1 g kg−1 g kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1

S1a (10–20 cm)

S1b (0–10 cm)

S1b (10–20 cm)

S2 (0–10 cm)

S2 (10–20 cm)

S3a (0–10 cm)

S3a (10–20 cm)

S3b (0–15 cm)

S4 (0–10 cm)

S4 (10–20 cm)

S5 (0–20 cm)

229 665 106 Silt loam 6.0 5.4 0.27 68.4 5.5 12.4 1.43 4952 1284 340 32 16 0.7 76.3 16.6 3.8 8 46 31 31 45 123

187 689 125 Silt loam 6.0 5.3 0.28 61.6 5.2 11.8 1.40 4117 1264 263 31 13 0.6 62.5 12.1 2.6 7 47 30 31 46 128

185 652 163 Silt loam 5.9 5.2 0.30 58.9 5.9 10.0 1.89 4374 764 203 28 18 0.6 51.3 8.0 3.0 8 53 36 33 52 123

168 674 158 Silt loam 6.1 5.4 0.18 58.1 5.7 10.2 1.88 4492 857 184 23 19 0.6 52.1 7.5 2.8 2 51 36 33 50 129

342 521 137 Silt loam 6.3 5.9 0.24 62.2 5.0 12.4 0.94 6805 1055 311 29 2 2.4 58.8 26.4 3.0 8 39 31 26 50 101

263 563 174 Silt loam 6.6 5.9 0.13 50.9 4.4 11.6 1.07 6300 898 281 32 2 2.1 57.2 18.8 1.5 8 39 31 25 51 109

589 355 57 Sandy loam 6.3 4.9 0.31 132.3 7.8 17.0 1.85 4633 342 431 65 10 1.2 99.6 33.5 5.9 6 28 35 32 47 139

603 347 50 Sandy loam 6.3 4.6 0.22 116.1 7.2 16.1 1.82 3801 353 365 36 12 1.2 89.9 30.2 4.2 6 29 33 28 49 140

639 297 64 Sandy loam 6.0 4.2 0.18 131 8.3 15.8 1.93 2101 235 260 41 18 0.7 137.3 25.0 5.2 6 31 26 26 53 138

115 616 269 Silt loam 6.3 6.2 0.12 28.1 2.6 10.8 0.77 5115 473 138 24 4 1.1 38.2 27.0 1.1 9 61 26 43 36 125

94 603 303 Silt clay loam 6.6 6.0 0.12 19.4 2.1 9.2 0.72 4912 346 100 13 2 1.2 29.8 20.2 0.9 65 27 49 49 39 119

537 355 108 Sandy loam 6.1 4.7 0.15 75.6 4.4 17.2 0.72 1511 181 811 21 7 2.6 136.5 87.6 3.3 13 1069 46 518 30 223

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

Sand Silt Clay Texture pHH2O pHKCl EC Corg Ntot C/N Ptot Caexc Kexc Mgexc Naexc Pava Cuava Feava Mnava Znava Cd Cr Cu Ni Pb Zn

S1a (0–10 cm)

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372 Table 3 Main physical and chemical properties of the two soils (L and SCL) used for the incubation study. Parameters

L

SCL

pHH2O pHKCl EC Clay Silt Sand Texture −33 kPa water content CEC Corg Ntot C/N

8.16 7.7 0.13 244 335 421 Loam 19 10.6 7.1 0.7 9.5

7.7 6.8 0.24 363 524 113 Silt clay loam 29 21.1 28.9 3.1 9.6

dS m−1 g kg−1 g kg−1 g kg−1 gwater/100 gsoil cmol(+) kg−1 g kg−1 g kg−1

added and alkalinized with 150 μl of 1 M NaOH solution to pH 12. After heating for 1 h at 45 °C, the solution was acidified to pH 2 by 2 N H2SO4 and extracted with 2 ml ethyl ether by sonication for 3 min. Aliquots of 100 μl of the extracts were concentrated to 50 μl. 2.6. GC–MS conditions and analysis The DSQ™ II, configured with the proven Thermo Scientific TRACE GC Ultra gas chromatograph and single quadrupole GC–MS, was used. The capillary column was TR CP-Wax 52 CB fused silica WCOT, 10 m × 0.1 mm, df = 0.2 μm. The GC–MS transfer line temperature was maintained at 200 °C. The following temperature program was employed: initial temperature 100 °C held for 1 min; ramped at 20 °C min−1 up to 245 °C, held for 7 min. Injector and ion source temperatures were 240 °C. Helium was the carrier gas at 0.9 ml min− 1; the sample (1 μl) was injected in the splitless mode. The ionization was used in the electron ionization (EI) mode with an ionization in the selected ion monitoring (SIM) mode of MS. Data were collected by evaluating areas of chromatograms (obtained in SIM mode for three ions) of analyte (Br-Pt) and an added internal standard (Br-pterosined2). Ions monitored were 187, 201, and 280 m/z for the analyte, and 189, 203, and 283 m/z for the deuterated reference. 2.7. Linearity, limit of detection, limit of quantification and recovery Internal standard calibration was performed to obtain the calibration curve by measuring the peak areas of the analyte relative to that of the internal standard. A good linearity was achieved up to 0.0001– 1 mg l−1 (y =0.2631x + 0.0201, R2 = 0.9973). The limit of detection (LOD) and the limit of quantification (LOQ) were calculated following the directives of the IUPAC and the American Chemical Society's Committee on Environmental Analytical Chemistry. The LOD and LOQ values obtained were 0.1 and 0.3 μg l−1, respectively. The concentrations were calculated by the standard curve and expressed as sample's dry weight. Humidity was determined by drying the samples in the oven at 105 °C until constant weight. Recovery yields and accuracy were determined for each type of soil used in the incubation study (L and SCL) as well as for plant material. Untreated samples from all soils without addition of lyophilized bracken were analyzed at the same time. Mean recovery for the SCL soil was 87.9%, with a coefficient of variation of 1.6%, while mean recovery for the L soil was 88.7%, with a coefficient of variation of 0.8%. LOD and LOQ were 0.01 and 0.015 mg kg−1, respectively, for all soils. The applied chromatographic parameters produced a measured separation between Br-Pt (tr = 8.63 min) and internal standard (Br-pterosine-d2) (tr = 8.69 min) in all soil samples (Supporting information Fig. S1). For

369

plant material, mean recovery was 94%, whereas LOD and LOQ were 0.005 and 0.01 mg kg−1, respectively. 2.8. Curve fitting and statistics The actual degradation rate was calculated for each sampling time using the ModelMaker software Version 4.0; graphical compartmental and system dynamic modeling software package provides the best fit line for the experimental data (FamilyGenetix, 2000). Two kinetic models were used to fit the degradation data of PTA: (a) a Single FirstOrder (SFO) equation (exponential kinetic model); and (b) a FirstOrder Multi-Compartment (FOMC) model (Gustafson and Holden, 1990). Model parameters were optimized according to recommendations given by FOCUS (2006) and using the least-squares method. A fit that results in an error level b15% is considered acceptable although this is not an absolute cut-off criterion and a visual assessment must be made. All data were analyzed using the General Linear Models Procedure (SAS Institute Inc., 2001). Mean separation among lines was accomplished using the Duncan multiple range test at p b 0.05. A stepwise regression method was used to define the most important soil and climatic parameters which were potentially effective on PTA production. Clearly, in building regressions, all parameters determined were considered, but those that did not give relevant results, and/or that were highly correlated with other parameters already considered in the model, were systematically excluded. Statistica Version 9.1 software (StatSoft Inc., 2010) was used for building stepwise regressions. 3. Results and discussion 3.1. Pedo-climatic differences in site features Besides genetic heritage, external growth factors like climate and soil properties also affect PTA content in bracken ferns (Smith et al., 1994). Consequently, the present study has been carried out selecting sites characterized by different phyto-climatic features (Table 1 and Supporting information Table S1). In particular, they show an elevation ranging between 665 and 790 m a.s.l., a mean monthly temperature between 11.6 ± 6.5 and 13.7 ± 6.4 °C, and an annual precipitation between 787 and 981 mm (Table 1). Studied sites are also characterized by soils having extremely different physical and chemical features (Table 2). In detail, the texture ranged from silt loam to sandy loam to silt clay loam, with a clay content in the top layer of 6 to 27%. The organic C content ranged from 19 to 132 g kg−1, whereas total N ranged from 2.1 to 8.3 g kg−1. Because the highest values of both organic C and total N were found in soils showing higher sand content (around 60%; S3a, S3b), it is likely that most of the organic matter is in the form of particulate organic material (Kooch et al., 2012). All soils were slightly acidic, with a pH ranging from 5.9 to 6.6. Among micro- and macro-nutrients, phosphorous seems to be the main limiting factor (Pava = 2–19 mg kg−1). Soil samples from one site (S5) showed high concentrations of heavy metals (Cr, Ni and Zn). 3.2. Incubation study The soils used for the incubation study have been selected as they are quite diffused in the whole Apulia region; their main physical and chemical properties are reported in Table 3. Degradation pathways of PTA in L and SCL soils, determined according to SFO and FOMC models, are presented in Fig. 2. Obtained DT50 (Disappearance Time 50) and DT90 (Disappearance Time 90) values, i.e., the time within which the percentage of the PTA is reduced by 50 and 90%, respectively, as well as corresponding χ2 error values, are reported in Table 4. DT50 values determined in the L soil were quite similar using both SFO (7.7 days) and FOMC (8.2 days) models, whereas a higher variability was observed for the

370

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

0,025

0,025

SFO

measured

0,020

Concentration (mg kg-1)

Concentration (mg kg-1)

measured

L

0,015

0,010

0,005

L 0,015

0,010

0,005

0,000

0,000 0

3

6

9

12

0

15

3

6

9

12

15

Time (days)

Time (days) 0,025

0,025

measured

SFO

Concentration (mg kg-1)

measured

Concentration (mg kg-1)

FOMC

0,020

0,020

SCL 0,015

0,010

0,005

FOMC

0,020

SCL 0,015

0,010

0,005

0,000

0,000 0

3

6

9

12

15

0

3

6

9

12

15

Time (days)

Time (days)

Fig. 2. Experimental data of PTA degradation in L and SCL soils, according to the Single First-Order (SFO; ○) and First-Order Multi-Compartment (FOMC; Δ) models.

SCL soil (11.4 vs. 7.5 days, respectively). A similar behavior was observed for DT90, whose values ranged between 25 and 51 days. Anyway, because 90% of PTA degradation was not achieved during the experimental period, DT90 values are presented but not discussed. Finally, the SFO model seems to fit the data slightly more closely than the biphasic FOMC one for the SCL soil (χ2 values, 5.8 vs. 11.1, respectively), while no significant differences were observed for the L soil (χ2 values, 7.9 vs. 8.0, respectively) (Table 4). Observed differences were significant at the p b 0.05 level. Degradation rates (K) were 0.061 to 0.090 × 10−2 mg kg−1 day−1 for the SCL and L soils, respectively. Consequently, it seems that the degradation rate is higher in the L soil, i.e. those characterized by a lower organic C content (0.7 vs. 2.9%), a higher sand percentage (42.1 vs. 11.3%) and pH sub-alkaline (8.6 vs. 7.7) (Table 2). These results are in agreement with Rasmussen et al. (2005) who reported that the Freundlich affinity coefficient of PTA increased linearly and positively with clay and organic matter contents. Thus, the degradation rate of PTA increases in the soils where this molecule is less adsorbed on the surfaces of organic and inorganic colloids, being, in this case, more easily available to microbial degradation. In fact, as reported by Engel et al. (2007), microorganisms might play a predominant role for a rapid

Table 4 Single First-Order (SFO) and First-Order Multi-Compartment (FOMC) DT50 and DT90 values for PTA (data are expressed in days). Percentage error levels (χ2) a are also reported. Soils

L SCL

SFO

FOMC

DT50

DT90

χ2 (%)

Kb

DT50

DT90

χ2 (%)

7.7 11.4

25.6 37.7

7.9 5.8

0.090 0.061

8.2 7.5

27.2 50.4

8.0 11.1

a The smaller the error, the better is the fit: errors b15% are considered acceptable (FOCUS, 2006). b K = factor of disappearance (10−2 mg kg−1 day−1).

PTA degradation, that occurs when the compound is in the soil–water phase. 3.3. PTA occurrence in the studied sites PTA concentration in P. aquilinum samples showed a wide variation, ranging from 2 mg kg−1 (S2) to 780 mg kg−1 (S1b) (Table 5). Besides living bracken ferns, and only for S5, dead plants and litter samples were also analyzed, and PTA concentrations resulted in 6 mg kg− 1 and bLOQ, respectively. The stepwise regression method underlined how PTA content in P. aquilinum samples is affected by soil features rather than by climatic conditions characterizing the studied sites (Table 6a). Although regression models might generally show some limitation, it is quite clear that the PTA content in P. aquilinum plants is significantly and positively affected mainly by the availability of P in the soil (Table 6a). Also the availability of other micronutrients (i.e., Mg, Fe, Cu) seems to influence the occurrence of PTA, although to a different extent. Finally, a negative correlation was found between PTA occurrence in P. aquilinum and both organic C content and pH in soil, probably due to their ability to affect the biogeochemistry and bioavailability of nutrients in the plant–soil system. The stepwise regression analysis also revealed a positive and significant correlation between PTA and total P concentrations in P. aquilinum plants (Table 6b). Again, it seems that bracken ferns showing higher concentrations of this element are those producing more PTA. On the opposite, a negative correlation was found between PTA and Ni concentrations in P. aquilinum plants. In contrast with incubation studies, the PTA concentration in soil samples was always bLOQ (0.015 mg kg− 1), independent of: i) the PTA concentration in the corresponding Pteridium samples (2 to 780 mg kg−1), ii) the soil organic C content (2 to 13%), iii) the soil pH (5.9 to 6.6), iv) the soil texture, v) the sampling depth (0–10 cm; 10–20 cm), and vi) precipitations (780 to 960 mm a− 1). Although

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

371

Table 5 Elemental content and PTA concentration in Pteridium aquilinum samples collected at each of the study sites. Values are reported as the average of three replicates.

Corg Ntot C/N Ptot Cd Cr Cu Ni Pb Zn PTA a b

g kg−1 g kg−1 g kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1 mg kg−1

S1a

S1b

S2

S3a

S3b

S4

S5

S5a

S5b

470.9 2.00 235.4 2.17 b0.1 28 18 16 2 39 505

469.1 1.81 259.2 1.98 b0.1 b0.1 20 8 16 34 780

478.4 0.70 683.4 0.77 b0.1 104 10 19 b0.1 59 2

472.9 1.09 433.9 1.18 b0.1 15 11 22 b0.1 40 32

464.7 1.29 360.2 1.40 b0.1 38 20 33 2 46 8

469.3 1.50 312.9 1.63 1 14 16 49 b0.1 26 12

469.6 2.32 202.4 2.55 b0.1 25 21 28 b0.1 36 390

473.1 0.68 695.7 0.75 1 39 10 18 4 38 6

385.8 0.91 424.0 0.97 4 1137 62 531 41 79 b0.015

Dead tissues of Pteridium aquilinum (dry and brown in color). Litter sample mainly consisting of Pteridium aquilinum, Quercus cerris and Fagus sylvatica debris at different stage of decomposition.

the microbial activity was not investigated, the absence of PTA in soils from the studied sites seems to suggest its degradation by the indigenous microbial community. These data are in general disagreement with those reported in literature and with incubation studies, both underlining a certain affinity of PTA for organic colloids and clay/silt particles. The presence of organic matter mainly in particulate form, having generally a lower specific surface, could also explain the absence of PTA in soils characterized by bracken ferns showing high concentrations of this molecule. The chemical hydrolysis of PTA, suggested to occur under either acidic (pH b 4) or alkaline (pH N 7) conditions (Agnew and Lauren, 1991), seems to be unlikely. Some of the results presented in this work are in contrast with data previously reported in literature. At the same time, it is important to underscore that most of the published studies on PTA fate in the soil–fern system have been carried out in Northern Europe, i.e., in soils showing lower pH and higher organic C content (e.g., Rasmussen et al., 2003b), and in completely different climatic conditions.

concentration in soil samples from the studied sites was always undetectable (b0.015 mg kg−1), independent of the PTA concentration in the corresponding Pteridium samples (2–780 mg kg−1) and all variables under investigation, including soil organic matter content, pH, texture, depth of sampling, and precipitations, although incubation studies underlined a certain affinity of PTA for both organic colloids and clay/ silt particles. This seems to suggest the degradation of PTA by indigenous microbial community characterizing the P. aquilinum–soil system. Anyway, because there could be many other factors affecting the production and fate of PTA, and taking into account the range of variability of pedo-climatic parameters potentially involved, possible environmental implications need to be investigated at a local/regional scale. Furthermore, studies on physiological and biochemical aspects regarding PTA production in ferns would be beneficial in order to better understand and predict the occurrence of this carcinogen in the environment. Acknowledgment

4. Conclusions The occurrence of PTA in the P. aquilinum–soil system has been investigated in different pedo-climatic areas from the South of Italy. The PTA content was determined in both soil and fern samples by GC–MS, and data correlated with the physical and chemical features of soils, as well as climatic parameters of studied sites, in order to better understand their influence on PTA occurrence. Stepwise regression analysis showed that PTA concentration in P. aquilinum ferns is significantly and positively affected by nutrient availability (mainly P) in soil rather than climatic conditions. Also pH and organic C content of soil seem to influence PTA concentration in P. aquilinum ferns. Furthermore, PTA concentration was higher in plants showing a higher P content, whereas a negative correlation was found with Ni. At the same time, PTA

The present study was financed by the Ministero Italiano delle Politiche Agricole, Alimentari e Forestali (MIPAAF) (Project AZBSASP — “Agrozootecnia biologica: considerazioni in termini di sicurezza alimentare e problemi di salute pubblica”). C.Z. would like to thank Prof. Vincenzo Lattanzio, University of Foggia, responsible for the UniFG Research Unit, and Luigi Forte, University of Bari “Aldo Moro”, for the help during the field work and his useful comments on botanical and phyto-climatic aspects. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.07.046.

Table 6 Stepwise regression analysis of (a) PTA concentration in Pteridium aquilinum plants (PTApl, dependent variable) and pedo-climatic features (independent variables) (n = 7); (b) PTA concentration in Pteridium aquilinum plants (PTApl, dependent variable) and elemental content in Pteridium aquilinum plants (independent variables) (n = 8; litter sample has been excluded). Regression models are significant at p b0.05 level. Step

Regression equation

a) 1 2 3 4 5 6

PTApl PTApl PTApl PTApl PTApl PTApl

b) 1 2

PTApl = 0.779 Ptot PTApl = 1.67 Ptot − 1.1 Ni

= = = = = =

0.762 Pava 1.51 Pava − 0.84 Corg 1.72 Pava − 1.7 Corg + 0.859 Mgexc 1.8 Pava − 1.0 Corg + 1.41 Mgexc − 1.3 Feava 1.86 Pava − 1.1 Corg + 1.26 Mgexc − 1.3 Feava + 0.166 Cuava 2.12 Pava − 1.0 Corg + 1.09 Mgexc − 1.4 Feava + 0.572 Cuava − 0.43 pH

Adj. R2

p

Standard error of the estimation

0.5112 0.6195 0.9102 0.9805 0.9828 0.9999

0.0279 0.0385 0.0049 0.0019 0.0122 0.0012

266.45 235.10 114.23 53.21 49.95 0.67

0.5507 0.9031

0.0133 0.0004

238.96 111.00

372

C. Zaccone et al. / Science of the Total Environment 496 (2014) 365–372

References Agnew MP, Lauren DR. Determination of ptaquiloside in Bracken fern (Pteridium esculentum). J Chromatogr 1991;538:462–8. Alonso-Amelot ME, Castillo U, Smith BL, Lauren DR. Bracken ptaquiloside in milk. Nature 1996;382:587–8. Alonso-Amelot ME, Castillo U, Smith BL, Lauren DR. Excretion, through milk, of Ptaquiloside in bracken-fed cows. A quantitative assessment. Lait 1998;78:413–23. Alonso-Amelot ME, Calcagno MP, Pérez-Inojosa M. Growth and selective defensive potential in relation to altitude in neotropical Pteridium aquilinum var. caudatum. In: Taylor JA, Smith RT, editors. Bracken fern: toxicity, biology and controlProceedings of the International Bracken Group Conference, Aberystwyth, Wales; 2000. p. 43–7. Bonadies F, Borzacchiello G, Dezzi S, Nicoletti R, Roperto S. Mass spectrometric analysis of ptaquiloside, the toxic sesquiterpene from bracken fern. Rapid Commun Mass Spectrom 2004;18:825–8. Bonadies F, Berardi G, Nicoletti R, Romolo FS, De Giovanni F, Marabelli R, et al. A new, very sensitive method of assessment of ptaquiloside, the major bracken carcinogen in the milk of farm animals. Food Chem 2011;124:660–5. Campo MS. Bovine papillomavirus and cancer. Vet J 1997;154:175–88. Conti F, Abbate G, Alessandrini A, Blasi CAn annotated checklist of the Italian vascular flora. Roma: Palombi & Partner; 2005. 420 pp. Engel P, Brandt KK, Rasmussen LH, Ovesen RG, Sørensen J. Microbial degradation and impact of Bracken toxin ptaquiloside on microbial communities in soil. Chemosphere 2007;67:202–9. FamilyGenetixModelMaker 4.0, FamilyGenetix Ltf: Beaconsfield. UK: Buckinghanshire; 2000. Fletcher MT, Reichmann KG, Brock IJ, McKenzie RA, Blaney BJ. Residue potential of norsesquiterpene glycosides in tissues of cattle fed Austral bracken (Pteridiuim esculentum). J Agric Food Chem 2011;59:8518–23. FOCUS. Guidance document on estimation persistence and degradation kinetics for environmental fate studies on pesticides in EU registration. Report of the FOCUS work group on Degradation Kinetics, EC document reference Sanco/10058/2005, version 1.0 (final draft awaiting approval). Brussels: European Commission; 2006. Gil da Costa RM, Bastos MMSM, Oliveira PA, Lopes C. Bracken-associated human and animal health hazards: chemical, biological and pathological evidence. J Hazard Mater 2012;203–204:1–12. Gustafson DI, Holden LR. Non linear pesticide dissipation in soil: a new model based on spatial variability. Environ Sci Technol 1990;24:1032–8. Hirono I. Carcinogenic principles isolated from bracken fern. CRC Crit Rev Toxicol 1987; 17:1–22. Jensen PH, Jacobsen OS, Hansen HCB, Juhler RK. Quantification of ptaquiloside and pterosin B in soil and groundwater using liquid chromatography–tandem mass spectrometry (LC–MS/MS). J Agric Food Chem 2008;56:9848–54. Kooch Y, Hosseini SM, Zaccone C, Jalilvand H, Hojjati SM. Soil organic carbon sequestration as affected by afforestation: the Darab Kola forest (North of Iran) case study. J Environ Monit 2012;14:2438–46.

Miele G, Costi R, Bonadies F, Nicoletti R. Simple synthesis of deuterated pterosines. Synth. Commun. 2008;38:1553–9. OECD. Guidelines for the testing of chemicals test no. 307: aerobic and anaerobic transformation in soils. Paris, France: Organisation for Economic Co-operation and Development; 2002. Ojika M, Wakamatsu K, Niwa H, Yamada K. Ptaquiloside, a potent carcinogen isolated from bracken fern Pteridium aquilinum var. latiusculum: structure elucidation based on chemical and spectral evidence, and reactions with amino acids, nucleosides, and nucleotides. Tetrahedron 1987;43:5261–74. Pathania S, Kumar P, Singh S, Khatoon S, Rawat AKS, Punetha N, et al. Detection of ptaquiloside and quercetin in certain Indian ferns. Curr Sci 2012;102:1683–91. Pignatti SFlora d'Italia, vol. 3. Bologna: Edagricole; 1982. Rasmussen LH, Jensen LS, Hansen HCB. Distribution of the carcinogenic terpene ptaquiloside in bracken fronds, rhizomes (Pteridium aquilinum), and litter in Denmark. J Chem Ecol 2003a;29:771–8. Rasmussen LH, Kroghsbo S, Frisvad JC, Hansen HCB. Occurrence of the carcinogenic Bracken constituent ptaquiloside in fronds, topsoils and organic soil layers in Denmark. Chemosphere 2003b;51:117–27. Rasmussen LH, Lauren DR, Hansen HCB. Sorption, degradation and mobility of ptaquiloside, a carcinogenic bracken (Pteridium sp.) constituent, in the soil environment. Chemosphere 2005;58:823–35. Rasmussen LH, Schmidt B, Sheffield E. Ptaquiloside in bracken spores from Britain. Chemosphere 2013;90:2539–41. Saito K, Nagao T, Matoba M, Koyama K, Natori S, Murakami T, et al. Chemical assay of ptaquiloside, the carcinogen of Pteridium aquilinum, and the distribution of related compounds in the Pteridaceae. Phytochemistry 1989;28:1605–11. SAS Institute, Inc. SAS/STAT® 9.22 User’s Guide. NC: Cary; 2010. Smith BL, Seawright AA. Bracken fern (Pteridium spp.) carcinogenicity and human health — a brief review. Nat Toxins 1995;3:1–5. Smith BL, Seawright AA, Ng JC, Hertle AT, Thomson JA, Bostock PD. Concentration of ptaquiloside, a major carcinogen in Bracken fern (Pteridium spp.), from eastern Australia and from a cultivated worldwide collection held in Sydney, Australia. Nat Toxins 1994;2:347–53. Sparks DL. Methods of soil analysis. Part 3: chemical methods. In: Sparks DL, editor. SSSA book series, No. 5. Madison, Wisconsin, USA: SSSA and ASA; 1996. p. 1390. StatSoft, Inc. Statistica (data analysis software system), version 9.1.; 2010. www.statsoft. com. Taylor JA. The bracken problem: a global perspective. In: Thomson JA, Smith RT, editors. Bracken biology and management, No. 40. Sydney, Australia: Australian Institute of Agricultural Science; 1990. p. 3–19. Yamada K, Ojika M, Kigoshi H. Isolation, chemistry, and biochemistry of ptaquiloside, a bracken carcinogen. Angewandte Chemie — International Edition, 37; 1998. p. 1818–26.