ARTICLE pubs.acs.org/est
Occurrence and Mass Balance of Isoflavones on an Experimental Grassland Field Corinne C. Hoerger,†,‡ Felix E. Wettstein,† Hans J€org Bachmann,† Konrad Hungerb€uhler,‡ and Thomas D. Bucheli*,† † ‡
Agroscope Reckenholz-T€anikon Research Station ART, CH-8046 Z€urich, Switzerland Institute of Chemical and Bioengineering, Swiss Federal Institute of Technology, CH-8093 Z€urich, Switzerland
bS Supporting Information ABSTRACT: Isoflavones and coumestrol (COU) are estrogenic compounds that are naturally produced by plants (e.g., red clover, soybeans). Although these compounds have been extensively studied in food and feed, only little is known about their environmental fate. Therefore, we investigated the occurrence of isoflavones (formononetin, daidzein, equol, biochanin A, and genistein) and COU over 3.5 years in red clover, manure, and soil of a grassland field with and without manure application, as well as their emission via drainage water. Isoflavones were regularly quantified in plant (e15 106 ng/gdry weight (dw)), manure (e230 103 ng/gdw), soil (e3.4 103 ng/gdw), and drainage water samples (e3.6 103 ng/L). In contrast, COU was observed only in manure and soil. Cumulative isoflavone loads emitted via drainage water were around 0.2 103 kg/ha/y, which is very little compared to the amounts present in red clover (105220 kg/ha/y), manure (0.51.0 kg/ha/y), and soil (0.15.1 kg/ha/y). Under good agricultural practice, no additional emission of isoflavones into drainage water was observed after manure application. With calculated 17β-estradiol equivalents up to 0.46 ng/L in drainage water, isoflavones can constitute a dominant and ecotoxicological relevant portion of the total estrogenicity in small rural river catchments.
’ INTRODUCTION The occurrence of endocrine disruptors such as steroid hormones and anthropogenic micropollutants in the aqueous environment has been extensively studied.1 In contrast, naturally produced phytoestrogens have rarely been investigated,25 although their estrogenic potencies are similar to, or even higher than, those of man-made micropollutants.6 Isoflavones and coumestrol (COU) are present in high concentrations (mg/gdry weight (dw)) in legumes, such as clover (Trifolium ssp.), soybeans (Glycine max), and alfalfa (Medicago sativa). Red clover (Trifolium pratense) is a common pasture and forage crop, which produces mainly formononetin (FOR) and biochanin A (BIO) at a concentration of up to 15 106 ng/gdw.7 Conversely, genistein (GEN) and daidzein (DAI) prevail in soybeans.8 FOR and BIO are demethylated in the intestinal track of humans and husbandry animals to DAI and GEN, respectively.9,10 Equol (formally an isoflavane, but attributed to the isoflavones in this study for reasons of simplicity) is not produced in planta,9,10 but is a degradation product of DAI. COU is mainly produced in alfalfa,8 but also at lower concentrations in clover and soybeans.11 The physical chemical properties and estrogenic activities of these compounds are given in Supporting Information (SI) Tables S1 and S2. The content of isoflavones and COU in feed and food has been monitored repeatedly (reviewed in refs 12 and 13), and their uptake by husbandry animals10 and humans (reviewed in ref 3) is well described. Less is known at present about their environmental occurrence and exposure. The presumed main pathways of isoflavones into the aqueous environment are briefly summarized below and include r 2011 American Chemical Society
(1) human excretion via wastewater treatment plants (WWTPs), (2) excretion from grazing livestock and manure application, and (3) runoff and drainage from fields cultivated with forage and grain legumes. The presence of isoflavones in WWTP influents and effluents was investigated by several authors (reviewed in ref 3). Influent concentrations of FOR, DAI, equol, BIO, and GEN were often in the order of several hundred ng/L,4,1416 whereas respective effluent concentrations were generally considerably lower (from not detected to a few hundred ng/L),4,1416 with variable removal rates of 50100%. So far, only a few studies reported isoflavones and COU in liquid or solid manures, in which equol prevailed with concentrations in the μg/gdw or mg/L range.1719 Although it constitutes only a minor fraction of the daily intake (approximately 0.050.1 kg/ ruminant/day),10,20 the excreted amount of isoflavones in the order of 0.5 103 to 2.0 103 kg/day19 is substantially greater than that of natural steroids (0.3 106 to 1 106 kg/ruminant/ day).21 In Switzerland, cattle are mainly fed with grass from permanent and temporary grassland, and their manure is reapplied to these areas to close nutrient cycles. Correspondingly, grassland areas and livestock units in a given catchment are highly correlated Received: February 18, 2011 Accepted: June 28, 2011 Revised: June 16, 2011 Published: June 28, 2011 6752
dx.doi.org/10.1021/es200567b | Environ. Sci. Technol. 2011, 45, 6752–6760
Environmental Science & Technology (total ruminants vs grassland: R2 > 0.94, using Cantonal data from the Swiss Federal Statistical Office22). Assuming 1 103 kg of isoflavones in some 5060 L of manure excreted per cow and day17,19 and an average manure application of 50100 m3/ha/y, some 11.6 kg/ha/y of isoflavones may be entering an agricultural soil. This amount is more than 60 times less than the one produced in clover pastures,23 but may now be present in a much more mobile, bioavailable, and estrogenically more potent form, because these compounds are no longer stored in intact plants as glucosides,24 but probably prevail in dissolved liquid form. Hence, the risk of isoflavone input into natural waters via manure application onto agricultural fields has yet to be comprehensively evaluated. The isoflavone emission from grassland (including red clover) and cropland (e.g., soybeans) by runoff and drainage water has not yet been investigated in great detail. Erbs et al.25 reported isoflavone concentrations between 4 and 157 ng/L in drainage water from a red clover pasture field. This preliminary study did not, however, distinguish between direct isoflavone emission from plant, e.g., after cutting, or during decomposition of material, and runoff from applied manure. The above-mentioned input sources result in the exposure of surface waters to isoflavones. Their occurrence, mostly in the low ng/L-concentration range, has been reported in several rivers.4,5,14,16 The study by Kawanishi et al.,26 with concentration of DAI and GEN up to 143 103 ng/L, is an exception. Only a few publications (e.g., refs 5 and 16) elucidated possible sources of isoflavones. In an earlier publication,16 we presented some circumstantial evidence that isoflavones were emitted in Switzerland mainly via runoff from grassland. To confirm this hypothesis, in the present work an experimental field was cultivated with a grasslandred clover mixture for 3.5 years. During this period, isoflavones and COU were regularly investigated in red clover, liquid manure (urine and feces from cattle/pig 3:1), soil, and drainage water. Additionally, to better distinguish between primary (plants) and secondary (manure applications) emissions, manure was not applied on the field until 2 years after the beginning of the study. The drainage water data of the field study were compared with those of percolating waters from a nearby lysimeter study. Finally, the ecotoxicological relevance of the presence of these compounds in drainage and surface waters is briefly evaluated. To our knowledge, this is the first time the environmental exposure to isoflavones and COU has been examined with such an approach.
’ MATERIALS AND METHODS Field Site Description, Instrumentation, and Cultivation. The study site has been previously described in refs 27 and 28. In brief, it is located near our research station (Agroscope ART), in the North of Zurich, Switzerland (47°250 7400 N, 8°300 8500 E). The field comprised 0.2 ha and had a gentle slope of 12°. The topsoil was a gleyic cambisol29 with 30% sand (2 mm63 μm), 39% silt (67 cm) is removed by harvest for husbandry animal feedstuff. Isoflavones in Manure. All isoflavones were detected in three manure samples (Table 1, Figures 1B, S3BS6B). Equol exhibited by far the highest concentrations (160 103230 103 ng/gdw), followed by FOR, DAI, BIO, and GEN. These results fit well into a larger data set with different types of manure from various Swiss dairy farms,18 and correspond largely with earlier data from literature.17,19 It is known that FOR and DAI are mainly converted to equol in the intestinal track of husbandry animals.10,19 Accordingly, the isoflavone pattern of manure (Figure 2B) was completely different from that of red clover (Figure 2A). Equol accounted for 97% of all isoflavones. Based on the concentration range in Table 1, an annual manure application of 6080 m3/ha, and a dry matter content of 5% for manure, the total isoflavone input from manure on our field was between 0.5 and 1 kg/ha/y. This amount corresponds to less than 1% of the total amount produced in red clover (see above). Still, the speciation of equol may be more environmentally mobile in manure (dissolved liquid form) than that of isoflavones in intact plants (stored as glucosides).24
ARTICLE
Isoflavones in Soil. Before manure application, FOR and BIO were detected in all topsoil (010 cm) samples (n = 17) (Table 1). The FOR concentrations were highest (e3.4 103 ng/gdw), followed by those of BIO, GEN, DAI, and equol (Table 1, Figures 1C, S3CS6C; red triangles). Interestingly, equol was detected in soil, although no manure application took place (Figure S4C). Until now, transformation of FOR and DAI to equol was only reported in the intestinal track of husbandry animals or humans under anaerobic conditions.9,10 Here, the production of equol in soil was observed under presumably aerobic conditions. The isoflavone pattern (Figure 2C) was similar to that of red clover, supporting its role as a major source. The FOR contribution was even higher in topsoil than in red clover (75% and 60%, respectively), while in contrast BIO decreased (14% and 34%, respectively). This suggests that BIO dissipated faster than FOR, which is in accordance with its halflife being shortest (26 d, Table S1; refs 25 and 32). The parent compounds accounted for 89% of the total isoflavone content, which is less than in plants. In subsoil (1020 cm), FOR, DAI, and BIO were quantified in all samples before manure application (n = 15) (Table 1). Generally, the concentrations in subsoil were lower than in topsoil, with FOR still prevailing (e250 ng/gdw) (Table 1, Figures 1C, S3CS6C; dark red triangles). This observation is in accordance with Hoerger et al.18 The isoflavone pattern (Figure 2E) was the same as in topsoil (Figure 2C), but the contribution of the parent compounds was reduced to 71%. Overall, the increase of metabolite contributions from red clover to top- and subsoil suggests that dissipation (including metabolization, transformation, and sequestration) of the parent compounds took place as soon as they were emitted into the environment. In fact, half-lives of isoflavones in soils were reported to range from about 2 to 20 days.32,33 After manure application, FOR, DAI, and BIO were found in all topsoil (010 cm) samples (n = 16; Table 1). Again, the FOR concentrations were highest (e2.2 103 ng/gdw) (Table 1, Figures 1C, S3CS6C; red triangles). The concentrations of equol (t test, p-value = 0.003) and DAI (t test, p-value
