|
|
||||||||
Southern Crop Protection and Food Research Centre, Agriculture and Agri-Food Canada, 1391 Sandford Street, London, ON, Canada N5V 4T3
* Corresponding author (toppe{at}agr.gc.ca)
Received for publication August 25, 2004.
| ABSTRACT |
|---|
|
|
|---|
-androstane-3,17-dione, and 1,4-androstadiene-3,17-dione), which accounted for 48.7, 23.7, and 9.6% of the remaining 14C, respectively. Periodic analysis of soil incubated at 23, 12, and 4°C showed that the rates of testosterone dissipation and metabolite appearance and subsequent dissipation were temperature dependent with rates decreasing with decreasing temperature. In sterilized loam, 4-androstene-3,17-dione was the only metabolite detected. We conclude that testosterone is rapidly and thoroughly biodegraded in agricultural soils under a range of conditions typical of a temperate growing season and thus is unlikely to pose a long-term risk to adjacent aquatic environments.
Abbreviations: DT50, time to dissipate 50% of material HPLC, high performance liquid chromatography LSC, liquid scintillation counting RD, radioactivity detection
| INTRODUCTION |
|---|
|
|
|---|
The most significant environmental sources of natural and synthetic androgenic substances appear to include pulp and paper mill effluent (reviewed in Thomas et al., 2002), sewage treatment plant effluent (reviewed in Kirk et al., 2002), and intensive livestock operations (reviewed in Lintelmann et al., 2003). Although the identity of the androgenic substance(s) in pulp and paper mill effluent remains unknown (Durhan et al., 2002), androgenic activities determined in sewage treatment plant effluent and livestock operations have been associated primarily with naturally produced gonadal steroids and their metabolites (Thomas et al., 2002; reviewed in Lange et al., 2002). For example, greater than 99% of the total androgenic activity determined for sewage treatment plant effluent in the United Kingdom was related to naturally produced steroids or steroid metabolites (Thomas et al., 2002).
Some farm animals excrete naturally produced androgens in the range of milligram quantities per animal per day (Lange et al., 2002) and recombinant yeast assays have shown androgen receptor gene transcription activities in livestock manures (Lorenzen et al., 2004). However, in addition to naturally produced hormones, many farm animals are also treated with exogenous sources of hormones or synthetic substances to synchronize reproductive cycles and improve feed efficiency. In particular, the widely used growth promoter trenbolone acetate has potent androgenic activity (Wilson et al., 2002; reviewed in Schiffer et al., 2001). Since the current trend is toward the use of exogenous hormones and intensive farming practices, the potential for androgenic effects arising from the usage of animal manure on agricultural lands adjacent to environmentally sensitive areas cannot be discounted.
The purpose of the present study was to determine the persistence and pathways of testosterone degradation in soil. Various soil types and environmental conditions were investigated.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Soil Microcosms
The properties of the loam, sandy loam, and silt loam soils used in this study have previously been described in Colucci and Topp (2002). Sieved (2-mm maximum particle size) moist soils were stored at 5°C for periods of up to 6 mo before experimentation. Sterile soil was prepared by autoclaving twice (45 min at 120°C), the second time following a 24-h room temperature incubation.
Microcosms consisted of a small baby-food jar placed within a sealable glass 1-L Mason jar. A scintillation vial containing 10 mL of water was placed in each jar to maintain a humid atmosphere and prevent desiccation of the soil. In experiments evaluating mineralization of [414C]-testosterone, a second scintillation vial was added containing 5 mL 1 M NaOH to trap 14CO2. The vial containing NaOH was periodically replaced and 10 mL of Universol liquid scintillation counting (LSC) cocktail were added before LSC analysis.
Soils were supplemented with unlabeled, 3H- or 14C-labeled hormone by adding an ethanolic stock solution to 1 g of air-dried pulverized soil, allowing the solvent to evaporate, and thoroughly incorporating this into 25-, 50-, or 100-g portions (moist weight) of the corresponding soil. Unless otherwise indicated, microcosms were incubated at 30°C with an initial hormone concentration of 1 mg/kg moist soil. Soil moisture content was adjusted gravimetrically, and since moisture contents were held constant during all incubations, hormone concentrations are expressed on a moist soil basis.
Soil Sample Extraction
Periodically, 5-g portions of soil were removed from each microcosm. Each was weighed into a clean glass vial and extracted twice with 10 mL of ethyl acetate and once with 10 mL of acetone. For each extraction samples were shaken vigorously by hand for 20 s, transferred to a Burrell wrist action shaker (Burrell Corporation, Pittsburgh, PA), and vigorously shaken automatically for 10 min. The sample was then centrifuged (Model GLC-1, 900 rpm [164 x g]; Sorvall, Asheville, NC) for 10 min after which the supernatant was transferred into a clean glass vial. The pooled supernatants were reduced to dryness under nitrogen in a 37°C water bath and the dried extracts were stored at 7°C. For use in the bioassays, the unlabeled extracts were dissolved in 1 mL of absolute ethanol and serial dilutions were prepared. For high performance liquid chromatography (HPLC)radioactivity detection (RD) or LSC the dried radiolabeled extracts were dissolved in 500 µL of ethanol of which 125 µL were used for LSC analysis after addition of 10 mL of LSC cocktail. The remaining three-quarters of the extract was used for HPLCRD analysis.
Preliminary experiments established that the efficiency of testosterone extraction following addition to the three soils used in this study was 84 ± 2% (n = 9).
Chemical and Radioisotope Methods
Radioactivity was measured with a Model LS 6500 liquid scintillation counter (Beckman Coulter, Irvine, CA). Each sample received 10 mL of UniverSol scintillation cocktail (ICN, Cosa Mesa, CA). Quenching was corrected using an external standard.
Parent compounds and transformation products were analyzed by reverse phase HPLCRD (Model LB509 radioflow detector; Berthold, Bad Wildbad, Germany). Instrument operating conditions were as follows: column, 25.0 cm x 4.6 mm (ODS-3 packing; Whatman, Maidstone, UK); mobile phase, acetonitrile and water (40:60). With solvent delivered at 1 mL/min testosterone had a retention time of 20.1 min. Using HPLCRD (60 µL sample injection), the lower limit of detection for [1,2,6,73H(N)]-testosterone in extracts prepared from the soils used in this study was 0.060 ng. Taking into account our extraction methods, this corresponded to a method detection limit of 13.6 ng/kg soil.
Degradation products corresponding to the radiolabeled materials detected by HPLCRD were isolated from extracts of soil treated with unlabeled testosterone by collecting appropriate fractions under the same chromatography conditions with the aid of UV detection at 220 nm. We used HPLCmass spectrometer detection to identify the degradation products in these fractions. The HPLCmass spectrometer detection system consisted of an Alliance 2690 HPLC/autoinjector and a model LCT orthagonal time-of-flight mass spectrometer (Waters, Milford, MA). Chromatography was performed on a 2- x 150-mm, 5-µm particle Prodigy ODS(3) column (Phenomenex, Torrence, CA) operated isocratically with 20:70:10 (v/v) wateracetonitrile1:99 (v/v) formic acid in water flowing at 0.2 mL/min. The ion source was equipped with a standard nebulizer assisted electrospray probe, which was operated in positive ion mode with nitrogen as desolvation gas at 225°C flowing at 460 L/h and a potential of 2.6 kV applied to the capillary. The sample cone was maintained at 30 V.
Androgen Receptor Gene Transcription Bioassay
The human androgen receptor recombinant yeast strain YPH500 was kindly provided by Dr. K.W. Gaido and Dr. D.P. McDonnell. The yeast were maintained and used for androgen receptor gene transcription assays as originally reported (Gaido et al., 1997), with modifications to allow the use of 96-well plates for both extract exposure and end-point spectrophotometer measurements as described previously (Lorenzen et al., 2004). The relative androgenic potencies of testosterone and three transformation products (4-androstene-3,17-dione, 5
-androstan-3,17-dione, and 1,4-androstadiene-3,17-dione) were determined using this assay.
Calculations
All treatments were in triplicate, and data in figures are expressed as mean ± standard deviation. The mineralization of [414C]-testosterone was determined by measuring liberated 14CO2 in NaOH traps by LSC and expressing the accrued radioactivity as a percentage of the total initial radioactivity in the microcosms. Alternatively, the mineralization of [1,2,6,73H(N)]-testosterone was estimated as follows. The radioactivity in a soil extract was determined by counting a portion by LSC. Another portion of the extract was evaporated to dryness under a stream of nitrogen. The residue was dissolved in ethyl acetate and this residue was counted. The difference was taken to be due to evaporative loss of 3H2O, due to mineralization of the corresponding tritium atoms in the testosterone. Extractable radioactivity measured in extracts by LSC is presented as a percentage of the initial radioactivity added. The disposition of extractable radioactivity in parent compound and transformation products was estimated according to the peak areas of compounds resolved by HPLCRD.
Full concentrationresponse data were obtained for each extract or standard compound using the recombinant yeast androgen receptor gene transcription bioassay exactly as described previously (Lorenzen et al., 2004).
| RESULTS |
|---|
|
|
|---|
|
|
-androstan-3,17-dione (39.9 min); and 1,4-androstadiene-3,17-dione (18.9 min). The percentage of total 14C present in these products at 30°C 6 h after addition of [414C]-testosterone was 48.7, 23.7, and 9.6%, respectively. The androgenic potencies of these products relative to testosterone were determined in the yeast androgen receptor gene transcription assay by determining the concentration of an analytical standard of each transformation product that was required to produce a response equivalent to 20% of the maximal response obtained with testosterone (EC20; Fig. 5)
. 4-Androstene-3,17-dione, 5
-androstan-3,17-dione, and 1,4-androstadiene-3,17-dione were 20, 80, and 3% as potent as testosterone, respectively.
|
|
|
| DISCUSSION |
|---|
|
|
|---|
-ethynylestradiol (Colucci et al., 2001; Colucci and Topp, 2001). Testosterone concentrations fell below our method detection limit of 13.6 ng/kg, consistent with the degree of dissipation previously observed with 17ß-estradiol and 17
-ethynylestradiol (Colucci and Topp, 2002). Dry conditions and reduced temperatures resulted in slower dissipation of testosterone as also previously observed with estrogenic hormones. Testosterone and androstenedione adsorption increases with the organic matter content of the soil (Lee et al., 2003; Casey et al., 2003, 2004). Our results support the notion that soil sorption should not significantly reduce bioavailability and biodegradation, and therefore enhance any environmental risk from testosterone or testosterone transformation products (Das et al., 2004).
Testosterone was converted into three transformation products that were identified by HPLC retention times and mass spectra as 4-androstene-3,17-dione, 5
-androstan-3,17-dione, and 1,4-androstadiene-3,17-dione. Application of testosterone to sterile soil resulted in the production of 4-androstene-3,17-dione only. All three of these transformation products had androgenic potencies less than that of testosterone and, based on their relative abundance and rapid dissipation under temperate conditions, would not be expected to contribute significantly to the total androgenicity detected in the soil. Both 4-androstene-3,17-dione and 5
-androstan-3,17-dione have been detected in sewer effluent in the UK (Thomas et al., 2002). In that study it was calculated that these compounds accounted for 36 and 33%, respectively, of the total androgenic activity detected in the effluent.
Testosterone was much more stable in sterile soil than nonsterile soil, suggesting that biodegradation is the key determinant of dissipation. The absence of a detectable lag phase in nonsterile soil suggests that rapid dissipation is not predicated on induction or proliferation of testosterone-degrading microorganisms. This finding is similar to that previously reported for the estrogenic hormones estradiol and 17
-ethinylestradiol (Colucci et al., 2001; Colucci and Topp, 2001). The rapid conversion of estradiol to estrone and testosterone to 4-androstene-3,17-dione in soil suggests that in both cases the first step in the biodegradation pathway is likely to be the oxidation of the alcohol at Position 17 to the corresponding ketone. A comparison of the fate of radiolabeled atoms in the A rings of testosterone or estradiol suggests that the dissipation pathways thereafter are quite different. Atoms at Positions 4 (14C) or 1, 2, 6, and 7 (3H) of testosterone were very rapidly and thoroughly mineralized. In contrast, [414C]-17ß-estradiol rapidly formed non-extractable soil-bound residues, and was only very slowly mineralized (Colucci et al., 2001). Steroids generally are subject to a wide variety of biotransformation reactions, with the attack initiated at different locations on the molecule in different ways. For example, Comamonas testosteroni dismembers the B ring, resulting in the A ring being cleaved off, which is then subject to complete degradation (Horinuichi et al., 2003). Androst-4-ene-3,17-dione monooxygenase, carried by the fungus Cylindrocarpon radicicola for example, catalyzes the lactonization of the D ring of 4-androstene-3,17-dione to form testololactone (Prairie and Talabay, 1963). Finally, P-450 mono-oxygenase enzymes widely employed in microbial steroid catabolism can hydroxylate various positions of the androstane hydrocarbon steroid skeleton. Overall, the ability to biodegrade natural steroidal hormones is likely to be ubiquitous in environments that are rich in microorganisms, such as soils.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
-ethynylestradiol. J. Environ. Qual. 30:20772080.Related articles in JEQ:
This article has been cited by other articles:
![]() |
H. A. Sangsupan, D. E. Radcliffe, P. G. Hartel, M. B. Jenkins, W. K. Vencill, and M. L. Cabrera Sorption and Transport of 17{beta}-Estradiol and Testosterone in Undisturbed Soil Columns J. Environ. Qual., October 27, 2006; 35(6): 2261 - 2272. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. N. J. Hemmings and P. G. Hartel Mineralization of Hormones in Breeder and Broiler Litters at Different Water Potentials and Temperatures J. Environ. Qual., April 3, 2006; 35(3): 701 - 706. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.-M. Jacobsen, A. Lorenzen, R. Chapman, and E. Topp Persistence of Testosterone and 17{beta}-Estradiol in Soils Receiving Swine Manure or Municipal Biosolids J. Environ. Qual., April 20, 2005; 34(3): 861 - 871. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||