Environ. Sci. Technol. 2007, 41, 1450-1456
Chlortetracycline Detoxification in Maize via Induction of Glutathione S-Transferases after Antibiotic Exposure MICHAEL H. FARKAS,† J A M E S O . B E R R Y , † A N D D I A N A S . A G A * ,‡ Departments of Biological Sciences and Chemistry, University at Buffalo, Buffalo, New York 14260
Soil contamination with nonmetabolized antibiotics is an emerging environmental concern, especially on agricultural croplands that receive animal manure as fertilizer. In this study, phytotoxicity of chlortetracycline (CTC) antibiotics on pinto beans (Phaseolus vulgaris) and maize (Zea mays) was investigated under controlled conditions. When grown in CTC-treated soil, a significant increase in the activities of the plant stress proteins glutathione Stransferases (GST) and peroxidases (POX) were observed in maize plants, but not in pinto beans. In vitro conjugation reactions demonstrated that the induced GST in maize catalyzed the conjugation of glutathione (GSH) with CTC, producing stable conjugates that were structurally characterized using liquid chromatography/mass spectrometry. The antibiotic-induced GST produced CTC-glutathione conjugate at relative concentrations 2-fold higher than that produced by constitutively expressed GST extracted from untreated maize. On the other hand, GST extracted from pinto beans (both treated and untreated) did not efficiently catalyze glutathione conjugation with CTC. These results suggest that maize is able to detoxify chlortetracycline via the glutathione pathway, whereas pinto beans cannot. This may explain the observed stunted growth of pinto beans after antibiotic treatment. This study demonstrates the importance of plant uptake in determining the fate of antibiotics in soil and their potential phytotoxicity to susceptible plants.
Introduction Antibiotics used in human medicine and animal production enter the environment as contaminants when sewage sludge and animal manure are used to fertilize croplands (1, 2). Up to 80% of administered antibiotics are unmetabolized in the body of the treated organisms, and thus are excreted in urine and manure in the active form (3). Antibiotic contamination of the environment has become an emerging issue, as the United States Environmental Protection Agency (USEPA) reported that approximately 10 million tons of sewage sludge and manure are applied to croplands each year (4). While the promotion of antibiotic resistance in naturally occurring pathogens is the major concern, other issues associated with antibiotics in the environment have recently gained attention. * Corresponding author phone: 716-645-6800 ext 2226; fax: 716645-6963; e-mail:
[email protected]. † Department of Biological Sciences. ‡ Department of Chemistry. 1450
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007
One issue relates to whether antibiotics are taken up by plants and are stored within edible parts, potentially exposing consumers to low doses of these drugs. Another concern deals with whether antibiotics have significant phytotoxicity that can lead to a reduction in crop yields. To this date, no studies have been performed to quantify the ecological toxicity of antibiotics. However, a recent study by Capelton and co-workers (5) has prioritized 83 veterinary pharmaceuticals based on their usage, likelihood to reach the environment, and toxicity profile, of which 32 were classified as medium to high priority for detailed risk assessment. A few studies have demonstrated uptake of pharmaceuticals from soil by crop plants. For example, chlortetracycline is taken up by cabbage and green onion from contaminated soil (6); carrots and lettuce leaves also took up one or more test drugs, which included diazinon, enrofloxacin, florfenicol, and trimethoprim (7). The first report of phytotoxic effects of antibiotics in crops was that by Batchelder (8) who observed that growth of pinto beans is negatively impacted by the presence of chlortetracycline in soil. Recently, sulfadimethoxine, another agriculturally important antibiotic, was shown to be phytotoxic toward both maize and barley (9). Similarly, the antibiotic enrofloxacin was shown to alter post-germinative development of several agricultural crops (10). Despite serious agricultural implications, very little research has been conducted to elucidate the mechanisms of antibiotic uptake, accumulation, and detoxification by plants, or to explain variations in plant response of different plant species when grown in antibiotic-contaminated soil. For example, studies conducted in our laboratory showed that pinto beans grown in chlortetracycline-treated soil exhibited smaller leaves and pods relative to the untreated plants (11). In contrast, maize plants appeared to be unaffected by antibiotic treatment. Similar results were observed by Batchelder (8) in his early studies. These results suggest that certain plants are sensitive to antibiotics while others may have a mechanism for detoxification of the drugs after uptake. Research on the impact of antibiotics in the environment has mainly focused on determining persistence, transport, and concentrations in the soil; little work has been done to understand plant-antibiotic interactions. In this regard, understanding the mechanisms and pathways of antibiotic phytotransformation after uptake by plants is warranted. For instance, prior to this work, the roles of stress proteins such as glutathione S-transferase (GST) and peroxidase (POX) have not been investigated in plant-antibiotic interactions. GSTs have long been known to function in normal plant metabolic processes (12-14), and have served crop plants by detoxifying the plethora of agriculturally applied pesticides via the glutathione conjugation pathway. The POX enzyme system is important in plant defense in which its general function is to fortify the cell wall, thus resisting disease progression (15). Their expression has been observed to increase within the roots as a response to xenobiotic stress where they oxidize and deactivate chemical xenobiotics (16, 17). The induction of peroxidase expression in plants is expected in response to many types of stressors. Thus, the increase in peroxidase activity was measured in this study to serve as an additional indicator of a plant’s physiological response to chlortetracycline. The objectives of this study are to (1) determine the ability of agricultural crops to detoxify chlortetracycline, using maize (Zea mays) and pinto beans (Phaseolus vulgaris) as our model plants, and (2) evaluate the induction of plant stress enzymes and their role in the transformation of antibiotics in plants. 10.1021/es061651j CCC: $37.00
2007 American Chemical Society Published on Web 01/11/2007
Experimental Section Materials. The Bradford assay kit for protein quantification, chlortetracycline (g97%, analytical grade), 1-chloro-2,4dinitrobenzene (CDNB, g97%), reduced glutathione (g99%), and trichloroacetic acid (g99%) were purchased from SigmaAldrich, CA. Glutathione-agarose affinity kits consisting of glutathione linked via the sulfur group to epoxy-activated agarose beads were purchased as a lyophilized powder from Sigma-Aldrich, CA. Tris-HCl, EDTA, and phenylmethanesulfonyl fluoride (PMSF) were all of molecular biology grade and also purchased from Sigma-Aldrich, CA. Guaiacol was purchased from Caymen Chemicals, MI. The organic solvents acetonitrile and methanol were ACS grade (Burdick & Jackson, MI). Soil consisting of peat moss, perlite, and vermiculite (Burton Flower and Garden, OH) was mixed on-site. Plant Material and Treatment with Chlortetracycline (CTC). CTC has been detected in the field at concentrations as high as 20 mg kg-1 (18). We used this concentration to allow us to observe plant response to antibiotic contamination close to a “worse case scenario”. CTC was prepared in methanol at a concentration of 20 mg L-1 and was slowly applied onto 1 kg of soil with constant mixing. Once treated, the soil was aged in a closed plastic bag for 24 h to allow equilibration. Maize and pinto bean seeds were allowed to germinate in untreated vermiculite soil incubated in a growth chamber. Growth conditions were maintained at 27 °C under a 14-h photoperiod at a light intensity between 170 and 200 µE m-2 min-1. Ten-day-old seedlings were transplanted into soil that had been spiked with 20 mg kg-1 CTC. Plants were harvested for analysis after growing in the CTC-treated soil at 1, 2, 3, and 30 days. During each sampling, harvested plants were uprooted and soil was washed away from the roots. The roots were then blotted dry, frozen in liquid nitrogen, and stored at -80 °C until analysis. Protein Isolation and Quantification. Approximately 70 mg of root material from each plant was ground on ice using a glass homogenizer, in the presence of 0.5 mL of protein extraction buffer (50 mM Tris-HCl, pH 7.2, 1 mM EDTA, and 10 mM β-mercaptoethanol, and 1 µL of 20 mg mL-1 PMSF). The homogenate was centrifuged at 10 000g for 1 min, and the supernatant containing the crude protein extract was collected. For affinity purification of GST from the crude extract, 1 mL of the crude extract was passed over a glutathione-agarose column and isolated following the procedure provided by the manufacturer, using elution buffer consisting of 75 mM reduced glutathione in 50 mM TrisHCl, pH 9.5. Protein concentrations in both crude extracts and affinity-purified GST were measured using the methods of Bradford (19), with bovine serum albumin as a standard. Representative protein samples were precipitated by ethanol and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a 15% acrylamide gel mixture. Measurement of GST Activity. The GST activities of the crude protein extracts were determined spectrophotometrically using CDNB as the standard substrate. The reaction mixture contained 1.0 mM CDNB, 5.0 mM GSH, phosphate buffered saline (PBS, pH 6.5), and 20 µL of the protein extract (20). The final volume of the reaction was 200 µL. The change in absorbance was measured at 25 °C for 5 min at 340 nm. The enzyme activity was calculated using an extinction coefficient of 9.6 mM-1cm-1 (21). Variable amounts of proteins, ranging from 25 to 75 µg per sample, were used in the assay because constant volumes of plant extracts were added to the reaction mixture. However, the reported enzyme activities were corrected to account for these variations by dividing the activity of each sample with its respective protein concentration. A total of six replicates were measured three
times and corrected for any nonenzymatic conjugation by subtracting the change in absorbance of a reaction mixture that did not contain the protein extract. Measurement of POX Activity. The POX activities of the crude protein extracts were also determined spectrophotometrically. The reaction mixture contained 68 µM guaiacol, 0.9 mM H2O2 as substrate, and 50 mM potassium phosphate buffer (pH 6.0). The change in absorbance was measured at 25 °C for 5 min at 420 nm. The enzyme activity was determined by using an extinction coefficient of 26.6 mM-1cm-1 (22). All data analysis was performed in the same manner as the GST activity assays described above. In vitro CTC-GSH Conjugation and LC/MS/MS Analysis. To evaluate the catalytic activity of the extracted GSTs for GSH conjugation with CTC, reactions were performed in vitro using the affinity-purified protein extracts. Samples containing 5.0 mM GSH, 0.02 mM CTC, and 100 µL of purified GST were added to PBS (pH 6.5) at a final volume of 500 µL. These samples were incubated for 1 h at 30 °C. As controls, nonenzymatic reactions (mixtures lacking GSTs) were incubated for 18 h at 45 °C. The reactions were stopped by the addition of 5% trichloroacetic acid. Samples were prepared for analysis by adding a 50:50 mixture of methanol/acetonitrile to a final volume of 1.0 mL. The reaction mixtures were analyzed by liquid chromatography (LC) with ion-trap mass spectrometry (IT-MS) using an LCQ Advantage IT-MS system equipped with an electrospray ionization source (ESI) and connected to a Surveyor LC system (ThermoFinnigan, San Jose, CA). Separation was performed on a BetaBasic-18 C18 column (100 × 2.1 mm i.d. with 3-µm particle size) equipped with a Uniphase guard cartridge (10 × 2.1 mm i.d. with 3-µm particle size), both purchased from Thermo Hypresil-Keystone (Bellefonte, PA). The flow rate was 200 µL/min with an injection volume of 10 µL. Sample components of interest were separated using a linear gradient mobile phase consisting of 95% water with 0.3% formic acid to 5% acetonitrile over a 15-min period, ending with a mobile phase consisting of 5% water and 95% acetonitrile. The ESI-IT-MS was operated in positive ion mode with a source voltage of 5.20 kV and capillary voltage of 19 V using a nitrogen sheath gas. MS/MS data were collected using a scan range between 185 and 800 m/z at a collision energy of 35% using helium as the dampening gas.
Results and Discussion CTC-Induced Changes on Protein Expression. Prior to this, no biochemical studies on plants grown in antibiotic-treated soil have been reported in the literature. Thus, the goal of our study was to determine if measurable biochemical changes could be observed when pinto beans and maize were grown in CTC-treated soil. Each day for 3 days after re-planting seedlings into antibiotic-treated soil, plants were harvested and root extracts were prepared to determine potential changes in protein expression, assayed by SDSPAGE. Roots were chosen for this study because the root extracts showed the most significant differences in SDS-PAGE banding patterns between the treated and control plants. Also, in our preliminary studies, the differences in the GST activities in the protein extracts from the upper plant tissues where not discernible between each treatment (23). Previous studies have shown that changes in protein expression levels can occur within a matter of hours, to a few days; this is dependent upon the rate at which the xenobiotic enters the plant because induction of proteins responsible for xenobiotic defense and detoxification normally occurs rapidly (9, 2426). Therefore, a 3-day exposure to antibiotics would be expected to induce changes in the plant biochemistry of the pinto beans and maize plants used for our analysis. In maize roots, SDS-PAGE revealed protein bands in the range of 20-30 kDa, which increased noticeably in the CTCVOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1451
FIGURE 1. GST activity measured from total protein extracts 1, 2, and 3 days after plants were treated with CTC: (A) Maize control (MCR) and CTC-treated (MTR) plants; (B) Pinto bean control (PCR) and CTC-treated (PTR) plants. Values represent the mean and standard deviation of six replicates. Asterisks denote statistically significant data (p < 0.05). treated maize roots relative to the untreated control roots (Supporting Information 1A). Because the SDS-PAGE was performed using crude protein extracts it is possible that the band at 20-30 KDa contains many proteins. However, we hypothesized that GSTs were primarily contained within this protein band based on their molecular weight and rapid inducibility. This premise was based on previous studies demonstrating that maize induces expression of GST enzymes to detoxify a variety of xenobiotics after exposure to phytotoxic compounds (such as herbicides) and nonphytotoxic xenobiotics (such as herbicide safeners) (27-29). Interestingly, and in contrast to maize, comparison of the expressed pinto bean proteins in the SDS-PAGE revealed no apparent difference in the banding patterns between the CTC-treated plants and the untreated plants (Supporting Information 1B). GST and POX Activities. To assay for enzyme activities, total proteins from root extracts were analyzed 1, 2, and 3 days after replanting the seedlings into CTC-treated soil. In the case of maize roots, samples from days 1 and 3 of the CTC-treated plants displayed a significant increase (p < 0.05) in GST activities, relative to their corresponding control plants (Figure 1A). Day 2 root samples also displayed increased GST activity, but not within the range of statistical significance. However, the GST activities in the roots of pinto beans showed no significant difference between the treated and control plants at any day (Figure 1B). The results of the protein activity assays were in agreement with the observed banding patterns manifested in the SDS-PAGE. It was predicted that the increase in the band intensity at 20-30 kDa was due to increased expression of GSTs by maize in response to CTC 1452
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007
FIGURE 2. POX activity measured from total protein extracts 1, 2, and 3 days after plants were treated with CTC: (A) Maize control (MCR) and CTC-treated (MTR) plants; (B) Pinto bean control (PCR) and CTC-treated (PTR) plants. Values represent the mean and standard deviation of six replicates. Asterisks denote statistically significant data (p < 0.05). exposure; hence, the significant increase in the specific activity of GSTs in the roots of CTC-treated maize was a predicted outcome. On the other hand, we expected that there would not be a difference in the GST activities between the CTC-treated and untreated pinto beans, based on the observations of the SDS-PAGE; results of the GST activity assays for pinto beans were consistent with our hypothesis. Interestingly, when GST activities between the CTC-treated maize and pinto bean plants were compared, activities of the pinto beans were 1.5 to 2-fold lower than that of maize. These results suggest that GSTs may be responsible for CTC detoxification in maize, but not in pinto beans, providing an explanation for observed growth differences in treated and untreated pinto beans reported in earlier studies (8, 10). Peroxidase (POX), like GSTs, is a commonly induced enzyme utilized in plant defense pathways; therefore, it is a good indicator of stress and helps corroborate the increase in GST expression in plants (24, 29-31). In this study, the same protein extracts previously assayed for GSTs were also assayed for POX activities using guaiacol as substrate. Guaiacol cannot be utilized by glutathione peroxidases; therefore it is a specific indicator of other endogenous peroxidase activity. The results of the POX assays followed the same trend as the GST activity assays. The CTC-treated maize roots had a significant increase (p < 0.05) in POX activity at days 1 and 3 posttreatment (Figure 2A), whereas pinto beans showed little difference in POX activity (Figure 2B). In addition, the POX activities in CTC-treated pinto beans were much lower than those of the CTC-treated maize plants. Since POX is a general defense enzyme, its activity is a good
indicator of a plant’s ability to recognize a stressor. These data, along with the GST activity data, are indicative of the ability of maize to recognize a stressor and respond to it. The lack of an increase in the GST and POX activities in the CTCtreated pinto beans, on the other hand, indicates that pinto beans are unable to efficiently induce a protective response against CTC. These differences in the biochemical responses between maize and pinto beans may explain why CTC affected the growth and development of pinto beans, but had no observable effect on maize. Characterization of In vitro CTC-GSH Conjugate Formation by LC/MS/MS. The identification of a CTC-GSH conjugate is important to show that GST has the ability to catalyze the conjugation of GSH with CTC. In vivo formation of CTC-GSH conjugates is difficult to measure using LC/MS methods because of the complex matrix of the plant. Other sensitive methods such as enzyme-linked immunosorbent assays are not sufficient because these techniques do not allow a distinction between the parent compound or other metabolites (18). For these reasons, we were unable to quantify uptake of CTC and the concentrations of its conjugates to determine the efficiency of the phytotransformtion of CTC in vivo. In this regard, in vitro conjugation reactions were performed using affinity-purified GSTs from both control and CTC-treated pinto beans and maize. This allowed for positive identification of both nonenzymatic and enzymatic products, as well as their relative abundances. The reaction products were characterized using ion-trap LC/ MS/MS. The chromatogram (Figure 3A) of the GST-catalyzed reactions displayed two characteristic features of the putative CTC-GSH conjugate: (1) the conjugate eluted sooner than CTC (Figure 3B), and (2) the conjugate eluted in multiple peaks. Based on the known GST detoxification reactions of chlorinated herbicides (32), conjugation is expected to occur via a nucleophilic substitution of chlorine in the toxicant by the sulfur atom of the GSH, forming a highly polar GSHconjugate, as illustrated in Figure 3A. The removal of chlorine and the addition of a glutathione moeity results in a highly water soluble CTC-GSH conjugate, thus a much shorter retention time relative to the more hydrophobic CTC is expected. The presence of multiple peaks with identical mass spectra can be explained by GSH conjugation with the different isomeric and epimeric forms of CTC, which can be separated chromatographically (33). Several fragmentation patterns in the mass spectrum of the CTC-GSH conjugate can be used to verify the identity of the enzymatic product. Since chlorine is electrophilic, it is expected to be removed during GSH conjugation, because electrophilic atoms are typically the targeted sites of GSTs (34). The expected molecular ion [M + H]+ of CTC-GSH conjugate (m/z 751) with chlorine removal was not observed in the positive electrospray ionization. Instead, a base peak of m/z 677, corresponding to the loss of glycine (MW ) 75 Da) was observed. The absence of [M + H]+ in the mass spectra is likely due to the electronic nature of the CTCGSH conjugate, which produces an unstable molecular ion. When m/z 677 was isolated in the ion trap and subjected to further fragmentation, characteristic peaks corresponding to the loss of glutamic acid (m/z ) 129) and a water molecule (m/z 18) were observed in the MS/MS spectrum as shown is Figure 3C. This fragmentation pattern has been observed with other GSH conjugates (35). The absence of the isotopic chlorine signature in the mass spectrum of the CTC-GSH conjugate further supports the proposed removal of chlorine in the CTC molecule via glutathione conjugation. To verify the catalytic role of GSTs in the conjugation of GSH with CTC, in vitro reactions were performed in the absence of GST enzymes. Incubation of the nonenzymatic reaction mixtures under the same conditions used in the
GST-catalyzed reactions did not produce any detectable CTC-GSH conjugates, as determined using the same LC/ MS/MS analysis. For additional verification, the reactions were incubated at 42 °C for 18 h to allow production of detectable amounts of conjugates by mass spectrometry. LC/ MS/MS analysis of the products formed in the nonenzymatic reaction displayed strikingly different results than that of the enzyme conjugated CTC-GSH products (Figure 4). First, the retention times of the nonenzymatically formed products overlapped with CTC peaks in the chromatogram. This suggests that if GSH did conjugate nonenzymatically, it occurred at a different site and chlorine was not removed, retaining the relatively nonpolar nature of chlorinated compounds. Indeed, the mass spectra of the conjugates exhibited the chlorine isotopic signature, indicating that the site of conjugation did not involve the nucleophilic substitution of chlorine typical of GST-mediated reactions (see inset in Figure 4). Second, two different products (m/z 654 and m/z 695) were formed, each eluting as a single peak, unlike the characteristic three peak isomers of chlortetracycline. This means that the site of conjugation occurred such that isomerization and epimerization of the CTC molecule was no longer favored. While the expected nominal mass of the molecular ion [M + H]+ at m/z 786 corresponding to the nonenzymatic CTC-GSH conjugate (no dechlorination) was not observed, fragment ions corresponding to the loss of glycine and other characteristic fragmentation patterns of glutathione conjugates confirmed the presence of CTC-GSH conjugates (Supporting Information 2A and 2B). MS/MS analysis of m/z 654 produced three distinctive product ions (SI Figure 2A). First, the characteristic losses of glutamic acid (m/z 525) and water (m/z 507) were observed. We also observed a product ion at m/z 422, which is suggestive of the loss of cysteine. The MS/MS ion spectrum of m/z 695 was more complex than the other conjugate, but characteristic fragmentation in the mass spectrum verified the formation of another CTC-GSH conjugate (SI Figure 2B). For instance, the fragment ion at m/z 566 resulted from the loss of glutamic acid. The fragment ions at m/z 548 and 531 resulted from further loss of two water molecules and an NH3 group that are characteristic of CTC, as typically observed in MS/MS analysis of CTC standards (33). The potential site of GSH conjugation may involve alkyl groups (CdC) (36) and result in the ring opening of CTC, hence the absence of epimerization in the chromatogram. However, the exact location of the GSH conjugation to CTC during the nonenzymatic reaction cannot be assigned without further NMR analysis, which is beyond the scope of our present study. Taken together, the information obtained from the MS/MS analysis of the CTC-GSH conjugates provides very important evidence supporting the catalytic activity of the GST enzymes during dechlorination and potential detoxification of CTC antibiotics in maize plants. CTC-Induced GST Activity. It is important to determine the contribution of the constitutively expressed GSTs in the enzymatic conjugation of GSH with CTC. Therefore, in addition to the standard CDNB activity assays we measured the relative amounts of GSH conjugates formed during enzyme-mediated in vitro reactions by LC/MS/MS using GST enzymes from treated and untreated plants. Based on the LC/MS/MS peak areas, we observed that the maize GSTs from the untreated control plants were only able to produce 53% of the CDNB-GSH (m/z 475 was monitored) product relative to GSTs from the CTC-treated maize plants (3 days after treatment) (Figure 5A). These data suggest that constitutively expressed GST enzymes have lower specific activities toward CDNB conjugation. Similarly, the peak areas of CTC-GSH conjugates using GSTs from the treated and untreated maize were also determined. Here, maize GSTs purified from the untreated control plants produced about VOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1453
FIGURE 3. In vitro LC/MS/MS data for GST-mediated CTC-GSH conjugation: (A) Chromatogram of m/z 677 conjugate and hypothesized product of a GST-mediated CTC-GSH reaction (inset); (B) Chromatogram of CTC demonstrating the difference in polarity relative to the observed conjugate and the similarities in isomeric peaks; and (C) chemical structure of the fragment ion m/z 677 and its MS/MS fragmentation spectrum. 56% as much of the CTC-GSH product as compared to the CTC-treated GSTs (Figure 5B). These results correlate well with the observed GST activities described earlier using CDNB activity assays. These data suggest that constitutively expressed GSTs from maize have the inherent ability to detoxify CTC. However, maize exposure to CTC in the soil induces expression of GST isoforms that are specific to the CTC as a stressor, which enhances its detoxification via GSH conjugation. 1454
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007
In comparison to the CDNB-GSH enzyme-mediated conjugates, the maize samples were able to form about 2% as much CTC-GSH products relative to the respective treatment (data not shown). The ability of GST to produce a CTC-GSH conjugate appears minor relative to the GST production of a CDNB-GSH conjugate, but it is comparable to other studies conducted using alachlor, a chlorinated herbicide readily detoxified by maize (20). It is unlikely that nonenzymatic reactions contribute to CTC detoxification in
FIGURE 4. LC/MS/MS chromatogram of nonenzymatic CTC-GSH products formed in vitro. The peaks at 9.89 and 10.00 min correspond to CTC-GSH conjugates with a m/z of 654.0 and 695.0 (inserts), respectively. Both products contain a characteristic (M + 2)+ chlorine isotopic signature.
FIGURE 5. Relative abundance of (A) CDNB-GSH conjugates and (B) CTC-GSH conjugates from CTC-treated and control maize. Data were determined using peak areas obtained from LC/MS data of products formed using GSTs isolated from day 3 plants for the in vitro reactions. maize, based on an analysis of the peak areas of the products formed. When the peak areas were determined and adjusted for incubation time, nonenzymatic conjugation accounted for only 3% of total product formation relative to CTC-GSH conjugation using CTC-treated GSTs. Implications in Agriculture and Environment. The study presented here is the first to demonstrate the ability of antibiotics to induce production of GST isozymes that are important in detoxification of chlortetracycline. Several interesting issues arise from this work. First, residues of veterinary antibiotics that are unintentionally applied to soil through the land application of manure may have phytotoxic effects on susceptible agricultural crops, which could lead to losses in production. Whether the negative effects of antibiotics on crop production is significant enough to outweigh the benefits of using manure as fertilizer is yet to be determined. Second, while plant uptake of antibiotics has been shown to be detrimental to some crops (6, 9, 10), it is not yet known whether antibiotics bioaccumulate in edible plant tissues and potentially expose consumers to low concentrations of antibiotics. For instance, uptake of other organic pollutants such as trichloroethylene has been
quantified in the fruit of apple and pear trees by Chard and co-workers (37). On the other hand, our research demonstrates the involvement of the glutathione conjugation pathway in dechlorination of CTC by maize. It is likely that endogenous soil microorganisms do not play a significant role in CTC detoxification because the major pathways for resistance toward tetracycline antibiotics involve efflux and ribosomal protection in microorganisms; hence the pathway for degradation of tetracyclines is much less prevalent (38). In fact, new studies have shown that efflux and ribosomal protection are induced by tetracycline introduced into the environment from wastewater lagoons (39). It appears that certain agricultural crops may be used for the natural remediation of antibiotic-contaminated soil and that this may be important for reducing antibiotic resistance in the environment. A recent study by Gujarathi and Linden showed that exudates from hairy roots of sunflower promote oxidation of oxytetracycline into products devoid of antibiotic activity. Together, these studies indicate that there may be several other agricultural crops that can be used in phytoremediation of antibiotics from crop fields. However, basic knowledge on the types of plants capable of detoxifying antibiotics and the conditions that favor effective biodegradation of these contaminants in the environment is needed. With more information, one could develop ways to utilize agricultural crops in phytoremediation of antibiotics and devise crop rotation programs that would prevent the accumulation of persistent antibiotics in soil. Phytoremediation is an expanding area of research where field studies have shown it to be effective in remediating sites contaminated with pollutants such as heavy metals, pesticides, and explosives (40). The findings that GST detoxification is involved in the biotransformation of chlortetracycline is encouraging, because it suggests that antibiotics may be removed by the same mechanisms as other known pollutants via glutathione conjugation.
Supporting Information Available Figure S1 of the Supporting Information demonstrates the differential expression of proteins between control and CTCtreated maize (S1A) and pinto beans (S1B). Central to our hypothesis is the increased protein expression within the range of 20-30 kDa. Figure S2 of the Supporting Information includes the LC/MS/MS spectrum of the nonenzymatic CTCGSH conjugates with m/z of 654 (S2A) and 695 (S2B). The insets are hypothesized structures based on the fragmentation VOL. 41, NO. 4, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1455
data for each conjugation product. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) McManus, P. S.; Stockwell, V. O.; Sundin, G. W.; Jones, A. L. Antibiotic use in plant agriculture. Annu. Rev. Phytopathol. 2002, 40, 443-465. (2) Khachatourians, G. G. Agricultural use of antibiotics and the evolution and transfer of antibiotic-resistant bacteria. CMAJ 1998, 159, 1129-1136. (3) Sedlak, D. L.; Pinkston, K. E. Factors affecting the concentrations of pharmaceuticals released to the aquatic environment. J. Contemp. Water Res. Educ. 2001, 1, 56-65. (4) U.S. EPA. Biosolids generation, use, and disposal in the United States; 1999; http://www.epa.gov/epaoswer/non-hw/compost/ biosolid.pdf. (5) Capleton, A. C.; Courage, C.; Rumsby, P.; Holmes, P.; Stutt, E.; Boxall, A. B. A.; Levy, L. S. Prioritising veterinary medicines according to their potential indirect human exposure and toxicity profile. Toxicol. Lett. 2006, 163, 213-223. (6) Kumar, K.; Gupta, S. C.; Baidoo, S. K.; Chander, Y.; Rosen, C. J. Antibiotic uptake by plants from soil fertilized with animal manure. J. Environ. Qual. 2005, 34, 2082-2085. (7) Boxall, A. B. A.; Johnson, P.; Smith, E. J.; Sinclair, C. J.; Stutt, E.; Levy, L. S. Uptake of veterinary medicines from soils into plants. J. Agric. Food Chem. 2006, 54, 2288-2297. (8) Batchelder, A. R. Chlortetracycline and oxytetracylcine effects on plant growth and development in soil systems. J. Environ. Qual. 1982, 11, 675-678. (9) Migliore, L.; Civitareale, C.; Cozzolino, S.; Casoria, P.; Brambilla, G.; Gaudio, L. Laboratory models to evaluate phytotoxicity of sulphadimethoxine on terrestrial plants. Chemosphere 1998, 37, 2957-2961. (10) Migliore, L.; Cozzolino, S.; Fiori, M. Phytotoxicity to and uptake of enrofloxacin in crop plants. Chemosphere 2003, 52, 12331244. (11) Farkas, M. H.; Berry, J. O.; Aga, D. S. In American Chemical Society Book of Abstracts: Atlanta, Ga, 2006. (12) Gujarathi, N. P.; Haney, B. J.; Park, H. J.; Wickramasinghe, S. R.; Linden, J. C. Hairy roots of Helianthus annuus: A model system to study phytoremediation of tetracycline and oxytetracycline. Biotechnol. Prog. 2005, 21, 775-780. (13) Nagao, A.; Usui, K.; Shim, I. S. Comparison of the induction of glutathione s-transferase in plants by treatment with substrates. J. Weed Sci. Technol. 1997, 42, 340-348. (14) Pflugmacher, S.; Schroder, P.; Sandermann, H., Jr. Taxonomic distribution of plant glutathione s-transferases acting on xenobiotics. Phytochemistry 2000, 54, 267-273. (15) Hirata, T.; Ashida, Y.; Mori, H.; Yoshinaga, D.; Goad, L. J. A 37-kDa peroxidase secreted from liverworts in response to chemical stress. Phytochemistry 2000, 55, 197-202. (16) Anderson, J. V.; Davis, D. G. Abiotic stress alters transcript profiles and activity of glutathione s-transferase, glutathione peroxidase, and glutathione reductase in Euphorbia esula. Physiol. Plant. 2004, 120, 421-433. (17) Gujarathi, N. P., Linden, J. C. Oxytetracycline inactivation by putative reactive oxygen species released to nutrient medium of Helianthus annuus hairy root cultures. 2005, 92, 393-402. (18) Aga, D. S.; O’Connor, S.; Ensley, S.; Payero, J. O.; Snow, D.; Tarkalson, D. Determination of the Persistence of Tetracycline Antibiotics and Their Degradates in Manure-Amended Soil Using Enzyme-Linked Immunosorbent Assay and Liquid Chromatography-Mass Spectrometry. J. Agric. Food Chem. 2005, 53, 7165-7171. (19) Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248-254. (20) Mozer, T. J.; Tiemeier, D. C.; Jaworski, E. G. Purification and characterization of corn glutathione s-transferase. Biochemistry 1983, 22, 1068-1072. (21) Habig, W. H.; Pabst, M. J.; Jakoby, W. B. Glutathione STransferases. The first enzymatic step in mercapturic acid formation. J. Biol. Chem. 1974, 249, 7130-7139. (22) Schroder, P.; Juuti, S.; Roy, S.; Sandermann, H.; Sutinen, S. Exposure to chlorinated acetic acids. Responses of peroxidase
1456
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 4, 2007
(23) (24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
(38)
(39)
(40)
and glutathione S-transferase activity in pine needles. Environ. Sci. Pollut. Res. Int. 1997, 4, 163-171. Farkas, M. H.; Berry, J. O.; Aga, D. S. In Society of Environmental Toxicology and Chemistry: Baltimore, MD, 2005. Nepovim, A.; Podlipna, R.; Soudek, P.; Schroder, P.; Vanek, T. Effects of heavy metals and nitroaromatic compounds on horseradish glutathione s-transferase and peroxidase. Chemosphere 2004, 57, 1007-1015. Andrews, C. J.; Cummins, I.; Skipsey, M.; Grundy, N. M.; Jepson, I.; Townson, J.; Edwards, R. Purification and characterisation of a family of glutathione transferases with roles in herbicide detoxification in soybean (Glycine max L.); selective enhancement by herbicides and herbicide safeners. Pestic. Biochem. Physiol. 2005, 82, 205-219. Deng, F.; Hatzios, K. K. Characterization and safener induction of multiple glutathione S-transferases in three genetic lines of rice. Pestic. Biochem. Physiol. 2002, 72, 24-39. Smith, A. P.; DeRidder, B. P.; Guo, W. -J.; Seeley, E. H.; Regnier, F. E.; Goldsbrough, P. B. Proteomic analysis of arabidopsis glutathione S-transferases, from benoxacor- and copper-treated seedlings. J. Biol. Chem. 2004, 279, 26098-26104. DeRidder, B. P.; Dixon, D. P.; Beussman, D. J.; Edwards, R.; Goldsbrough, P. B. Induction of glutathione S-transferases in arabidopsis by herbicide safeners. Plant Physiol. 2002, 130, 1497-1505. Edwards, R.; Dixon, D. P.; Walbot, V. Plant glutathione stransferases: enzymes with multiple functions in sickness and in health. Trends Plant Sci. 2000, 5, 193-198. Durner, J.; Klessig, D. F. Inhibition of ascorbate peroxidase by salicylic acid and 2,6-dichloroisonicotinic acid, two inducers of plant defense responses. PNAS 1995, 92, 11312-11316. Passardi, F.; Cosio, C.; Penel, C.; Dunand, C. Peroxidases have more functions than a swiss army knife. Plant Cell Rep. 2005, 24, 255-265. Andrews, C. J.; Skipsey; Townson, J. K.; Morris, C.; Jepson, I.; Edwards, R. Glutathione transferase activities toward herbicides used selectively in soybean. Pestic. Sci. 1997, 51, 213-222. Eichhorn, P.; Aga, D. S. Identification of a photooxygenation product of chlortetracycline in hog lagoons using LC/ESI-Ion Trap-MS and LC/ESI-Time-of-Flight-MS. Anal. Chem. 2004, 76, 6002-6011. Marrs, K. A. The functions and regulation of glutathione S-transferases, in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1996, 47, 127-158. Davoine, C.; Douki, T.; Iacazio, G.; Montillet, J. L.; Triantaphylides, C. Conjugation of keto fatty acids to glutathione in plant tissues. Characterization and quantification by HPLCtandem mass spectrometry. Anal. Chem. 2005, 77, 7366-7372. Cysyk, R. L.; Parker, R. J.; Barchi, J. J.; Steeg, P. S.; Hartman, N. R.; Strong, J. M. Reaction of geldanamycin and C17-substituted analogues with glutathione: product identifications and pharmacological implications. Chem. Res. Toxicol. 2006, 19, 376381. Chard, B. K.; Doucette, W. J.; Chard, J. K.; Bugbee, B.; Gorder, K. Trichloroethylene uptake by apple and peach trees and transfer to fruit. Environ. Sci. Technol. 2006, 40 (15), 47884793. Rysz, M.; Alvarez, P. J. J. Amplification and attenuation of tetracycline resistance in soil bacteria: aquifer column experiments. Water Res. 2004, 38, 3705-3712. Peak, N.; Knapp, C. W.; Yang, R. K.; Hanfelt, M. M.; Smith, M. S.; Aga, D. S.; Graham, D. W. Abundance of six tetracycline resistance genes in wastewater lagoons at cattle feedlots with different antibiotic use strategies. Environmental Microbiology, In Press. Arthur, E.; Rice, P.; Rice, P.; Anderson, T.; Baladi, S.; Henderson, K.; Coats, J. Phytoremediation - An overview. Crit. Rev. Plant Sci. 2005, 24, 109-122.
Received for review July 12, 2006. Revised manuscript received November 16, 2006. Accepted December 5, 2006. ES061651J
