SEARCH

SEARCH BY CITATION

Keywords:

  • Antimicrobials;
  • Triclosan;
  • Seed germination;
  • Plant morphology;
  • Methyl-triclosan

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Three wetland macrophytes, Sesbania herbacea, Bidens frondosa, and Eclipta prostrata, were exposed (0.4–1,000-ppb nominal concentrations) to the antimicrobial triclosan for 28 d in a flow-through system. Sesbania herbacea had decreased seed germination at the 100-ppb exposure level at days 7, 14, and 21, and B. frondosa germination was reduced at the 1,000-ppb exposure level at day 7. Eclipta prostrata germination was unaffected. Seedling effects monitored were total fresh weight, shoot and root fresh weights, root length, and root surface area. Root metrics were most affected by exposure. Total root length was diminished at all exposure levels in S. herbacea and B. frondosa and at the 10-ppb and higher concentrations for E. prostrata. Root surface area decreased at all exposure levels in B. frondosa and at the 10-ppb level and above in S. herbacea and E. prostrata. Root and shoot bioconcentration factors (BCFs) were estimated for S. herbacea and B. frondosa. While BCFs were low in shoots of both species and roots of S. herbacea (<10), they were elevated in B. frondosa roots (53–101). Methyl-triclosan was formed in the system and accumulated in shoot and root tissues of S. herbacea to concentrations that exceeded those of the parent compound. However, methyl-triclosan was nontoxic in an Arabidopsis thaliana enoyl-acyl carrier protein reductase (the putative enzymatic target of triclosan) assay and did not appear to contribute to the effects of exposure. Two of the three plant species assessed exhibited reduced root systems at environmentally relevant concentrations, raising the concern that wetland plant performance could be compromised in constructed wetlands receiving wastewater treatment plant discharges.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

North Texas (USA) is typical of many urbanized areas in which plans for management of future drinking water supplies include reuse of water from sources that receive municipal effluent. Many of the these plans include pretreatment of effluent-dominated sources through natural or constructed wetlands to reduce risk of human exposure to pharmaceuticals and personal care products (PPCPs) released by wastewater treatment plants (WWTPs), yet there is little known of the implications of such measures [1,2]. Improved understanding of the accumulation and effects of PPCPs in wetland ecosystems will be required to ensure sustainable long-term success of these water resource management plans.

Triclosan (TCS; 5-chloro-2-[2,4-dichlorophenoxy]phenol) is an antimicrobial additive in a wide range of personal care products and as such is typical of down-the-drain PPCPs. Several studies have reported that TCS enters WWTPs at low μg/L concentrations and is largely but incompletely removed, resulting in typical final effluent concentrations in the range of 100 to 400 ng/L in activated sludge systems with somewhat less efficient removal by trickling filter systems [3–7]. Despite the relative efficiency of WWTP removal, TCS has been reported to be one of the most frequently detected PPCPs in surface waters [8]. Agricultural application of sewage sludge represents an additional source of environmental exposure for plants.

Triclosan partitions to sludge during wastewater treatment resulting in elevated mg/kg concentrations [5,9–12].

Methyl-triclosan (MTCS; 5-chloro-2-[2,4-dichlorophenoxy] anisole) is an environmental metabolite of TCS with increased hydrophobicity [13–15]. Triclosan and methyl-triclosan have been shown to have bioaccumulation factors of several orders of magnitude in animals [13,16] and algae [17] and have been suggested as marker compounds for the environmental distribution of WWTP organic contaminants [13,17] and microbes [18].

In addition to serving as marker compounds, there is concern over the widespread distribution of antimicrobials and their potential environmental effects. Triclosan has a specific mode of action inhibiting fatty acid synthesis in prokaryotes and plants at the enoyl-acyl carrier protein (ACP) reductase (FabI) step [19,20]. To our knowledge, the effect of MTCS exposure on this enzymatic pathway has not been investigated. A number of other potential modes of TCS toxicity have been investigated in animals [21–24]. Thus, development of microbial resistance to TCS and possible cross-resistance to antibiotics [25] as well as susceptibility of plants [17] and animals represent potential environmental risks. Although ecological risk evaluations of TCS have generally shown little threat for aquatic animals, aquatic plants have been considered more vulnerable [1,26,27]. These assessments have been based on TCS only and have not considered the possible contributions of MTCS or triclocarban (TCC). Triclocarban is an antimicrobial additive used primarily in solid bar soaps that is less efficiently removed by WWTPs than TCS and has received relatively little attention until recently. Triclocarban is now known to co-occur with TCS, typically at somewhat higher concentrations and shares a similar mode of action [16–18,28–33]. Thus, risk assessments based on TCS only may underestimate the combined effects of antimicrobials and their metabolites, especially those risks to the aquatic plant community.

Previously, we reported on seasonal TCS concentrations in influent and effluent from a North Texas activated sludge WWTP [7] and the bioconcentration of TCS, MTCS, and TCC in aquatic algae [17] and snails [16] in an effluent-receiving stream. As part of our continuing study into the environmental effects of PPCP, here we report results of laboratory flow-through exposure of three wetland macrophyte species (Sesbania herbacea, Bidens frondosa, and Eclipta prostrata) to TCS nominal concentrations ranging from 0.4 to 1,000 ppb. The 0.4-ppb exposure concentration was selected as a conservative estimate of the sum of TCS and TCC concentrations in WWTP effluent observed in our earlier studies. The higher concentrations represent exaggerated exposures, the lower range of which might be typical of pore-water concentrations under some conditions. Seed germination rates and root and shoot system morphology were evaluated for exposure effects on all three species. Bioconcentration of TCS and MTCS in seedling root and stem tissues was evaluated for the two larger species (S. herbacea and B. frondosa). Finally, because MTCS was found to accumulate in the TCS-exposed plants, the effects of acute TCS and MTCS exposure on ACP-reductase activity in the model plant Arabidopsis thaliana were compared.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Plants

Three rooted emergent wetland plant species were used in the present study: S. herbacea (Mill.) McVaugh, B. frondosa L., and E. prostrata (L.) L. These are common herbaceous, annual plants in north-central Texas wetlands and have characteristics amenable to laboratory studies (high germination percentage, uniform and early germination times, and robust seedlings). Seeds were collected from established populations on the Elm Fork of the Trinity River, in the Greenbelt Corridor between Lake Ray Roberts and Lake Lewisville (Denton, TX), in the fall of 2007 and stored in the laboratory until use.

Chemicals

Neat TCS was purchased from Fluka. Internal standards 13C12 TCS, 13C12 MTCS, and unlabeled MTCS were purchased from Wellington Laboratories. All other chemicals were purchased from Fisher Scientific.

Flow-through exposure system

A flow-through exposure system was designed to establish each of six exposure solutions (control, solvent control, and TCS nominal concentrations of 0.4, 10, 100, and 1,000 ppb). Each delivery system consisted of one channel from a 12-channel peristaltic pump (Carter 12/6, Cassette pump) carrying laboratory-deionized water diluent at a flow rate of 342 ml/h amended by a flow of 240 μl/h from one channel of a 10-channel syringe pump (kdScientific, Model 200 series) carrying TCS, methanol as a carrier solvent, and additional deionized water diluent (control channels received no syringe pump flow; solvent control channels received only methanol and deionized water). Flow from the two pumps was combined in individual 25-ml mixing chambers placed on continuously operated stir plates. After mixing, each of the exposure solutions was delivered to two plastic exposure trays plumbed in series. Retention time in the exposure trays was 4.42 h. Exposure trays were constructed from standard non-draining potting trays (54 × 28 × 6 cm, Summit Plastic Company) with a 0.6-cm drain placed on the side of the tray 1 cm from the tray bottom. The first exposure tray was used for seed germination trials. Twenty seeds of each species were sown on the surface of filter paper in covered Petri dishes (5-cm diameter). Each filter-paper disc in the Petri dishes communicated with the exposure solution in the tray via a 1-cm-wide filter-paper wick. Four replicate Petri dishes were placed in each exposure tray for each test species. The second exposure tray was used for assessment of seedling growth and development and received exposure solution that drained gravimetrically from the first exposure tray. The seedlings were placed in individual cells in seedling line inserts (4.3 × 4.3 × 6 cm; Dillen Products) with approximately 40 g of glass beads as a growth substrate (0.5-mm diameter; Clemtex). Exposure solution was delivered to seedling root systems by capillary action. Dilution rates for TCS were first estimated for the nominal target concentrations. After a 14-d equilibration period, actual TCS system concentrations were measured. The entire system was maintained in a growth room with a fluorescent light illumination system providing an average of 146 μmol/s/m (seed germination trays) and 459 μmols/m2 (seedling growth trays) light energy on a 14:10 light:dark cycle at an average temperature of 23°C.

Germination, characterized by radicle emergence, was monitored daily for four weeks. Immediately following radicle emergence, seedlings were transferred to pots for assessment of growth and development. Seedling transplant continued until each replicate pot had a viable seedling. Transplanting was completed within a 3-d period. Plants were exposed to treatment conditions for 28 d and then harvested for morphometric and chemical residue analyses.

Harvesting and morphometric analyses

To prevent cross contamination among treatments, seedlings were harvested first from the control and then sequentially from the lowest to the highest TCS concentrations. All plants were harvested within a 12-h period. Plants were transferred from the growth room to the laboratory, shoot heights were obtained, and plants were photographed. Plants were then removed from their pots, and the glass bead substrate was gently washed off. Roots were rinsed thoroughly with deionized water in order to remove surface-sorbed TCS and MTCS, separated from shoots, and blotted dry with paper towels, and the fresh weight of each was obtained. Individual root systems were placed in water-filled clear plastic trays and scanned at 400 dpi using an Epson Expression 10000XL (Seiko Epson) with transmitted light. Total root length, volume, and surface area were obtained following analysis of the digitized root system using WinRHIZO PRO (ver 2007c, Regent Instruments). Once processed for morphometric analysis, root and shoot systems were collected in plastic vials. Triclosan and methyl-triclosan internal standards were added, and samples were stored at 4°C for bioconcentration study.

Tissue preparation for chemical residue analyses

Triclosan and methyl-triclosan concentrations for root and shoot tissues were determined for individual plants. The amount of tissue available for analysis varied among individuals and treatments and, to a more significant degree, among species (range of fresh weight for root samples was 0.024–0.68, 0.003–0.129, and 0.0002–0.022 g and for shoot samples was 0.038–0.204, 0.012–0.080, and 0.001–0.006 g for S. herbacea, B. frondosa, and E. prostrata, respectively). Tissue analysis was limited to larger specimens of the two larger species, S. herbacea (∼100 mg) and B. frondosa (∼50 mg). Tissue extraction and isotope dilution gas chromatography/mass spectrometry (GC/MS) analysis were similar to methods previously described [17]. Because S. herbacea were to be used for lipid quantification and had higher individual masses, samples were processed somewhat differently than B. frondosa. Undamaged S. herbacea tissue was carefully placed into a 3-ml polypropylene vial containing one-third volume of 2.5-mm glass beads. Two milliliters of 70°C isopropyl alcohol (IPA) were added, and vials were placed in a water bath at 70°C for 30 min to halt enzyme activity. The samples were then homogenized using a MiniBeadbeater-8™ (Biospec Products) and placed back in the water bath. Ten microliters each of 5 ng/μl 13C12 TCS and 5 ng/μl 13C12 MTCS (internal standards) were added to the vials. The contents of the vials were transferred to a 25-ml tube with three rinses. The tubes were sealed with a polytetrafluoroethylene-lined cap, vortexed, and held overnight at 4°C. Samples were then vortexed and centrifuged for 5 min at 1,500 rpm. The extracts were then decanted into a clean 25-ml tube (transfer process was conducted three times with 2 ml of 2:1 IPA:CHCl3). Aqueous back-extraction and IPA removal were conducted with repeated 3-ml aliquots of 1 M KCl until the aqueous layer was clear and no additional volume changes occurred in the organic layer. The organic layer was reduced under nitrogen to less than 0.5 ml and filtered through a 0.22-μm Polyvinylidene fluoride single-use syringe filter into preweighed sample vials where they were evaporated to dryness to determine lipid mass and reconstituted in 100 to 500 μl dichloromethane (DCM) for GC/MS analysis. Bidens frondosa were analyzed as described previously, except the initial extraction solvent was hexane:ethyl acetate 1:1 and the 1 M KCl aqueous back-extraction was omitted. Bioconcentration factors (BCF) for root and stem tissues were calculated as the ratio of measured fresh tissue concentrations to measured exposure water concentrations.

Exposure water preparation for chemical residue analysis

Small samples (20 ml for controls, solvent controls, and the TCS 0.4-μg/L exposure treatments; 1 ml for the higher exposure treatment concentrations) were collected from the outflows of both the seed germination trays and the seedling growth trays at approximately even intervals (six times) throughout the course of the experiment. Internal standards were added as described previously and were subjected to liquid/liquid extraction with DCM. After the extraction, the samples were evaporated under nitrogen and reconstituted to 100 μl with DCM for GC/MS analysis.

Instrumental analysis

Triclosan and methyl-triclosan analyses were conducted on an Agilent 6890 GC coupled with a 5973 mass selective detector (MS, 70 eV). The MS was operated in single ion monitoring mode. Gas chromatography conditions were helium carrier gas at 480 hPa, inlet temperature at 260°C (2 μl, pulsed pressure at 1,700 hPa for 0.5 min, splitless injection), and column (Alltech; EC-5 30 m, 0.25-mm inner diameter, 0.25-μm film) temperature initially at 40°C with a 1-min hold followed by a 50°C/min ramp to 140°C with a 5-min hold followed by a 10°C/min ramp to 300°C with a final 17-min bake-out. A standard curve was constructed from 10 to 1,000 pg/μl injected analytes and 500 pg/μl internal standards. Practical quantitation limits (PQL) were based on the lowest concentration from the standard curve and corresponding minimum masses and volumes used for unexposed samples. Quality control included analysis of blanks, blank spikes, matrix spikes, and matrix spike duplicates for both tissue and water samples. Bidens frondosa tissues were used for matrix spike analysis.

Enoyl reductase assay

A modified protocol based on that described in Serrano et al. [34] was used to assess the acute effects of TCS and MTCS exposure (2 h) on enoyl-ACP reductase enzyme activity in the model plant A. thaliana. Total protein was extracted from healthy adult plant tissues using a 10-mM sodium phosphate buffer at a pH of 9, along with 1 mM dithiothreitol. Protein concentration was determined by the Bradford assay using a commercial kit (Bio-Rad Laboratories). Protein (50 μg) was then dissolved in reaction buffer (10 mM sodium phosphate at pH 6.2 and 140 μM nicotinamide adenine dinucleotide [NADH]). Crotonoyl-CoA (120 μM; Sigma) was used as substrate. Total reaction volume was 1.0 ml. The reaction was initiated after addition of substrate and continuously monitored for at least 25 min at room temperature (25°C). Enoyl-ACP reductase activity was monitored by the decrease in absorbance at 340 nm due to oxidation of NADH. Changes in rate were calculated as a percentage change relative to rates in solvent control samples rates for each minute monitored (n ≥ 25).

Statistical analyses

Triclosan and methyl-triclosan concentrations in shoots and roots were compared only between treatments in which concentrations exceeded PQLs. In this case, multiple comparisons were conducted using Proc Mixed and the Tukey-Kramer method in SAS® (SAS Institute). Germination rates were compared for each species separately using repeated-measures analysis and Proc Mixed in SAS. Treatments were compared to controls using contrast statements. To meet assumptions of normality and equal variance, the percentage of seeds germinating was square-root transformed for B. frondosa, logarithmically transformed for S. herbacea, and squared for E. prostrata. Shoot and root fresh weight, ratio of shoot to root fresh weight, total root length, and root surface area were compared within species using a one-way analysis of variance with Proc Mixed in SAS, and treatments were compared to the controls using Dunnett's tests. To meet assumptions of normality and equal variance, all end points for E. prostrata were logarithmically transformed; total weight, root weight, shoot:root ratio, and root surface area were square-root transformed, while total root length was logarithmically transformed for S. herbacea; total weight, shoot weight, shoot:root ratio, and total root length were logarithmically transformed, while root weight and root surface area were square-root transformed for B. frondosa. The relationship between plant responses and TCS and MTCS bioconcentration in roots and shoots was investigated using stepwise multiple regression procedures in SAS's Proc Reg. Enoyl-reductase activity was compared between treatments and controls using InStat version 3.06 (GraphPad Software). For all figures, means are presented with standard error calculated from untransformed data. Differences among means were considered statistically significant at p < 0.05.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Analytical quality control and measured exposure concentrations

Analytical quality control results (Table 1) indicated good recovery of known spike concentrations from clean B. frondosa and no blank concentrations above the PQLs. Under the conditions of analysis, underivatized TCS exhibits peak tailing that can reach unacceptable levels when the GC inlet is not maintained properly. Frequent inlet liner replacement is required to preserve quantitative results. This is especially important when analyzing shoot samples with substantial chlorophyll content. Even though the flow-through system had equilibrated for 14 d, measured exposure concentrations were less than half the nominal concentrations at the two highest exposures (Table 2). Also, measured concentrations in the second tray (used for seedling exposure) of the two-tray series tended to be lower than those in the first tray (used for seed exposure). This decline in concentration through the flow-through system ranged from 14 to 25% and appeared to be independent of exposure concentration. Methyl-triclosan in exposure water was below PQL (0.02 ppb) for all treatments. The six measured TCS concentrations were taken at approximately even intervals throughout the course of the experiment and showed no apparent relationship with time.

Table Table 1.. Quality control data for triclosan (TCS) and methyl-triclosan (MTCS) in exposure water, root, and shoot tissues of Biden frondosa. Spike additions were 50 ng for each sample. TCS = triclosan, MTCS = methyl-triclosan
 Blank (ppb)Mean blank spike recovery (%)Mean matrix spike recovery (%)Replicate matrix spike deviation (%)Practical quantitation limits (ppb)
Water
TCS<0.05 (n = 4)114 (n = 4)92.4 (n = 5)38.80.05
MTCS<0.02 (n = 4)111 (n = 2)87.7 (n = 2)3.90.02
Roots
TCS<10.0 (n = 10)124 (n = 2)127 (n = 4)4.010
MTCS<10.0 (n = 10)99.0 (n = 2)99 (n = 4)3.410
Shoots
TCS<10.0 (n = 10)124 (n = 2)139 (n = 4)8.310
MTCS<10.0 (n = 10)99.1 (n = 2)99.9 (n = 4)10.210

Tissue concentrations and bio concentration factors

Samples of root and shoot tissues were analyzed for TCS and MTCS in S. herbacea and B. frondosa. Eclipta prostrata seedlings yielded too little tissue over the duration of this experiment to permit analysis. At the lowest levels of TCS exposure, TCS concentrations in shoot and root tissues were below PQL. At higher levels of exposure, TCS levels in tissues tended to increase with increasing exposure concentrations in both S. herbacea and B. frondosa in roots and shoots (Fig. 1).

Methyl-triclosan concentrations showed a similar pattern in S. herbacea but were much greater than TCS concentrations in roots and shoots (Fig. 2) at the higher exposure concentrations. Methyl-triclosan concentrations in B. frondosa were below PQL for shoots at all concentrations and, surprisingly, were measurable in root tissues only at the intermediate exposure concentrations of 10 and 100 ppb (Fig. 2). Thus while MTCS root concentrations at the 10-ppb nominal exposure were similar for S. herbacea and B. frondosa, they continued to increase at higher exposures in S. herbacea, while they declined at higher exposure concentrations in B. frondosa.

Table Table 2.. Nominal and measured triclosan (TCS) concentrations (ppb) in the exposure trays averaged over the course of the experiment. Data are shown as means ± one standard error (n = 6)
 Measured concentration
Treatment and target concn.First tray (seed germination)Second tray (seedling growth)
Control<0.05<0.05
Solvent<0.05<0.05
TCS 0.4 ppb0.8 ± 0.150.6 ± 0.10
TCS 10 ppb9.9 ± 2.87.8 ± 2.1
TCS 100 ppb55 ± 4.147 ± 3.3
TCS 1,000 ppb440 ± 58.9356 ± 52.8
thumbnail image

Figure Fig. 1.. Triclosan (TCS) levels in Sesbania herbacea and Bidens frondosa seedlings grown for four weeks under exposure to four nominal concentrations of TCS (0, 0.4, 10, 100, and 1,000 ppb). Data are shown as means ± one standard error. Within an organ system (shoot or root), similar lowercase letters identify treatment means that do not significantly differ (Tukey-Kramer p > 0.05). PQL = practical quantitation limits.

Download figure to PowerPoint

Bioconcentration factor estimates were based on measured water and tissue concentrations of TCS (Table 3). No BCF estimates were calculated for bioconcentration levels below PQL or for MTCS since it was not detected in the exposure water. Bioconcentration factors for shoot TCS concentrations were low for both species (maximum of 2.8) and unrelated to exposure concentration. Bioconcentration factors for S. herbacea roots were also low (1.4–2.8) but were markedly elevated in B. frondosa (53–102).

thumbnail image

Figure Fig. 2.. Methyl-triclosan (MTCS) levels in Sesbania herbacea and Bidens frondosa seedlings grown for four weeks under exposure to four nominal concentrations of TCS (0, 0.4, 10, 100, and 1,000 ppb). Data are shown as means ± one standard error. Within an organ system (shoot or root), similar letters identify treatment means that do not significantly differ (Tukey-Kramer p > 0.05). PQL = practical quantitation limits.

Download figure to PowerPoint

Seed germination

Seed germination began within 24 h of seeding. The number of seeds germinating increased over time and generally peaked by the third week; few seeds germinated after three weeks (Fig. 3). No significant effects of TCS exposure on seed germination were observed for E. prostata, and by day 28, 90.0 ± 1.9% of all seeds germinated (Fig. 3). Sesbania herbacea seed germination was significantly lower in the 100-ppb treatment compared to the controls at day 7 (16.09 ± 3.7% vs 31.3 ± 4.1%), day 14 (19.4 ± 3.8% vs 34.4 ± 4.3%), and day 21 (19.4 ± 3.8 vs 34.4 ± 12.1%); however, by day 28, there were no significant differences among treatments. Despite the depression of germination rates seen at 100-ppb exposure level, no significant differences in germination were observed at 1,000 ppb. Overall, 31.3 ± 1.8% of S. herbacea seeds germinated by day 28 (Fig. 3). Bidens frondosa seed germination in the 1,000-ppb treatment was significantly lower than controls at day 7 (65.0 ± 6.9% vs 83.8 ± 3.9%); however, there were no significant effects of TCS exposure on germination at subsequent sampling periods, and by day 28, 96.0 ± 0.8% of seeds had germinated (Fig. 3).

Morphometric measurements

Although there were no detectable effects of the TCS exposure on total fresh weight of B. frondosa, total fresh weight of S. herbacea was significantly lower than the controls at 1,000 ppb, while total fresh weight of E. prostrata was significantly lower than the controls at 10, 100, and 1,000 ppb (Fig. 4). Shoot fresh weight of E. prostrata was not significantly affected by TCS exposure; however, shoot fresh weight was significantly greater in the 100-ppb treatment compared to the controls for S. herbacea and significantly greater in the 100- and 1,000-ppb treatments compared to the controls for B. frondosa (Fig. 4). While B. frondosa root fresh weight was not significantly affected by TCS exposure, S. herbacea root fresh weight was significantly lower in the 100- and 1,000-ppb treatments compared to the controls, and E. prostrata root fresh weight was significantly lower in the 10-, 100-, and 1,000-ppb treatments compared to the controls (Fig. 4). Triclosan exposure affected patterns of resource allocation between shoots and roots (Fig. 4); S. herbacea and E. prostrata shoot:root ratios were significantly greater in the 100- and 1,000-ppb treatments compared to the controls, while B. frondosa shoot:root ratios were significantly greater in the 1,000-ppb treatment compared to the controls. Triclosan exposure had a significant effect on total root length and root surface area for all species (Fig. 5). Total root length was significantly lower than controls at all levels of TCS exposure for S. herbacea and B. frondosa and was significantly lower at 10, 100, and 1,000 ppb compared to the controls for E. prostrata (Fig. 5). Sesbania herbacea and E. prostrata root surface area was significantly lower than controls at 10, 100, and 1,000 ppb, while B. frondosa root surface area was significantly lower than the controls at 0.4, 100, and 1,000 ppb (Fig. 5). The relationship between root morphology and exposure concentration was readily apparent on visual inspection for all plant species. Since root system morphology was similar for all species, scanned images are presented only for S. herbacea (Fig. 6).

Table Table 3.. Concentration of triclosan (TCS) in water and tissues and bioconcentration factors (BCF) in Sesbania herbacea and Biden frondosa root and shoot samples obtained from seedlings grown for four weeks under exposure to three concentrations of TCS (10, 100, and 1,000 ppb). No values are shown for concentrations below practical quantitation limit
 S. herbaceaB. frondosa
Exposure level (ppb)Water concn. (ppb)Concn. in shoots (ppb)BCFConcn. in roots (ppb)BCFConcn. in shoots (ppb)BCFConcn. in roots (ppb)BCF
108.925.12.8    901101
10054.938.10.71512.8  4,93690
1,0003961560.45401.497.70.2521,05453
thumbnail image

Figure Fig. 3.. Seed germination of three wetland plant species (Eclipta prostrata, Sesbania herbacea, and Bidens frondosa) grown for four weeks under exposure to four nominal concentrations of TCS (0, 0.4, 10, 100, and 1,000 ppb). Data shown as means ± one standard error. The asterisk (*) indicates significant differences between treatments and controls (p < 0.05). All treatments were assessed at the same time at weekly intervals; data have been staggered for legibility.

Download figure to PowerPoint

Regression analyses: TCS, MTCS, and morphometric measurements

Regression analyses conducted on the relationships among root concentrations and root morphometric measurements indicated that changes in root metrics were almost completely explained by root TCS concentrations, while MTCS concentrations had little influence. Sesbania herbacea root weight was negatively related to root TCS concentration (r2 = 0.78; p < 0.05), while regressions based on root MTCS concentration and the sum of root TCS and MTCS were not as pronounced (r2 = 0.69 and 0.69, respectively; p > 0.05). In B. frondosa, TCS root concentrations were most strongly related to decreased root area (r2 = 0.94; p < 0.05), while there was no significant relationship between MTCS and root area (r2 = 0.02; p > 0.05). The sum of TCS and MTCS only modestly improved the strength of the relationship over that of TCS only (r2 = 0.95; p < 0.05).

Enoyl A CP reductase assays

Results of duplicate assays for acute effects of TCS and MTCS on enoyl ACP reductase enzymatic activity in A. thaliana are presented in Figure 7. Depressions in activity at 1 and 10 ppm were greater than those observed at 0.1 ppm but exhibited no dose-response relationship at these higher concentrations. In contrast, MTCS exposures to the same range of concentrations resulted in no significant depression relative to the solvent control. Methylation of TCS eliminated toxicity to enoyl ACP reductase enzymatic activity at the concentrations tested.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Performance of the experimental system

Challenges associated with trace-level exposure to toxicants often preclude use of static exposure systems. This can be especially true of phenolic toxicants such as TCS. The flow-through system used in the present study resulted in measured exposure concentrations that were relatively consistent and exhibited no trend toward increasing or decreasing concentration over the four-week exposure period, indicating that sorption dynamics had largely stabilized during the initial equilibration period. The degree to which the measured concentrations departed from the targeted nominal concentrations underscores the requirement for analytical confirmation of exposure concentrations under these conditions. Once set up, the system was automated and reliable and required little maintenance.

Formation of MTCS

Methyl-triclosan was not detected in the exposure water analyses. Volumes used for these analyses were selected for detection of TCS at the expected nominal concentrations. Although the analytical procedures had an estimated PQL for MTCS that was approximately half that of TCS for aqueous samples, the volumes used were not sufficient to permit detection of the formation of MTCS at levels that would be indicative of trace metabolite formation by microbial activity in the exposure water. However, if MTCS had been formed at concentrations that even remotely approximated those indicated by the ratio of MTCS/TCS concentrations found in WWTP effluent [17], it would have been easily detected. It was concluded, therefore, that there was no significant exposure to MTCS originating from the exposure water per se. Nevertheless, MTCS concentrations in roots exceeded TCS concentrations by as much as two orders of magnitude in S. herbacea seedlings and increased with increasing TCS exposure. Bidens frondosa MTCS root concentrations were similar to S. herbacea at the 10-ppb nominal exposure concentration but declined at increasing TCS concentrations rather than continuing to increase as was the case for S. herbacea. This difference between the two species is interesting and may be related to the mechanisms of formation of MTCS. Microbial methylation of TCS to form MTCS is well documented [5,13,14,35] and could have occurred to a greater degree in the root zone of S. herbacea than in B. frondosa. Perhaps the microbial community characteristic of B. frondosa is more susceptible to elevated TCS concentrations than the community that formed in the S. herbacea root zone. Also, plants are known to be capable of endogenous O-methylation of xenobiotics [36]. Perhaps this endogenous capability for formation of MTCS is more susceptible to TCS toxicity in B. frondosa than in S. herbacea. Methyl-triclosan concentrations in S. herbacea shoots also exceeded TCS concentrations at the two highest exposure concentrations (concentrations were below PQL at lower exposures), but the dominance of the metabolite was much less marked than seen in the roots. Methyl-triclosan in concentrations in B. frondosa shoots were below PQL in this much smaller plant.

thumbnail image

Figure Fig. 4.. Total fresh weight, shoot fresh weight, root fresh weight, and the ratio of shoot to root fresh weight of three wetland plants grown for four weeks under exposure to four nominal concentrations of TCS (0, 0.4, 10, 100, and 1,000 ppb). Data shown as means ± one standard error. The asterisk (*) indicates significant differences between treatments and controls (p < 0.05).

Download figure to PowerPoint

thumbnail image

Figure Fig. 5.. Total root length and root surface area of three wetland plants grown for four weeks under exposure to four nominal concentrations of triclosan (0, 0.4, 10, 100, and 1,000 ppb). Data shown as means ± one standard error. The asterisk (*) indicates significant differences between treatments and controls (p < 0.05).

Download figure to PowerPoint

Although the elevated MTCS concentrations described above may represent concern for its biomagnification potential, we did not see evidence of its contribution to toxicity. Multiple regression analyses indicated that TCS concentrations accounted for essentially all the observed decline in root structure metrics with little improvement when MTCS concentrations were included in the regressions. Additionally, when TCS and MTCS were evaluated for direct inhibition of enoyl reductase activity in A. thaliana, the potent TCS toxicity observed here and in previous studies [34] was completely absent with MTCS. Finally, in B. frondosa at elevated TCS exposures, we observed declining root metrics with increasing TCS concentration even though MTCS concentrations were declining, indicating that morphometric effects were largely tracking TCS concentrations rather than MTCS concentrations.

thumbnail image

Figure Fig. 6.. Scanned roots of Sesbania herbacea grown for four weeks under exposure to four nominal concentrations of triclosan: (a) = 0, (b) = 0.4, (c) = 10, (d) = 100, and (e) = 1,000 ppb. Scale bar = 10 cm.

Download figure to PowerPoint

thumbnail image

Figure Fig. 7.. Percentage changes in enoyl reductase enzyme activity using four-week-old Arabidopsis thaliana. Exposure treatments expressed as percentage of mean solvent control response (n > 25). Assays were conducted on two replicate plants (a and b) for each level of triclosan (TCS) exposure. Similar letters identify treatment means that do not significantly differ (Tukey-Kramer p > 0.05). MTCS = methyl-triclosan.

Download figure to PowerPoint

Bioconcentration

Plants can play a pivotal role in remediation of soils and waters contaminated with organic pollutants through direct uptake of contaminants, transformation, and accumulation into various plant tissues. Previous studies regarding plant uptake of organic compounds have focused on polychlorinated hydrocarbons (polychlorinated biphenyl and trichloroethylene), polycyclic aromatic hydrocarbons (phenanthrene, anthracene, and pyrene), and explosives (2,4,6-trinitrotoluene and hexahydro-1,3,5-trinitro-1,3,5-triazine) [37,38]. These studies have revealed that the quantity and efficiency of uptake and distribution within living plants are determined by a large number of factors, including contaminant concentration, contaminant properties (i.e., water solubility, octanol-water partition coefficient [KOW], ionization, and molecular weight), plant characteristics (i.e., root anatomy and morphology, enzymes in plants, and lipid content), and environmental characteristics (temperature, pH, organic matter, soil moisture, and soil particle size distribution) [38–44]. Generally, it has been shown that lipophilic organic compounds (log KOW > 3) are poorly translocated into aboveground tissues and are concentrated in belowground tissues [45], while more water-soluble organic compounds are more readily translocated into aboveground plant parts [46–48]. The results of the present study are in agreement with previous studies. Triclosan has a log KOW of 4.8 [49], and in all plant species and treatments except S. herbacea seedlings exposed to 10 ppb TCS, bioconcentration in roots exceeded shoot bioconcentration. The basis for the discrepancy at 10 ppb for S. herbacea is not presently known. Additional studies that assess the relationships among plant physiology, anatomy, morphology, uptake, transformation, and accumulation may be required to understand the phylogenetic differences in BCFs observed in the present study.

Seed germination

Seeds and seed banks are crucial to the reestablishment of vegetation following disturbance (fire, overgrazing, flooding, drought, and so on) [50], and the importance of seed banks in reestablishing wetland plant communities following flooding and drought is well established [51–53]. The three wetland plant species assessed in the present study differed in their response to TCS exposure. While seed germination in E. prostrata was not significantly affected by TCS exposure, B. frondosa displayed a lower germination at week 1 in the 1,000-ppb treatment, while germination was lower at weeks 1 to 3 for S. herbacea seeds exposed to 100 ppb TCS. The basis for interspecific differences in response to TCS exposure and the reduction in germination rates in S. herbacea seeds exposed to 100 ppb but not 1,000 ppb is unclear. Although significant differences in germination rates did not persist and were not detected at week 4, this should not be interpreted as implying that community-level effects would not ensue. Early seedling establishment imparts a significant competitive advantage, and differences in germination of 15 d [54] or 1 or 2 d [55] can substantially impact seedling survival when seedlings are competing for resources.

Morphometrics

Although total fresh weight was not significantly affected by TCS exposure in B. frondosa, it was lower at the highest level of exposure for S. herbacea and at the 10- to 1,000-ppb levels of exposure for E. prostrata. No nutrient amendments were made in the present study. The water source was deionized water, and since the potting media consisted of silica beads, an inert substrate, we interpret the results as not being attributable to a reduced capacity for nutrient foraging but rather reflective of a reduced capacity to utilize storage reserves in the seed for plant growth. In agreement with a lack of a detectable response to TCS exposure below 100 ppb for S. herbacea and B. frondosa, Orvos et al. [56] found no detectable effect of TCS on Lemna gibba biomass in exposure concentrations up to 62.5 ppb, their maximum exposure concentration. The significant effect on total biomass in the 10-ppb treatment for E. prostrata, however, identifies the presence of species-specific responses to TCS and underscores the necessity for toxicity assessments that are multispecies in nature to adequately reflect potential community-level responses.

The functional equilibrium theory of shoot:root ratios states that plants maintain a balance between shoot and root growth reflective of resource availability [57]. Corroborating studies have shown that when belowground resources limit plant growth, belowground growth will take precedence over aboveground growth, and, conversely, when aboveground resources are limiting, aboveground growth predominates [58–60]. Thus, altered patterns of resource allocation between shoots and roots are generally associated with changing levels of resource availability; in wetland plants a reduced root system in response to increased levels of water availability is considered an adaptation to wetland habitats [61,62]. In the current study, however, patterns of resource allocation were not associated with levels of resource availability but with TCS exposure. The three species included in the present study exhibited reduced patterns of resource allocation to roots, although this was restricted to the highest levels of exposure. In contrast, total root length and root surface area were more sensitive to TCS exposure; total root length and/or root surface area was significantly lower than the controls at all levels of exposure for S. herbacea and B. frondosa and at the 10- to 1,000-ppb concentration for E. prostrata. Reduced root length with increasing levels of TCS exposure has also been found for Oryza sativa and Cucumis sativus [63], although direct comparisons are limited since Liu et al. focused on exposure under terrestrial conditions. Nevertheless, a trend toward a reduction in root system morphology on exposure to TCS has been shown to exist in terrestrial and wetland plant species. Plant root systems are the principal organs for nutrient acquisition and anchorage, functions that are highly dependent on the extent of the root system [64–67]. Therefore, any stressors that affect root system morphology have the potential to influence these functions. For example, reduced root length may lead to increased potential for uprooting [66], which may be of particular relevance for plants inhabiting areas with frequent flood pulses. Furthermore, a reduced root system limits the area in which a plant can forage for resources, impacting a plant's competitive ability [64,68], and influences the ecosystem services they provide. The fact that two of the three wetland species assessed in this study exhibited a reduced root system at environmentally relevant levels of TCS exposure emphasizes the need for more comprehensive investigation into the effects of TCS on plant performance in wetland habitats.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Although TCS is a ubiquitous compound found throughout North American watersheds, there is little understanding of the environmental impacts associated with TCS exposure. We are aware of no other study that has examined the effects of TCS exposure on wetland vascular plants beyond the model genus Lemna. Despite being limited to an assessment of only three wetland plant species, the present study has revealed that TCS bioaccumulates in plant tissues, that TCS exposure affects seed germination and plant development, and that plant responses to TCS are species specific. By affecting seed germination and seedling development, most notably root system development, TCS has the potential to impact plant community structure and the ecosystem services provided by wetland plants. The latter is of particular interest since many wetland ecosystem functions (i.e., nutrient cycling, erosion control, and carbon storage) are influenced by plant root systems and the communities they support (i.e., rhizosphere bacteria, mycorrhizae, and nitrogen-fixing bacteria). Further research is under way to determine the effects of TCS on plant community structure and ecosystem processes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

We thank Joanna Blaszczak, Joel Janssen, Regina Edziyie, Gopi Nallani, Chris Wall, and Misty Wellner. This research was funded in part by an Environmental Protection Agency Texas Institute for Environmental Assessment and Management grant to the University of North Texas, Institute of Applied Sciences, and funding from Tetra Point Fuels, Denton, Texas, USA.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES