Article pubs.acs.org/est
Interaction of Heavy Metals and Pyrene on Their Fates in Soil and Tall Fescue (Festuca arundinacea) Mang Lu,†,‡ Zhong-Zhi Zhang,*,† Jing-Xiu Wang,† Min Zhang,† Yu-Xin Xu,‡ and Xue-Jiao Wu§ †
State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Beijing 102249, China School of Materials Science and Engineering, Jingdezhen Ceramic Institute, Jingdezhen 333403, Jiangxi Province, China § Library , Jingdezhen Ceramic Institute, Jingdezhen 333001, Jiangxi Province, China ‡
S Supporting Information *
ABSTRACT: 90-Day growth chamber experiments were performed to investigate the interactive effect of pyrene and heavy metals (Cu, Cd, and Pb) on the growth of tall fescue and its uptake, accumulation, and dissipation of heavy metals and pyrene. Results show that plant growth and phytomass production were impacted by the interaction of heavy metals and pyrene. They were significantly decreased with heavy metal additions (100−2000 mg/kg), but they were only slightly declined with pyrene spiked up to 100 mg/kg. The addition of a moderate dosage of pyrene (100 mg/kg) lessened heavy metal toxicity to plants, resulting in enhanced plant growth and increased metal accumulation in plant tissues, thus improving heavy metal removal by plants. In contrast, heavy metals always reduced both plant growth and pyrene dissipation in soils. The chemical forms of Cu, Cd, and Pb in plant organs varied with metal species and pyrene addition. The dissipation and mineralization of pyrene tended to decline in both planted soil and unplanted soils with the presence of heavy metals, whereas they were enhanced with planting. The results demonstrate the complex interactive effects of organic pollutants and heavy metals on phytoremediation in soils. It can be concluded that, to a certain extent, tall fescue may be useful for phytoremediation of pyrene−heavy metalcontaminated sites. Further work is needed to enhance methods for phytoremediation of heavy metal−organics co-contaminated soil.
1. INTRODUCTION
fast-growing high biomass plants that accumulate moderate levels of metals in their tissues for metal phytoremediation. Extensive studies conducted both in the laboratory and in the field have demonstrated that tall fescue (Festuca arundinacea) has an important value for remediation of sites contaminated with organic pollutants including PAHs.10,15−19 Meanwhile, it had been reported that tall fescue had the potential of ecological rehabilitation on land contaminated by heavy metals including copper, lead, cadmium, zinc, nickel, etc.20−23 However, there is lack of information on the interactions of heavy metals and organic pollutants on their uptake, transport, and chemical form in plants especially in tall fescue. Organic and inorganic contaminants could interact with plants or with themselves and could influence the phytoremediation potential of plants.24 For example, Chigbo et al.25 found that the degradation of pyrene was significantly decreased by the presence of copper (50 and 100 mg/kg) in planted and nonplanted soils. In contrast, the addition of phenanthrene and pyrene could significantly promote Cd accumulation and removal by Juncus subsecundus, whereas
In nature, absolute single contamination is very scarce, and copresence of organic pollutants and heavy metals is often found, which have become major environmental and human health concerns worldwide.1,2 The co-occurrence of polycyclic aromatic hydrocarbons (PAHs) with heavy metals are frequently found and extensively evaluated in many types of anthropogenic industry contaminated sites, such as lumber and wood production sites,3,4 manufactured gas plant sites,5 mining and metallurgy industry sites,6 etc., and even in the sediments of natural water bodies.7,8 Extensive studies have been performed on phytoremediation of heavy metals and organic pollutants alone.9−11 However, phytoremediation of organic and inorganic co-contaminants is not well understood, and this has emerged as an environmental concern because many soils are exposed to co-contamination of heavy metals and organic compounds.12 Although using hyperaccumulating plants for phytoextraction of metals may have advantages over other methods, many limitations exist for this technology.13 A serious limitation is that these hyperaccumulators are usually small and grow slowly, or strongly region dependent, making them difficult to harvest mechanically, and they are unsuitable for generalization and application on a large scale.14 An alternative approach is to use © XXXX American Chemical Society
Received: July 30, 2013 Revised: September 27, 2013 Accepted: October 25, 2013
A
dx.doi.org/10.1021/es403337t | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
accumulation of PAHs by plants and dissipation of PAHs from soils were not significantly affected by Cd additions.26 Nevertheless, previous studies were mainly focused on a single metal pollution case,24−26 and the knowledge about the influence of these interactions between co-contaminants on the uptake, translocation, and degradation of metals and organic pollutants (e.g., PAHs) in plants is still poor. The objectives of this study were to investigate (1) the influence of metal ions (Cu2+, Cd2+, and Pb2+) and pyrene as combined contamination on growth of tall fescue, (2) the impact of the co-contaminants on uptake and translocation of heavy metals and pyrene in plants and their removal from the substrate by plants, and (3) the effect of the co-contaminants on pyrene mineralization and microbial counts in the soil.
microcosm was prepared by adding HgCl2 (5 g/kg) and chloramphenicol (500 mg/kg) to the pyrene spiked soil. Tall fescue seeds were germinated and grown for two additional weeks in a clean sand bed. Ten seedlings of similar size were carefully transplanted to the soils. Afterward, a layer of mixture of paraffin wax and petroleum jelly (1:4) was used to physically seal the opening between the foliar chamber and the root chamber.16 All experiments were performed in a greenhouse with natural light and day/night temperature of 28/21 °C and humidity of 73/86%. During the experimental period, deionized water was added to the root chamber by using a syringe to compensate the water losses. Soil moisture content was maintained at approximately 60% of its water holding capacity by using weighing method. Air was continuously evacuated from both the foliar chamber and the root chamber separately by using vacuum pumps at a constant flow rate of 30 mL/min, and the outlet air was passed through a series of sampling traps. Similarly, air was evacuated from the root section at a constant air rate of 10 mL/min and passed through a similar but separate series of traps. The gas-phase sampling and radioactivity analysis were performed following the descriptions introduced by Cheng and Wong,29 and the details are given in Section II of the Supporting Information. 2.2. Analytical Methods. 2.2.1. Sampling. After 90 days of growth, the plants and soils were removed from the growth chambers. The mixes of paraffin wax:petroleum jelly were completely removed to exclude their possible interferences on the subsequent analysis. The plants were washed with deionized water and separated into roots and shoots. After drying with filter paper, the subsamples were freeze-dried, and dry weights were recorded. Following removal of plant tissues, the entire soil from each chamber was thoroughly homogenized, air-dried at room temperature, and ground sufficiently to pass through a 100-mesh sieve. 2.2.2. Pyrene Analysis. Residual pyrene in soils and plants were analyzed using an Agilent 7890−5975c gas chromatography−mass spectrometer (GC−MS). Approximately 10 g of soil sample or 1 g of plant biomass was spiked with p-terphynyld14 and mixed with 10 g anhydrous magnesium sulfate. The mixture was then Soxhlet-extracted with dichloromethane:acetone (1:1, v/v) for 16 h. The extract was condensed to approximately 2 mL by rotary evaporation, loaded onto a silicagel column (10 cm ×6 mm ID), and then eluted consecutively with 15 mL hexane and dichloromethane mixture (1:1, v/v). The filtrate was concentrated by evaporating the solvent under N2, and the residue was dissolved in hexane with a final volume of 1.0 mL for GC−MS analysis. The instrumental conditions are shown in Section III of the Supporting Information. Additionally, residual 14C in the soil and plant samples were determined after oxidation into 14CO2 by using wet combustion method,29 of which the details are given in Section IV of the Supporting Information. 2.2.3. Chemical Forms of Metals. Metal chemical forms in plant samples were measured using a sequential extraction method mentioned by Sahuquillo et al. 30 with some modifications. A total of five designated extraction solutions were used in the following order: (1) 80% ethanol, extracting ethanol-soluble protein-bound metal ions, (2) deionized water, extracting water-soluble metal−organic acid complexes, (3) 1 M NaCl, extracting protein integrated Cd, (4) 2% acetic acid (HAc), extracting sparingly soluble phosphate-bound metals, and (5) 0.6 M HCl, extracting insoluble metal oxalates. Metal ion concentrations were determined by using inductively
2. MATERIALS AND METHODS 2.1. Experimental Design. A loamy sand soil without detectable PAHs employed in the present study was the same as that used in our previous work.27 The main characteristics of the soil are provided in Section I of the Supporting Information. In this study, the soil was treated with both 14C-pyrene (2.0 μCi/kg soil) and unlabeled pyrene (50−500 mg/kg soil). Single pyrene treatments were 50, 100, 200, 300, and 500 mg/ kg, respectively. In the co-contamination experiment, the initial concentration of pyrene was 100 mg/kg. The 14C-labeled contaminant was pyrene-4,5,9,10-14C with 55 mCi/mmol (>98% purity, Sigma-Aldrich, St. Louis, MO). Both unlabeled and labeled pyrene samples were first dissolved in acetone and then were added drop by drop to 25% by weight of the required quantity of soil. After acetone evaporation, the soils were blended thoroughly with the remaining 75% by weight of the required quantity of soil.28 The same amount of acetone was used in all treatments. Then, two levels of Cu2+ (100 and 500 mg/kg), Cd2+ (100 and 500 mg/kg), and Pb2+ (500 and 2000 mg/kg) were added to pyrene-spiked soils, respectively. The soils were fertilized with a NPK fertilizer mixture (1 g/kg of soil) containing N:P2O5:K2O = 1:0.35:0.8 ratio. Afterward, the spiked soil samples were thoroughly mixed by sieving, covered with aluminum foil, and equilibrated in the dark for two weeks prior to the experiment. In the present study, a series of plant growth chambers made of plexiglass were used for monitoring the fate of 14C-pyrene in the soil−plant system. The growth chamber design was modified from Chen et al.16 The schematic diagram of experimental setup is shown in Figure S1 of the Supporting Information. Each chamber consisted of an upper (foliar) and a lower (root) section. The upper section had a height of 50 cm with a 20 cm ×20 cm base area that gave a volume of 20 L. The lower section serving as the soil chamber consisted of a cylindrical column, 15 cm in height and 10 cm inner diameter, resulting in a total volume of 1.18 L. Each soil chamber received 2.4 kg soil. An air inlet constructed from hard plastic tubes (PVC) with 25 mm ID was installed on the side face of both the shoot and root chambers. The air intake tube was connected to two charcoal bottles and a bacterial filter in series. A completely randomized block design (eight metal treatments × two pyrene treatments × one plant treatment) with three replicates was employed: (1) control (CK), plant grown in nonspiked soil; (2) bulk (unplanted), spiking soil with metals and/or pyrene without plant growth; and (3) plant grown in spiked soil. To investigate abiotic loss of pyrene, a soil B
dx.doi.org/10.1021/es403337t | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
coupled plasma−atomic emission spectrometry (ICP-AES) (Thermo Electron Co., Franklin, U.S.A.). The details are given in SectionV of the Supporting Information. 2.3. Bacterial Counts. Heterotrophic bacteria were counted on nutrient agar. Pyrene-degrading bacteria were enumerated by using the plate spread method with the basal mineral medium (g/L: pyrene 1.0, NH4Cl 1.0, K2HPO4 0.3, KH2PO4 0.2, MgSO4 0.5, agar 15, pH 7.0).27 The total number of colony forming units (CFU) of bacteria was counted after incubation at 28 °C in the dark for 2 days for heterotrophs and 14 days for pyrene-degrading bacteria, respectively. 2.4. Quality Assurance. Pyrene recovery was measured by adding a known concentration of pyrene standard (1 mg/kg) to uncontaminated soil and plant. The recoveries were 96.3 ± 6.32% and 102.4 ± 8.43% for soil and plant samples, respectively. Similarly for heavy metals, certified soil and plant reference standards (GBW07409 and GBW10014, Centre of Standard Materials of China) were used to evaluate the accuracy and precision of sample digestion and subsequent analysis. The mean recoveries of Cu, Cd, and Pb in soil standard were 97.5 ± 3.65%, 98.5 ± 2.43%, and 101.2 ± 3.52%, respectively. The average recoveries of Cu, Cd, and Pb in plant standard were 97.4 ± 2.21%, 98.7 ± 2.65%, and 98.5 ± 3.16%, respectively. 2.5. Data Analysis. Tolerance index (TI) was calculated by the following equation31 TI (%) =
growth in soil + metal × 100 growth in soil − metal
Figure 1. Shoot (upper) and root (lower) dry weight of tall fescue after 90-day growth. CK: control, without any spike. Py0: without pyrene spike. Py100: spiked with 100 mg/kg pyrene. Values displaying different superscript letters are significantly different based on LSD (p < 0.05).
(1)
The translocation factor (TF) was calculated as31 TF (%) =
Metal content in shoot × 100 Metal content in root
(2)
In this study, all experiments were performed in triplicate to get reliable data, and the results are reported as means ± standard deviations. Statistical significance was evaluated using SPSS package (version 11.0) with two-way ANOVA, and least significant difference (LSD) was applied to test for significance at P < 0.05 between the means.
Figure 2. Total number of heterotrophs and pyrene degraders in soils, averaged over various metal treatments, before and after 90 days of plant growth. There was no significant difference between Cd treatments with or without vegetation. Treatment: Py represents 100 mg pyrene per kg soil.
3. RESULTS AND DISCUSSION 3.1. Growth Response. Figure 1 shows the dry matter yield of tall fescue grown in the soil under different contamination cases. It can be observed that the growth of tall fescue was significantly affected by heavy metals and pyrene treatments and their interactions. In the case of single pyrene contamination, no obvious difference was observed among the biomasses in soils spiked with 50 or 100 mg/kg pyrene and those with control. When pyrene concentration exceeded 200 mg/kg soil, the plant biomass decreased significantly with increasing pyrene concentration. The prohibitive effect of pyrene on root growth was more pronounced than that on shoot growth. When the initial pyrene concentration was 500 mg/kg soil, shoot and root biomass was only 66.9% and 46.7% of the control, respectively. The ability of contaminated soil to provide water and nutrients to plants may be reduced by the presence of high levels of PAHs, leading to a decrease in biomass yield.32,33 Figure 1 shows that 500 mg/kg of Cu and Cd reduced growth at the end of 90 days to approximately 75% and 66% relative to control, respectively. The concentration of 500 mg/ kg of Cu and Cd proved to be toxic, affecting the plant growth
severely. Nevertheless, adding 500 mg/kg of Pb did not cause significant reduction in phytomass production of tall fescue, and the shoot and root biomasses only decreased by 22.0% and 37.3% respectively at 2000 mg/kg of Pb. This indicated that tall fescue has high resistance to Pb pollution in soil applied in this study. Actually, it has been demonstrated that the biomass yield of tall fescue in soil spiked with 1000 mg/kg of Pb was almost the same as that of unspiked control soil and approximately 85% of that of the control in the case of 2000 mg/kg Pb.20 Tolerance indices of tall fescue are listed in Table S1 of the Supporting Information. The most common effect of heavy metal toxicity in plants is reduction in seed germinability, stunted growth, leaf chlorosis, inactivation of enzymes, and inhibition of photosynthesis and mineral/nutrient absorption by plants, etc.34 However, some plant species exhibit resistance or tolerance to heavy metals with varying degrees through a number of ways depending on the metals, either through chelation, compartmentalization, or precipitation.35 C
dx.doi.org/10.1021/es403337t | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
Figure 3. Heavy metal concentrations (stack column) in plants and percentage removal (line) by plants influenced by different metal treatments with or without pyrene (100 mg/kg) after 90 days of tall fescue growth. Treatment: Py represents 100 mg pyrene per kg soil.
Figure 5. Residual concentration of pyrene in planted and nonplanted soil after 90 days of planting. CK: abiotic control, soil supplemented with HgCl2 (5 g/kg) and chloramphenicol (500 mg/kg) without planting. Py: spiked with 100 mg/kg pyrene. Bars (means ± SE, n = 3) with different letters indicate a significant difference based on LSD (P < 0.05).
3.2. Microbial Numbers. Figure 2 shows that addition of heavy metals caused a significant reduction in bacterial counts due to metal toxicity effects. Indeed, in trials with single metal amendment without vegetation, the heterotrophic counts were reduced by 1 order of magnitude after 90 days of incubation. In all cases studies, bacterial numbers were significantly higher in planted soils than in unplanted soils (P < 0.05). In 90 days, the addition of pyrene increased the degrader amounts by one to 2 orders of magnitude in unplanted and tall fescue treatments. Meanwhile, the ratio of degraders/heterotrophs increased from 0.056% in raw soil to 2.63% in unplanted and 6.15% in planted treatments. The plant growth and development can also be improved by phytohormones such as auxins, cytokinins, and gibberellins, which are yielded by plant-associated bacteria.41 Thus, we inferred that the pyrene amendment could favor the microbial prosperity and/or microbial population alteration in the planted soils, thus alleviating metal stress and promoting the phytomass yield (Figure 1). 3.3. Metal Accumulation, Translocation, and Extraction by Plant. 3.3.1. Total Metals. Figure 3 and Table S2 of the Supporting Information show the metal uptake and partitioning in tall fescue grown in artificially contaminated and unspiked soils, respectively. The concentrations of heavy metals in the plants were in the order: root > shoot. In general, the concentrations of Cu, Cd, and Pb in the plants increased with the increasing spiking level of metals in soils with or without pyrene spike (Figure 3). The translocation factor (TF) was significantly influenced by the different levels of contaminants (P < 0.05) (Table S1, Supporting Information). In the absence of pyrene, the highest TF was observed for Pb, followed by Cu, and the lowest value ( 0.05) in the Pb treatment (Figure 1). Table S1 of the Supporting Information shows that the tolerance indices of tall fescue were significantly increased in Cu and Cd treatments by 100 mg/kg pyrene spike (P < 0.05). The growth stimulation under coexposure indicates that tall fescue possesses the potential to tolerate moderate metal and pyrene co-contamination. An addition of phenanthrene and pyrene significantly enhanced Juncus subsecundus growth in Cd treatments compared to the single Cd treatments,26,36 but the spike of pyrene (250 and 500 mg/kg) markedly exacerbated Cu (50 and 100 mg/kg) toxicity to Brassica juncea.25 Similarly, certain concentrations of pyrene could alleviate the growth inhibition by Cu to maize (Zea mays)37 and by Pb to ryegrass (Lolium multiflorum),38 but pyrene could not alleviate Cd toxicity to maize.39 The metal-caused growth inhibition effects could also be alleviated by the presence of other organic pollutants, whereas high concentrations strengthened that toxicity.40 D
dx.doi.org/10.1021/es403337t | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
plants (maize and Sedum alfredii, respectively) with PAH additions. Our results demonstrate that co-contamination with PAH can change metal uptake extent by plants (Figure 3). PAHs including pyrene can passively penetrate the root cell membranes of plants without any carrier, which can therefore facilitate the penetration of metal or metal complexes into the cells.44 The pyrene penetration to root cell membranes could be one of the explanations for the observed increment in shoot and root concentration of metals as well as increased translocation factor with pyrene addition in the present study. Because of experimental condition limitations in this study, this work is still at a preliminary stage. We could not conclude on the existence of metal complexes with pectate or organic acids (mainly oxalate) other than phytochelatins through successive extraction. Therefore, further studies are required to examine distribution of heavy metals and its bonding state at the ultrastructural level by using micro X-ray absorption analysis or other advanced analytical methods. This will be helpful for us to improve our understanding of the joint effects of heavy metals and PAHs on cell ultrastructure within the plant. A better understanding of metal chelation sequestration in plants may eventually contribute toward the development of biorecovery techniques for the restoration of soils cocontaminated with organics and metals. The TF and percentage removal of the three metals also increased due to pyrene addition (Table S1, Supporting Information and Figure 3). The uptake of Pb for tall fescue was higher than Cu, while that of Cd was the lowest. Although metal concentrations in plants rose when raising metal spiking level, the percentage removal of heavy metals fell remarkably (Figure 3), which is due to the great reduction in phytomass yields at higher metal concentrations in soils. The soil pH could influence metal bioavailability, but in the present study, the pH was not significantly impacted by metal treatments with or without pyrene addition before and after the experiment (data not shown). 3.3.2. Chemical Form of Metals. The results of metal uptake shown in Figure 3 suggest that the absorption mechanism of these metals could be quite different due to plant species and metal characteristics. Nevertheless, total concentrations of metals can not provide sufficient toxic information. Metal mobility, toxicity, bioavailability, and chemical interactions are dependent on not only their total concentration, but also their chemical forms.45 Table S3 of the Supporting Information shows the chemical forms of Cu, Cd, and Pb in roots and shoots of tall fescue grown in soils contaminated with heavy metals with or without pyrene spike. In the roots of 100 mg/kg Cd-treated plants, the amount of CdHAc was dominant, followed by CdHCl, CdNaCl, Cdethanol, Cdwater, and Cdresidue. Pyrene supply increased the Cd concentration of various chemical forms in contrast to Cd treatment alone. Moreover, the proportion of Cdethanol, Cdwater and CdNaCl was increased respectively by 9.5%, 23.6%, and 33.4% relative to a single Cd treatment, suggesting that adding pyrene increased the proportion of Cd mobile forms (Cdethanol, Cdwater, and CdNaCl). In the shoot tissue of the Cd treatment, Cd concentration of various chemical forms was distributed as the decreasing order: CdHCl > CdNaCl > CdHAc > Cdethanol > Cdwater > Cdresidue. Supplement of pyrene did not alter this distribution pattern. Nonetheless, Cd concentrations in various chemical forms of pyrene-supplied plants were higher than that in Cd treatment alone. Overall, the absolute concentrations of
Figure 6. Percentage of cumulative 14CO2 evolved from (A) unplanted soils, (B) root chambers, and (C) shoot chambers of different treatments during the 90-day cultivation period. The percentage was calculated based on the initial total 14C applied. Py: spiked with 100 mg/kg pyrene.
increment of metal content in roots was more than that in shoots when raising metal concentrations in the soils. Interestingly, pyrene addition significantly increased metal concentrations of both roots and shoots at all levels of applied Cu, Cd, and Pb (P < 0.05) (Figure 3). These results are in agreement with the findings reported by others,25,26,36 who demonstrated that co-contamination with PAHs can increase the extent of heavy metal uptake by plants. However, contradictory responses were also obtained by Lin et al.37 and Wang et al.,43 who reported a decrease in metal uptake by E
dx.doi.org/10.1021/es403337t | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
4). Thus, in the present study, plant uptake was not a main pathway for pyrene removal in contrast to biodegradation. 3.5. Dissipation of Pyrene in soil. After 90 days of growth, the percentage of pyrene removal from soils was significantly influenced by the interaction of heavy metals, pyrene, and planting/nonplanting treatments accounting for 36.6−91.4% and 28.5−73.7% for planted and unplanted soils, respectively (Figure 5). The residual pyrene in soil vegetated with tall fescue was significantly (P < 0.05) lower than in the unplanted soil when soil was treated with 100 mg/kg of pyrene, indicating enhanced pyrene dissipation by planting. The results showed that without heavy metal spike, the extractable pyrene concentration in nonplanted and planted soils was 26.3 and 8.6 mg/kg, respectively, corresponding to a removal percentage of 73.7% and 91.4%, respectively. Figure 5 shows the effect of Cu, Cd, and Pb on pyrene dissipation in soils planted/nonplanted with tall fescue. The results showed that pyrene removal decreased significantly with metal addition levels, which was more obvious in the nonplanted treatments. There was a decrease in significant difference in pyrene removal between planted and nonplanted treatments with metal additions. Pyrene removal could be due to biodegradation, plant uptake and metabolism, or abiotic dissipation including photodegradation, volatilization, and incorporation into soil organic material.37 In the present study, the abiotic loss of pyrene amounted to 21.6% of the initial quantity (Figure 5), indicating that biotic process was the major dissipation pathway for soil pyrene. Despite the possible high toxicity of heavy metals to soil microorganisms, the residual pyrene in soil decreased even at 2000 mg Pb per kg, implying that even at higher metal concentration, highly adapted metal-resistant microbes could have enhanced pyrene degradation. It is known that plants could promote microbial biodegradation of PAHs as compared to that from unplanted treatment due to increased microbial activity and degradation mediated by plant-secreted enzymes in the root zone.48 In the present study, the abundance of PAH degraders in soils significantly increased due to planting of tall fescue (Figure 2). An increment in microbial counts, enhancement of microbial activity, and/or modifications in the microbial community structure in the rhizosphere as a result of the input of easily degradable organic substances such as plant exudates may improve the microorganism resistance to the heavy metal stress and enhance pyrene biodegradation. Effect of heavy metals on PAH biodegradation can be negative or positive depending on the type and concentration of both PAHs and heavy metals. For example, Pb can increase the dissipation rate of pyrene in Pb-pyrene co-contaminated soil,49 while the presence of heavy metals including Cd inhibited a broad range of microbial processes.50 The presence of plant also played an important role on impact of metals on PAH removal. Either positive or negative differences in PAH biodegradation may occur due to the distinction in the quality or quantity of nutrients released by plant roots as well as dead roots.51 In the present study, there seemed to be a favorable effect on PAH degradation. Plants may alleviate heavy metal toxicity to microorganisms through complexation of metal ions with root exudates.52,53 Further intensive study is needed to better understand the effect of plants on PAH dissipation in soils co-contaminated with metals and PAHs. In this study, the performance of tall fescue on metal phytoremediation did not meet the criteria for a hyperaccumulator, namely, metal concentrations in aerial parts are
all forms of Cd in roots were far greater than those aboveground. In the Pb treated group, the percentage distribution of various chemical forms of Pb changed drastically (Table S3, Supporting Information). Ethanol extracted Pb appeared in roots with a dominant percentage, followed by NaCl extracted, HAc extracted, HCl extracted, water extracted, and residue forms. In shoots, Pb mostly existed in HAc extracted form. Between roots and shoots, HAc and NaCl extracted Pb had higher concentrations in shoots, and ethanol extracted Pb had higher concentrations in roots. Pyrene spike did not alter distribution patterns of various Pb chemical forms like the case of Cd treatment. In the Cu treated group, the percentage distribution of various chemical forms of Pb was similar with that of Pb. We also observed from Table S3 of the Supporting Information that the chemical form distribution of Cd was quite different from that of Cu and Pb. These results suggest that the tolerance of tall fescue to heavy metals is highly metal specific.46 The existence of Cd in the roots of tall fescue was mainly in the form of difficult soluble phosphate that included divalent phosphate and orthophosphate. The mobility of these metal forms was small. Therefore, they mostly existed in the roots, and only a small portion of them was transferred to the aerial parts of the plants. This indicates that the shoots of tall fescue were more sensitive to Cd toxicity than to Cu and Pb toxicity. 3.4. Pyrene Concentration and Partitioning in Tall Fescue. Pyrene concentrations in plant roots and shoots are shown in Figure 4. Pyrene in roots grown in unspiked control soil was not detected. Figure 4 shows that the pyrene concentration in the plants markedly increased with an increase in the initial content of pyrene in the soil. However, the ability of pyrene uptake and translocation by tall fescue was weak, as suggested by the low root-to-shoot transfer factor of pyrene in the plants (100 mg/kg for Cd and 1000 mg/kg for Cu and Pb, and TF values are greater than 1.0.32 However, the ability of tall fescue to tolerate and accumulate various heavy metals may be useful for phytostabilization. Tall fescue cannot compare with hyperaccumulators in the phytoextraction efficiency of heavy metals. Unlike some hyperaccumulators such as Sedum alfredii,43 tall fescue grew better and could accumulate more heavy metals in contaminated soils with PAH addition as demonstrated in the present study. Moreover, increased pyrene degradation could be obtained by planting tall fescue. These facts indicates that tall fescue may be a good candidate for remediation of metal and PAH co-contamination. 3.6. Pyrene Mineralization. To investigate the effect of vegetation and/or heavy metals on pyrene mineralization, 14 CO2 generated in the root and shoot chambers of different treatments was collected and analyzed over the 90-day cultivation period. Figure 6 illustrates the cumulative evolution of 14CO2 in the treatments. The unplanted control showed minor 14CO2 production (4.8−11.5%) from the soil, and mineralization of 14C-pyrene was obviously lower in the metal treatments than in the unspiked counterpart (Figure 6A). A long lag phase (30−50 days) of 14CO2 production was observed from the unplanted soils, and metal additions prolonged this lag phase. In planted treatments, the lag phase of 14CO2 yield in root chambers was about 20−40 days (Figure 6B). Afterward 14CO2 evolution increased almost linearly until the end of the experiment, reaching 14.3−35.8% of the total initial 14Cactivities at the 90th day. The lag period for the onset of pyrene mineralization was shorter in the planted treatments relative to the unplanted control, suggesting that the establishment of tall fescue could enhance not only the transformation of pyrene molecules into their metabolites (Figure 5) but also pyrene mineralization (here 14CO2 evolution) in soils. By comparing pyrene dissipation efficiency (28.5−91.4%, Figure 5) with 14CO2 evolution (4.8−35.8%, Figure 6), we could infer that there existed large amounts of pyrene intermediates in the soil. In all the treatments tested, 14CO2 production was significantly higher from the root chambers compared to the shoot chambers (
