Microbial transformation of chlorinated aromatics in sediments

<p>Numerous contaminants like heavy metals, polycyclic aromatic hydrocarbons (PAHs), chlorinated benzenes (CBs), polychlorinated biphenyls (PCBs), polychlorinated dibenzo- <em>p</em> -dioxins (PCDDs) and polychlorinated furans (PCDFs) are detected in the major rivers in the Netherlands. These contaminants have relatively low aqueous solubilities and bind substantially to the suspended solids in river water. Due to decreasing stream velocities in the downstream stretches of a river suspended solids will settle to the river bed. Thus, the deposition of suspended solids in the downstream stretches results in transport of contaminants from the water column to the sediments.<p>In the past decade widespread contamination of sediments with heavy metals, PAHs, and chlorinated aromatics (CBs, PCBs, PCDDs and PCDFs) has been identified in the Netherlands. The extent and seriousness of sediment contamination is most pronounced in the sedimentation areas of the rivers Rhine, Meuse and Scheldt. The total amount of heavily contaminated sediments in these areas is estimated at more than 100 million m <sup><font size="-2">3</font></SUP>. The reduced fertility of cormorants in the Biesbosch, a sedimentation area of the Meuse, can be put forward as an example of the adverse ecological impact of contaminated sediments.<p>Microorganisms (bacteria and fungi) play an important role in the decomposition of organic matter in sediments. The diversity of metabolic processes in microorganisms enables them to degrade a variety of organic contaminants as well. Two degradation processes can be distinguished: transformation and mineralization. Transformation is the process by which the chemical structure of the organic substrate (or contaminant) is altered and organic products are formed. Mineralization is the process whereby, besides biomass, only inorganic products (CO <sub><font size="-2">2</font></sub> , CH <sub><font size="-2">4</font></sub> , H <sub><font size="-2">2</font></sub> O, chemical elements) are formed during microbial metabolism.<p>In the presence of oxygen, as in the water column of rivers, the microbial degradation of chlorinated aromatic compounds decreases with an increasing number of chlorine atoms in the molecule. The relative persistence of higher chlorinated aromatics in the water column, combined with the hydrophobic properties of the aromatics, results in deposition of these contaminants in sediments. The upper few centimeters of sediment contain oxygen; most of the sediment is anoxic. Higher chlorinated compounds can, particularly under these anoxic conditions, be transformed by anaerobic microorganisms into lower chlorinated compounds. In the last 5 to 10 years microbial dechlorination reactions, i.e. the replacement of chlorines in the contaminant molecule with hydrogen atoms, have been demonstrated in laboratory tests for a variety of chlorinated aromatics. Some of these tests, conducted with anaerobic sediment, suggest that dechlorination reactions may occur in contaminated sediments.<p>Determining the occurrence of <em><em>in situ</em></em> microbial dechlorination was the general objective of the research described from which the following issues were deduced for application to the compound classes selected, i.e.: CBs, PCBs, PCDDs and PCDFs:<br/>1) What are the <em><em>in situ</em></em> dechlorination rates?<br/>2) What types of reactions occur and what are the reaction products?<br/>3) Does a general reaction pattern in dechlorination reactions exist?<br/>4) What are the consequences of <em><em>in situ</em></em> microbial dechlorination in sediments from an environmental point of view?<p>A dual approach, including field studies and laboratory experiments, was followed to<br/>determine <em><em>in situ</em></em> reactions.<p><strong>Field studies</strong><p>The hypothesis that <em><em>in situ</em></em> dechlorination would lead to a shift from higher to lower chlorinated compounds in anaerobic sediments was verified in the field studies, which consisted of sediment cores collected from Lake Ketelmeer, a sedimentation area of the Rhine River. The cores were sectioned into thin layers and the year of deposition of each layer was determined using radiochemical analyses. In addition, the concentrations of the chlorinated aromatics were determined in each layer. The contaminant concentrations were plotted against the year of deposition of the individual sediment layers (chapters 2 and 3). The possibility of a shift from higher to lower chlorinated compounds was verified with the aid of historical samples, collected in 1972 from the Lake Ketelmeer sediment top layer, and subsequently dried and stored for almost 20 years. The historical samples were analyzed together with the sediment core layers, thus applying presently available, identical analytical techniques. The historical samples reflect the contaminant concentration at time of deposition in the past (around 1972), either without or with only limited influences of degradation processes. The concentrations of chlorinated aromatics in the historical samples were compared to the concentrations found in sediment core layers deposited around 1970. In this way, the chlorinated aromatic concentrations at the time of deposition are compared with the concentrations after 20 years of "environmental incubation". Six highly toxic PCBs, representing the so-called non-ortho and mono-ortho PCBs that cause dioxin-like toxic effects, were selected and compared with the concentrations in the historical samples. Four of the six PCB congeners showed a significant disappearance in the sediment core layers relative to the historical samples. Increased concentrations of lower chlorinated congeners were not observed, probably due to the limited number of congeners analyzed. Maximum half-lives (t <sub><font size="-2">1/2</font></sub> ) for the four PCB congeners showing significant disappearances were estimated at about 10 years. Of the 17 chlorinated dioxins and furans studied, only four congeners showed significant disappearances, indicating slow or complete absence of reactions (t <sub><font size="-2">1/2≥</font></sub> 12 years). Compared to historical input, significant disappearances of hexachlorobenzene (t <sub><font size="-2">1/2≤</font></sub> 7 years), pentachlorobenzene and 1,2,3,5- tetrachlorobenzene were found. On the other hand, the concentrations of 1,3,5-tri-, 1,2-di-, and 1,3- dichlorobenzene increased significantly as compared to the historical input, indicating their formation in the anaerobic sediment.<p><strong>Laboratory studies</strong><p><em>Transformations in Lake Ketelmeer</em><em>sediment</em><p>The field studies make the occurrence of selective disappearances of chlorinated aromatics in Lake Ketelmeer sediment clear, but these observations do not reveal the responsible mechanisms. Therefore laboratory experiments were conducted using Lake Ketelmeer sediment under conditions resembling the field conditions (chapter 3). These experiments demonstrated that the indigenous anaerobic microbial population in Lake Ketelmeer sediment is able to dechlorinate hexachlorobenzene (HCB) to 1,3,5-tri-, and 1,3-dichlorobenzene. Based on the field and laboratory observations it was concluded that <em>in situ</em> microbial dechlorination of HCB in Lake Ketelmeer had occurred, resulting in an 80% loss of HCB. The tri-, and dichlorobenzenes produced are less toxic and have a lower bioaccumulation potentials than the parent compound HCB. Moreover, tri-, and dichlorobenzenes are, opposite to HCB, suitable substrates for aerobic mineralization. The fact that the transformation products are less hydrophobic and therefore can be transported more easily with infiltrating water than HCB can be regarded as a disadvantage of dechlorination reactions in the environment.<p>An enrichment culture from Lake Ketelmeer sediment that readily dechlorinates HCB under methanogenic conditions was obtained (chapter 4). Substantial dechlorinating activity was still present at low temperatures (3°C), whereas the optimum temperature for HCB dechlorination was around 30°C. The dechlorinating capabilities of the enrichment culture were verified with a variety of chlorinated aromatics. Individual incubations with 11 chlorinated benzenes demonstrated that from the 19 dechlorination reactions possible, only the seven reactions with the highest energy release took place.<p>The enrichment culture was also incubated with a selected number of PCB congeners in accordance with those selected in the field study. The results show a selective removal of chlorines from para and meta positions if the departing chlorine atom was surrounded by chlorines on both sides (chapter 5). PCB 118 does not meet this structural prerequisite and was not dechlorinated by the enrichment. This experimental result is in agreement with the field observation where no significant disappearance of PCB 118 is observed. The structural prerequisite can also be deduced from the observed selectivity in chlorobenzene dechlorination. The correlation with thermodynamics, as observed for chlorinated benzenes, appeared to be valid for PCBs as well. Only the energetically most favourable dechlorination steps are catalyzed by the enrichment, except ortho-dechlorinations.<p>The dechlorinating capacity of the enrichment culture for chlorinated dioxins was tested using 1,2,3,4-tetrachloro- <em>p</em> -dioxin (1,2,3,4-TeCDD) as a model compound (chapter 6). Microbial dechlorination of 1,2,3,4-TeCDD resulted in the formation of 1,3- and 2,3-dichlorodibenzo- <em>p</em> -dioxins as the main products. The results suggest that microbial dechlorination. of PCDDs in Lake Ketelmeer sediment might be possible. However, the field data indicate that <em><em>in situ</em></em> reactivity of PCDDs is extremely slow or absent. This may be due to extremely low concentrations of PCDDs in porewater and the slow desorption of PCDDs from sediments, resulting in a low microbial availability.<p>The similar experimental protocols used for the laboratory incubations with CBs, PCBs and PCDDs allow the calculation of relative dechlorination. rates for these three compound classes. Dechlorination half-lives for tetra-, penta-, and hexachlorobenzene are between 1 and 2 days for 1,2,3,4-TeCDD 15.5 days, and for the tested reactive PCBs, between 10 and 120 days. Remarkable is the difference between half-lives determined in the laboratory and half-lives estimated in the field study (up to a decade). In the contaminated sediments densities of the actively dechlorinating microbial populations may be smaller and activities may be lower due to lower concentrations of co-substrates and/or nutrients and lower temperatures than in the laboratory incubations. In addition, in the laboratory interference of sediment was eliminated; experiments were conducted in liquid media. Binding of the hydrophobic chlorinated aromatics to sediments reduces the concentration in the sediment pore water substantially and the slow release of bound compounds to the pore water may limit <em><em>in situ</em></em> dechlorination rates.<p><em>Transformation in coastal sediments</em><p>The microbial dechlorination of HCB in sediment from the Ems estuary is described in chapter 7. Under sulfate-reducing conditions, prevailing in the estuarine sediment, an HCB dechlorinating- culture was obtained. The dechlorination of HCB occurred concomitantly with sulfate reduction but was not directly coupled to sulfate reduction. These results indicate that microbial dechlorination of chlorinated aromatics is not necessarily restricted to fresh water sediments but may also occur in estuarine sediments.<p><strong>Environmental significance</strong><p>The PCB data from the field and laboratory study support the conclusion that the occurrence of <em><em>in situ</em></em> dechlorination of PCBs in Lake Ketelmeer sediment is highly likely. The environmental significance of <em><em>in situ</em></em> dechlorination, particularly for those congeners that cause dioxin-like biochemical and toxic effects is the subject of chapter 8. The significant disappearance of four PCB congeners in Lake Ketelmeer sediment, as observed in the field study, has resulted in a 75% decrease in the dioxin-like toxicity during the last 20 years. The selectivity in the dechlorinating activity of the enrichment culture from Lake Ketelmeer was used to predict the reactivity of all non- ortho and mono-ortho PCBs. All congeners, except PCB 77 and PCB 118, appear to be reactive. Dechlorination could result almost exclusively in products that have no dioxin-like toxicity. Other advantages of PCB dechlorination, besides decreased toxicities, are a lowered bioaccumulation potential and dechlorination. products being more suitable substrates for aerobic mineralization than the parent compounds.<p>The results of this study have been included in an overview of currently available <em><em>in situ</em></em> half-lives of chlorinated aromatics (chapter 9). Although the available data are limited, it is clear that <em><em>in situ</em></em> microbial dehalogenation is a relatively slow process that may proceed with half-lives of several years to decades in contaminated sediments. Despite the low rates, it is important to include these transformation processes in long-term sediment-quality prognoses. After all, these transformations with their significant toxicological implications are the essential first steps that may eventually lead to a complete elimination of PCBs from the environment. The correlation between energy yield and catalyzed dechlorination steps is further verified with data from the literature. The thermodynamically most profitable dechlorination steps appear to be catalyzed preferentially, except for ortho-dechlorinations. The correlation with thermodynamics appears to be an appropiate instrument for predicting microbial dechlorination pathways.<p>New prospects for the biotechnological clean-up of sediments contaminated with halogenated aromatics have emerged from the results of this study. The clean-up process should include two steps; an anaerobic dehalogenation and a subsequent aerobic mineralization of dehalogenation products. The stimulation of dehalogenation rates and the application of the two-step process on pilot scale are two examples of future challenges