Background of the Invention

The present invention relates to the field of ecology and more particularly to the remediation of environmental contaminants by enzymatic degradation.

Contamination of the air, water and soil is a severe problem endangering the lives of many plants and animals, including humans. Many attempts have been made to reduce contamination by either preventing escape of the contaminants into the environment, containing the contaminants at one site, or treating the contaminants in some way to make them less harmful.

Extensive soil, water, sediment and aquifer contamination has occurred from the manufacture and widespread use of explosives by both civilians and the military. For example, the compound 2,4,6-trinitrotoluene (TNT) is a highly oxidized nitroaromatic that is stable on soil surfaces in the environment for as many as 40 years.

Currently, the preferred technology for remediating TNT-contaminated soil is by burning the soil and then cashing the incinerated soil in an enclosure for an indefinite amount of time. The burning treatment is expensive, costing $300-400 per ton, and, for a large contaminated site encompassing several acres, could cost tens of millions of dollars.

Scientists have shown that it is possible to degrade certain organic pollutants in soil through oxidation/reduction (redox) reactions. The goal has been to chemically or biologically oxidize or reduce the contaminants or functional moieties of the contaminants to innocuous compounds or

compounds that can be easily degraded and eliminated from the soil by known processes. Nitroorσanic Pesticides

Environmental contaminants that have been partially degraded through redox reactions include nitroaromatic pesticides such as parathion, methyl parathion, trifluralin, profluralin, benefin, nitrofen and pentachloro-nitrobenzene as described by Williams, P.P., J?esidue Jev. 66:63-135 (1977); Wahid, P.A. et al . , J. Environ . Qual . 9:127-130

(1980); Graetz, D.A. et al. , J . Water Poll . Control Fed . 42:R76-R94 (1970); Adhya, T.K. et al . , J. Agric . Food Chem . 29:90-93 (1981); Camper, N.D. et al., J. Environ . Sci . Health B15:457-473 (1980); Golab, T. et al., J. Agric. Food Chem . 27:163-179 (1979); Probst, G.W. et al . , J . Agric . Food Chem . 15:592-599 (1967); Willis, G.H. et al . , J. Environ . Qual . 3:262-265 (1974); Golab. T. et al . , J. Agric. Food Chem . 18:838-844 (1970); Lee, J.K. et al . , Misaengmul Hakhoe Chi . 20:53-66 (1982); Qian, W.W. et al . , Huanjing Kexue 3:36-39 (1982); and Murthy, N.B.K. and D.D. Kaufman, J. Agric . Food Chem . 26:1151-1156 (1978).

These nitroaromatic pesticide treatments generally involve the anaerobic transformation of the pesticides to reduced compounds by unidentified substances naturally present in the soil. However, the reduced products of nitroaromatic pesticide degradation include anilines and other compounds that are considered to be environmental hazards. Nitroorσanic Explosives

Many unsuccessful attempts have been made to oxidize the explosives 2,4,6,-trinitrotoluene (TNT) , hexahydro-1,3,5-trinitro-l,3,5-triazine (RDX) , octahydro-1,3,5,7-tetranitro-l,3,5,7- tetraazocine (HMX) and N-methyl-N- 2,4,6-tetranitroaniline (Tetryl) , nitrocellulose

and red water (a by-product of TNT production) to innocuous products. None have found any practical or commercial application because these compounds are highly oxidized, and further oxidation generally requires excessive amounts of energy. Attempts have also been made to remediate these compounds microbially by reduction under anaerobic conditions. (Alvarez, M. et al . , Enzyme Catalyed Transformation of 2,4,6-Trinitrotoluene, Abstracts of the General Meeting of the Am . Soc . for

Microbiology 91:217 (1991)) However, the reduction products of these compounds include the corresponding amines and several other less well defined hydroxyazo compounds. The analogs of these reduction products are potential carcinogens and are considered to be environmentally hazardous. Haloσenated Hydrocarbons

Halogenated hydrocarbons as a class of compounds are one of the most ubiquitous pollutants in the United States. They have been and still are widely used in many industries as cleaning solvents, refrigerants, fu igants and starting materials for the syntheses of other chemicals. Because of their extensive use, there are hundreds of contaminated groundwater and landfill sites in the United

States, many of which are superfund sites for which there is no inexpensive, effective remediation technology available. Also, industrial waste treatment technology is expensive and not always effective.

In contaminated ground water systems, the water is pumped out of the reservoir and treated with the "air stripping" treatment procedure. Halogenated hydrocarbons have also been remediated by a photolysis procedure wherein contaminated soil or sediment is placed on an oxide film and irradiated with concentrated sunlight to remove chloride

atoms. These procedures are expensive and only successful if all of the contaminated material has been successfully removed from the site of contamination. Effective in situ treatment is not practiced because of a lack of treatment technology. Bioremediation has not been successful because maintenance of a viable microorganism population is not generally feasible in subsurface ecosystems. Chemical remediation processes have not been utilized because of the delivery of large amounts of the necessary chemicals and problems associated with groundwater hydrology.

Bioremediation has received considerable attention as an in situ remediation process of contaminated waste sites. The parent pollutants, however, are often resistant to degradation and must first be transformed to more degradable compounds for the processes to be effective. Although many microorganisms have been isolated that are capable of degrading halogenated hydrocarbons in the laboratory, they are not always effective when ported to the field situation. Cvano Compounds

Aliphatic and aromatic cyano compounds are used as solvents and intermediates in the chemical industry in a variety of synthetic processes including textiles and pesticides. For example, acrylonitrile is a high production compound with output exceeding more than 2.3 billion pounds a year. These compounds can enter the environment through manufacturing waste waters and from the polymers of which they are associated and as a result of applications of pesticides such as dichlobneil (2,6-dichlorobenzonitrile) and bromoxynil (3,5-dibromo-4-hydroxybenzonitrile) .

Biodegradation of selected cyano compounds has been demonstrated in waste water treatment systems.

In soils, however, degradation is more difficult and high concentrations of the pollutants are often not readily degraded. Also, not all soils, in particular sandy soils, have the necessary microbial populations to degrade nitriles. Anisoles

Anisoles are used as intermediates in the chemical industry for the manufacture of a large number of polymer, dye and pesticide compounds. These compounds find their way into the environment through point source and non-point source pathways. For example, the pesticide Methoxychlor™ has been one of the most widely used pesticides in the United States. In general, anisoles are hard to degrade because the methyl-oxygen bond is very strong. Other Contaminants

Remediation of other environmental contaminants such as metals has also been largely unsuccessful. Metal contaminants have been treated by the "pump and treat" procedure wherein water is pumped out of the contaminated area and passed over a tube containing titanium to transform the contaminants to compounds that are less hazardous to the environment. The "pump and treat" method is very costly, is only applicable for removal of volatile contaminates from surface water or aquifers, and is not successful until the source of the contamination is depleted. It would be of great environmental benefit to have an inexpensive method of degrading contaminants in soils, waters, sediments, and aquifer materials that results in products that are environmentally acceptable. It is therefore an object of the present invention to provide an improved method of remediating environmental contaminants.

It is a further object of the present invention to provide a method of remediating environmental contaminants that can be carried out in situ and in batch reactors. It is a further object of the present invention to provide a composition for rapid reduction of contaminants.

It is a further object of the present invention to provide a process for the production of contaminant-reducing agents from soil.

It is a further object of the present invention to provide a method of remediating nitro-, halogenated-, cyano-, methoxy-organic, and metal contaminants from the environment. It is a further object of the present invention to provide a method of oxidizing reduced pollutants.

It is a further object of the present invention to provide a method of remediating environmental contaminants that is cost-effective.

It is a further object of the present invention to provide a method of remediating soils, sediments, and aquifers that maintains the integrity of the environmental compartment.

Summary of the Invention

A method for the remediation of contaminants of soil, water, sediment and aquifers is disclosed wherein contaminants are degraded by a remediation protein found in soil and sediment having a substantially high organic content and are transformed to environmentally safe products. For example, remediation of nitroorganic compounds is achieved by reduction by a reducing agent followed by oxidation. Reduction preferably takes place in a substantially anaerobic

environment. An anaerobic environment is naturally present in water and aquifers and can be created in contaminated soil or sediment by flooding the soil or sediment with water. Remediation of contaminants is achieved by combining the contaminated soil, water, sediment or aquifer material with either soil containing an adequate amount of the remediation protein, a crude enzyme preparation containing the remediation protein, the remediation protein as a semi-purified or purified enzyme specific for reduction of a particular contaminant, or a combination thereof. The contaminated soil, water, sediment or aquifer material is incubated with the remediation protein for a sufficient amount of time to allow degradation of the contaminants. The addition of a reducing metal, such as iron, to the soil further accelerates the remediation process.

Oxidation is achieved by oxygenating the water, aquifer material, flooded soil, or sediment containing the reduced contaminant or by simply removing the water from flooded soil or sediment containing the reduced contaminant.

A method for preparing the crude soil enzyme extract or isolated enzyme specific for remediation of specific environmental contaminants is disclosed wherein the remediation protein is extracted from soil by combining the soil with a solvent that solubilizes proteins. Preferably the extract is prepared from soil having a relatively high carbon content so that it contains a higher concentration of the remediation proteins. Further purification of the remediation protein can be achieved by protein precipitation and fractionation on chromatography columns. Semi-purified enzymes particularly useful for reducing specific classes

of environmental contaminants and the processes for isolating these proteins are provided herein.

Contaminants that can be degraded by the remediation method described herein include nitroorganics in general and specifically munitions such as TNT, RDX, HMX, nitrocellulose, and red water and pesticides such as methyl parathion and 2-(sec-butyl)-4,6-dinitrophenol, also known as Dinoseb™; halogenated organic compounds such as halogenated organic solvents, halogenated pesticides and other industrial halogenated compounds such as and pentachlorophenol; cyano compounds such as benzonitrile, acetonitrile, and other industrial chemicals; anisoles such as anisole, dyes and pesticides containing methoxy moieties; and metals such as chromium.

Brief Description of the Drawings

Figure 1 is a graph of the relative reducing activity of a fractionated crude enzyme sediment extract. Fractionation was performed on a QAE Cellulose ion exchange chromatography column. The symbol ■ represents the concentration of protein as μg protein per 50 μl aliquot of each fraction. The symbol • represents relative TNT-reducing activity. The symbol ▲ represents relative 4- chlorobenzonitrile-reducing activity.

Figure 2 is a graph of relative reducing activity of the void volume fractions of Figure 1 after fractionation on a Sepharose™ CL-6B size exclusion chromatography column. The symbol ■ represents the concentration of protein as μg protein per 50 μl aliquot of each fraction. The symbol • represents relative TNT-reducing activity. The symbol ▲ represents 4-chlorobenzonitrile-

reducing activity. The open square symbol represents relative PCE-reducing activity.

Figure 3 is a graph of relative reducing activity of fraction 18 of Figure 2 after fractionation on a Phenyl Sepharose™ hydrophobic interaction chromatography column. The symbol ■ represents the concentration of protein as μg protein per 50 μl aliquot of each fraction. The symbol • represents relative TNT-reducing activity. Figure 4 is a graph of relative reducing activity of fraction 18 of Figure 2 after fractionation on a Zn:Iminodiacetic Acid Sepharose™ 6B chromatography column. The symbol ■ represents the concentration of protein as μg protein per 50 μl aliquot of each fraction. The symbol • represents relative TNT-reducing activity. The symbol ▲ represents 4-chlorobenzonitrile-reducing activity.

Figure 5 is a bar graph of percent TNT-reducing activity inhibition after addition of protease to fraction 18 of Figure 2.

Figure 6 is a graph of relative reducing activity for fractions 4, 13, and 19 from Figure 2 after exposure to various temperatures. The symbol • represents fraction 4. The symbol ▲ represents fraction 13. The symbol ■ represents fraction 19.

Figure 7 shows chemical structures for the nitroorganics Disperse Blue 79™, Disperse Red 5™, Parathion™, Dinoseb™, RDX™, HMX™, and TNT. Figure 8 is a schematic representation of the stepwise reduction of substituted nitrobenzenes to the corresponding anilines in anaerobic sediments.

Figure 9 is a graph of the reduction of nitrobenzene to nitrosobenzene, phenylhydroxyamine and aniline after incubation under anaerobic conditions with a sediment sample containing a nitrobenzene reducing agent. The black triangle

symbol represents nitrobenzene, the black square symbol represents phenylhydroxyamine, the black circle symbol represents nitrosobenzene, the open square symbol represents aniline, and the asterisk symbol represents the sum of components.

Figure 10 is a schematic representation of the stepwise oxidation of anilines to catechols, carbon dioxide and the corresponding acetates.

Figure 11 is a graph showing the relationship between the disappearance rate constant of nitrobenzene and the organic carbon content of sediment when nitrobenzene is incubated under anaerobic conditions with various sediment samples containing a nitrobenzene reducing agent. (R ² = 0.968; XI = 0.442; C - -2.54)

Figure 12 is a representative HPLC chromatogram showing the relevant retention times and separation of TNT from the reduction products of TNT after incubation with a TNT reducing agent. Figure 13 is a graph showing the rate of reduction of various concentrations of TNT in an anaerobic sediment sample. The symbol ■ represents an initial TNT concentration of 125 ppm. The symbol + represents an initial TNT concentration of 2.5 ppm. The open diamond symbol represents an initial TNT concentration of 0.25 ppm. (p = 0.12 ± 0.01; pH = 6.5; Eh = -368 mv (Ag/AgCl) )

Figure 14 is graph comparing TNT reduction kinetics in aquifer material after incubation with a protein extract containing a TNT reducing agent. The x-containing open square symbol represents TNT in aquifer material alone. The t _m is 70.6 hours. The x-containing open diamond symbol represents TNT plus protein extract. The t _m is 6.5 hours. The x- containing open triangle symbol represents TNT in aquifer material plus protein extract. The t _m is 4.2 hours.

Figure 15 is a graph showing the reduction of 1.6 ppm TNT in an aqueous solution of 1% iron (w/v) in the absence of reducing agent. The t _1/2 is 1.14 days and r ² = 0.99. Figure 16 is a graph showing TNT reduction in flooded contaminated soil samples. The first arrow indicates that iron was added after 48 days, and the second arrow indicates that the sample was inverted to mix the contents after 51 days. The solid square represents supernatant. The open triangle represents sediment. The open square represents Eh.

Figure 17 is a graph showing the enzymatic reaction of trichloroethane (TCE) as concentration divided by initial concentration versus time. The initial concentration was 97.0 μM. The black circle symbol represents the control. The dash symbol represents TCE.

Figure 18 is a graph showing reactivity (percent activity) of tetrabromoethene (PBE) and hexachloroethane (HCA) with protein as a function of temperature. The black circle symbol represents HCA. The black square symbol represents PBE.

Figure 19 is a graph showing reactivity (percent activity) of tetrabromoethene (PBE) and hexachloroethane (HCA) with protein as a function of pH. The black circle symbol represents HCA. The black square symbol represents PBE.

Figure 20 is a graph showing a decrease in the concentration of hexachloroethane (HCA) after reaction with the dehalogenase enzyme.

Figure 21 is a graph showing the Michaelis- Menton Kinetics for tetrabromoethene (PBE) .

Figure 22 is a graph showing a Lineweaver-Burke analysis of the degradation of TNT as the inverse of substrate concentration versus the inverse of initial velocities. The black square symbol

represents enzyme alone and the grey diamond represents enzyme in combination with NADPH. Figure 23 is a graph showing benzonitrile reducing activity versus fraction number during chromatographic isolation of the nitrilase enzyme using a metal affinity column.

Figure 24a is a graph showing a Lineweaver-Burke analysis of the degradation of the halogenated hydrocarbon tetrabromoethene (PBE) as the inverse of substrate concentration versus the inverse of initial velocities. Figure 24b is a graph showing a Lineweaver-Burke analysis of the degradation of the halogenated hydrocarbon hexachloroethane (HCA) as the inverse of substrate concentration versus the inverse of initial velocities.

Figure 25 is graph of concentration versus time in minutes of the reduction of trichloroethane (TCE) by the dehalogenase in the presence of either copper nitrate, ascorbic acid or both. The black square symbol represents the control, enzyme alone. The grey diamond symbol is ascorbic acid. The star symbol is copper nitrate and ascorbic acid. The open square symbol is copper nitrate. The open diamond symbol is the enzyme in the presence of copper nitrate and ascorbic acid. The asterisk symbol is the TCE products.

Figure 26a is a graph of hexachloroethane concentration versus time for the reduction of hexachloroethane by incubation with nitella, parrot feather, hornwort and algae spirogyra. Figure 26b is a graph of perchloroethene concentration versus time for the reduction of perchloroethene by incubation with the same plants. For both Figure 26a and 26b, the black square symbol is the control, the black diamond symbol is nitella, the black triangle symbol is parrot feather, the open

square symbol is hornwort, and the open diamond symbol is algae spirogyra.

Figure 27 is a graph of concentration area versus time in hours showing degradation of hexachloroethane by enzyme-coated sand. The black square symbol is hexachloroethane (HCA) , the grey diamond symbol is trichloroethene (TCE) , the black triangle is tetrachloroethene (PCE) , and the open diamond is believed to be pentachloroethane (PC) .

Detailed Description of the Invention

Methods and compositions for remediating a contaminant in soil, water, sediment, and aquifers to environmentally safe products are provided.

The remediating protein is an enzyme or combination of enzymes found in soil or sediment and is generally found in greater concentrations in soil having a higher organic content. Preferably, the environmental contaminant is combined with soil, a crude soil extract containing an amount of the remediation enzymes, or semi- purified or purified remediation enzymes specific for the degradation of a selected class of compounds sufficient to cause remediation to an environmentally innocuous product. For remediation of compounds that can be reduced, such as nitroorganic compounds, the contaminant is first reduced by combining the contaminant with soil, a crude soil extract containing an amount of the reducing enzymes, or semi-purified or purified reducing enzymes specific for the reduction of a selected class of compounds sufficient to cause reduction. The reduced contaminant is then oxidized to an environmentally innocuous product by exposure to oxygen or air.

The method is applicable to both in situ and batch processing and can be used to remediate contaminants such as nitroorganics, halogenated hydrocarbons, anisoles, cyano compounds, and metals as discussed in more detail below.

Reduction of Soil or Sediment Contaminants

In the method described herein, contaminated soil or sediment is first deprived of oxygen to create a substantially anoxic or anaerobic environment. While an anoxic environment is not essential for degradation of the contaminants by the remediation proteins, it facilitates remediation by making the environment unfavorable to aerobic organisms. It has been discovered that aerobic organisms produce proteases that cleave and thereby inactivate the enzymes used in the remediation method provided herein.

In a preferred embodiment, the contaminated soil or sediment is deprived of oxygen by flooding the surface area of the soil or sediment with water. Contaminated sediment need not be flooded with water if it already exists in an anoxic state. The contaminated soil or sediment is then incubated for a sufficient period of time under the anoxic conditions to allow reduction of the contaminants by reducing agents either naturally present in the contaminated soil or sediment or by reducing agents added to the contaminated soil or sediment in the form of a crude enzyme extract or semi-purified enzyme extracted from soil. Mass transfer limitations can be overcome by mechanical mixing.

The term "remediation proteins" as used herein, refers to proteins that facilitate degradation of a compound to environmentally safe products and includes the term "reducing agents", which, as used herein, refers to substances which facilitate the reduction of a compound. If the contaminated soil

or sediment has a relatively low carbon content or if the degree of contamination is extensive and the soil therefore fails to contain an adequate concentration of remediation proteins for rapid or complete remediation of the contaminants, a crude soil extract containing an adequate concentration of remediation enzymes for degradation of the contaminants, a semi-purified or purified enzyme or enzymes specific for degradation of the particular contaminant to be remediated, or a combination thereof may be added to the contaminated soil or sediment for more rapid and complete contaminant remediation. Soil or sediment having a "substantially high organic or carbon content" as used herein generally excludes soils containing predominantly sand or clay. For example, sediment collected from the Beaver Dam stream, near Athens, Georgia, having a organic carbon content of 3% contains approximately 100 μg protein per liter of sediment. The amount of crude or purified enzyme added depends on the compound being remediated and the activity of the enzyme. For example, 1 μg of protein reduces 10" ⁶ M TNT in approximately 15 minutes. The extent of incubation depends on the carbon content of the soil or sediment which directly relates to the concentration of remediation enzymes in the soil and the amount of contamination as described above. In general, the higher the organic carbon content, the faster the contaminants are reduced. Preferably, the soil or sediment is incubated from one to four days. The addition of iron, preferably in the form of iron powder or filings provides an additional source of electrons, and, when added to the incubation mixture, greatly accelerates the reduction reaction.

Preferably, the soil or sediment being remediated is maintained in the pH range of 5 to 8 for both the oxidation and reduction steps. The heterogenous soil and water phase provides the pH buffering capacity. The Eh of the system is preferably -50 mV or less (relative to Ag/AgCl) . In addition, the temperature of the contaminated area during remediation preferably ranges from 10 to 115°C with the optimal temperature ranging between 25 and 37°C.

The sediment or flooded soil preferably contains a soil to water ratio ranging from 0.02:1 to 0.5:1 (g:g) . It has been discovered that, in batch processes, the rate constants for reduction increase with increasingly greater soil to water ratios up to a ratio of 0.5:1. Reduction of Water or Aquifer Material Contaminants

In the method described herein for remediation of contaminated water or aquifers, a sufficient amount of soil having a substantially high organic carbon content, a crude soil extract containing reducing enzymes, or a semi-purified or purified enzyme specific for reduction of the contaminant is combined with the contaminated water or aquifer material for reduction of the contaminants. As described above, the enzyme-containing soil and the water can be mixed by mechanical means to overcome mass transfer limitations. Contaminated water and aquifers are naturally anoxic and need no removal of oxygen during the incubation step.

The pH and Eh values, soil to water ratio, and extent of incubation are preferably the same as described above for the reduction of soil and sediment contaminants. As described above, the addition of iron to the incubation mixture, preferably in the form of iron powder or filings, greatly accelerates the reduction reaction.

Isolation of Remediation Agents a) Extraction

The crude enzyme extract and semi-purified or purified enzyme described above for reduction of soil, sediment, water, or aquifer material contaminants are extracted from a soil sample with one or more of a variety of different solvents. Preferably, the soil sample has a substantially high organic carbon content and therefore contains a concentrated amount of the remediation protein or proteins as described above. Most preferably, the soil sample is a sediment rich in organic material such as the type of sediment found in a marsh or bog. The source of the remediation protein, a degradation enzyme, is a plant, preferably a common aquatic plant, or weed. The source of the nitroreductase is preferably the hornwort moss. Other useful nitroreductase sources include the stonewort, the roots and leaves of the soft rush and the leaves of the waterwort. The source of the nitrilase is preferably the hornwort moss, stonewort, or oak leaves. The source of the dehalogenase is preferably water oak leaves, dogwood leaves, or weed quakegrass. Other useful dehalogenase sources include parrot feather, stonewort and algae spirogyra.

Therefore, the soil from which the nitroreductase or nitrilase is extracted should contain hornwort or should have previously supported the growth of the hornwort. The hornwort is a group of common aquatic weeds floating on ponds in warm weather areas. Similarly, the soil from which the nitroreductase or nitrilase is extracted should contain water oak leaves, dogwood leaves, weed quakegrass, parrot feather or

stonewort or should have previously supported the growth of such plants.

Optimal enzymatic yield is obtained when the soil sample is removed from beneath a water surface and is stored and transported under substantially anoxic conditions to limit proteolytic cleavage by aerobic organisms naturally present in the sample.

The crude enzyme extract is prepared from the soil sample by gently mixing approximately one volume of the sample with one or more volumes of solvent and then separating the solubilized proteins from the soil by mechanical methods known to those skilled in the art, such as centrifugation. Any solvent or mixture of solvents in which proteins are solubilized may be used to solubilize the remediation protein. A list of solvents useful for solubilizing the remediation protein, or reducing reagent, is set forth below in Table 1. Preferred solvents contain glycerol or methanol. The most preferred solvent is 20% glycerol in either water or a buffer, such as Tris or phosphate at pH 8. High quality solvents such as commercial high pressure liquid chromatography (HPLC) grade solvents are preferred to minimize additional contamination. HPLC grade solvents can be obtained from commercial chemical suppliers such as Burdick and Jackson, Musegon, MI. The solvents sodium dodecyl sulfate (SDS) , CTMA BR (cetyltrimethylammonium bromide) , Triton X-100 and Tween 20 are detergents, whereas ammonium sulfate, urea, potassium phosphate and ammonium phosphate are salts.

Table 1. Solvents useful for Extracting Reducing Reagent

0.5% sodium dodecyl sulfate (SDS) in water

0.5% cetyltrimethylammonium bromide (CTMA Br) in water

0.5% Triton X-100 in water

0.5% Tween 20 in water

20% glycerol in water

20% glycerol + 0.5% SDS in water

20% glycerol + 0.5% CTMA Br in water 20% glycerol + 0.5% Triton X-100 in water 20% glycerol + 0.5% Tween 20 in water 500 mM KC1 500 mM KC1 0.5% SDS in water 500 mM KC1 0.5% CTMA Br in water 500 mM KC1 0.5% Triton X-100 in water 500 mM KC1 0.5% Tween 20 in water 2 M NH ₄ S0 ₄ 2 M NH ₄ S0 ₄ + 0.5% SDS in water 2 M NH ₄ S0 ₄ + 0.5% CTMA Br in water 2 M NH ₄ S0 ₄ + 0.5% Triton X-100 in water 2 M NH ₄ S0 ₄ + 0.5% Tween 20 in water 6 M urea 6 M urea + 0.5% SDS in water 6 M urea + 0.5% CTMA Br in water 6 M urea + 0.5% Triton X-100 in water 6 M urea + 0.5% Tween 20 in water

K ₂ HP0 ₄ + KH ₂ P0 ₄ PH ( 0. 5 M) K ₂ HP0 ₄ + KH ₂ P0 ₄ PH (0. 5 M) + 0.5% SDS in water K ₂ HP0 ₄ KH ₂ P0 ₄ pH ( 0. 5 M) + 0.5% CTMA Br in water K ₂ HP0 ₄ KH ₂ P0 ₄ pH ( 0. 5 M) + 0.5% Triton X-100 in water K ₂ HP0 ₄ KH ₂ P0 ₄ PH ( 0. 5 M) + 0.5% Tween 20 in water

0 . 5 M NH ₄ HP0 ₄ 0. 5 M NH ₄ HP0 ₄ + 0.5% SDS in water 0. 5 M NH ₄ HP0 ₄ + 0.5% CTMA Br in water 0. 5 M NH ₄ HP0 ₄ + 0.5% Triton X-100 in water 0 . 5 M NH ₄ HP0 ₄ + 0.5% Tween 20 in water 20% glycerol + 50 mm Tris-HCl, pH 8.0 in water

Most preferably, the sample and solvent mixture is allowed to stand for several minutes until the majority of the particular matter of the sample has precipitated and the proteins have been solubilized. The supernatant is decanted and centrifuged for approximately 20 minutes at 5000 rpm to remove the finer particulate matter. The extraction and centrifugation steps can be repeated

to increase yield. The resulting supernatant is a crude enzyme extract that can be used to reduce environmental contaminants without further purification if desired. b) Protein Precipitation

The crude enzyme extract can be concentrated and further purified by precipitating the proteins from the extract. Salt precipitation with a concentration sufficient to achieve 80-90% saturation with a salt such as ammonium sulfate is preferred. c) Desalting

A semi-purified enzyme extract can be prepared by resuspending the precipitate in water and simultaneously removing the salt and fractionating the proteins on a desalting or size exclusion column such as a Sephadex™ G-25sf column, available from Pharmacia, Inc., Piscataway, NJ. The Sephadex™ G-25sf column fractionates molecules having a molecular weight greater than 5000, therefore the fraction eluting in the void volume of the G-25 column contains molecules having a molecular weight greater than 5000. This semi-purified high molecular weight enzyme fraction contains most of the reducing activity for all classes of contaminants remediated by the method described herein except metals. Therefore, the enzymes useful for remediation of nitroorganics, halogenated hydrocarbons, anisoles, and cyano compound contaminants all have a molecular weight greater than 5000. Metal contaminants, such as K ₂ Cr0 ₄ , are reduced by the low molecular weight fraction from the G-25 column (less than 5000 daltons) or the total extract. Therefore, the enzyme or compounds specific for reduction of metal contaminants have a molecular weight less than 5000 daltons.

d) Additional Purification

The enzyme extract can be further purified using conventional protein purification techniques known to those skilled in the art, alone or in combination, for the isolation of substantially pure enzyme. Protein purification techniques include separation of the active enzyme by charge fractionation using an ion exchange column such as an QAE anion exchange or DEAE Sepharose™ anion exchange column (Pharmacia, Piscataway, NJ) ; by size fractionation using a size exclusion column such as a CL-6B Sepharose™ column (Pharmacia) ; by hydrophobicity fractionation using a phenyl column such as a phenyl Sepharose™ CL-4B column (Pharmacia) ; and by amino acid metal binding properties using a zinc affinity column such as a Zinc Iminodiacetic Acid-Sepharose™ 6B Fast Flow Column (Sigma Chemical Co., St. Louis, MO). Enzymes specific for the reduction of nitroorganics, halogenated hydrocarbons, anisoles, cyano compounds and metals can be isolated from the crude enzyme extract using a combination of the above-mentioned protein fractionation techniques as described in more detail below. Fractions are assayed for activity as described below, and active fractions are pooled prior to fractionation by a subsequent separation mechanism. e) Assay for Reducing Activity Fractions collected during each step of the enzyme purification method described herein can be assayed for remediation activity by combining an aliquot of each fraction with a solution containing a compound from the class of contaminants to be remediated. The mixture is incubated and assayed for the presence of degraded or reduced contaminants by analytical methods known to those skilled in the art, such as HPLC and GC.

Protein concentrations for each fraction can be determined using commercially available reagents such as the BioRad™ Protein Reagent (BioRad, Richmond, CA) in accordance with the manufacturer's instructions or by conventional protein analysis methods well known to those skilled in the art. f) Isolation of Remediation Protein with Monoclonal Antibodies

The remediation protein, or a particular enzyme specific for the degradation or reduction of a particular class of contaminants such as a nitroreductase, nitrilase, demethylase, or dehalogenase can be isolated from a soil sample, or a soil sample can be tested for the presence of such an enzyme, with the aid of a detectable polyclonal or monoclonal antibody in accordance with methods well known to those skilled in the art. The production of monoclonal antibodies specific for a particular enzyme or protein is also achieved by methods well known to those skilled in the art, such as the methods described in Antibodies - A Laboratory Manual , Ed Harlow, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 1988. Basically, monoclonal antibodies are produced from a hybridoma cell generated by fusing a normal antibody-producing lymphocyte from the spleen of an experimental animal, such as a mouse or rat, recently immunized with the protein of interest, to a myeloma cell line that does not synthesize its own immunoglobulin and is deficient in the enzyme hypoxanthine-guanine phophoribosyltransferase (HGPRT) , which catalyzes reactions of the bases hypoxanthine and guanine with 5-phophoribosyl-l- pyrophosphate to form the nucleotides inosine-5 ^, -P (IMP) and guanosine-5 ^, -P (GMP) , respectively. The hybrid cells are selected over either parental cell

by culturing in a medium containing hypoxanthine, aminopterin, and thymidine (HAT medium) . The myeloma cells, lacking HGPRT, cannot survive because the de novo synthesis of GMP is blocked by the folate antagonist aminopterin, and the normal lymphocytes grow very slowly in HAT medium, whereas the hybrid cells, in which a functional HGPRT gene is supplied by the lymphocyte genome, grow rapidly and form large colonies that are readily distinguishable from those of the slowly growing lymphocytes. Clones of hybrid cells producing the desired antibody are identified by a suitable assay procedure and grown into larger cultures. The homogenous immunoglobulin produced by such a cloned hybrid is a monoclonal antibody having specificity for the enzyme.

The monoclonal antibody can be made detectable by attachment of a detectable label. The various types of labels and methods of labelling monoclonal antibodies are well known to those skilled in the art. Several specific labels are set forth below. For example, the label can be a radiolabel such as, but not restricted to, ³² P, ³ H, ¹⁴ C, ³⁵ S, ¹²⁵ I, or ¹³¹ 1. Detection of a radiolabel can be by methods such as scintillation counting, gamma ray spectrometry or autoradiography.

The label can also be a Mass or Nuclear Magnetic Resonance (NMR) label such as, for example, ¹³ C, ^,5 N, or ¹⁹ 0. Detection of such a label can be by Mass Spectrometry or NMR.

Fluorogens can also be used to label the monoclonal antibodies. Examples of fluorogens include fluorescein and derivatives, phycoerythrin, allo-phycocyanin, phycocyanin, rhodamine, Texas Red or other proprietary fluorogens. The fluorogens are generally attached by chemical modification.

The fluorogens can be detected by a fluorescence detector.

The monoclonal antibodies can alternatively be labelled with a chromogen to provide an enzyme or affinity label. For example, the antibody can be biotinylated so that it can be utilized in a biotin-avidin reaction which may also be coupled to a label such as an enzyme or fluorogen. The probe can be labelled with peroxidase, alkaline phosphatase or other enzymes giving a chromogenic or fluorogenic reaction upon addition of substrate. For example, additives such as 5-amino-2,3-dihydro-l,4-phthalazinedione (also known as Luminol™) (Sigma Chemical Company, St. Louis, MO) and rate enhancers such as p-hydroxybiphenyl (also known as p-phenylphenol) (Sigma Chemical Company, St. Louis, MO) can be used to amplify enzymes such as horseradish peroxidase through a luminescent reaction; and luminogeneic or fluorogenic dioxetane derivatives of enzyme substrates can also be used.

A label can also be made by detecting any bound antibody complex by various means including immunofluorescence or immuno-enzymatic reactions. Such labels can be detected using enzyme-linked immunoassays (ELISA) or by detecting a color change with the aid of a spectrophotometer. Remediation Protein Characteristics

The active remediation protein consists of one or more enzymes believed to be derived from a plant, preferably an aquatic plant such as the hornwort, soft rush, waterwort, oak leaves, dogwood leaves, weed quakegrass, parrot feather and algae spirogyra. The enzyme is soluble in water and may be a membrane-bound enzyme complex or portion thereof, possibly a glyco-protein. It is believed that the enzyme is an electron carrier that

transfers electrons to the contaminants in the presence of an electron source (anoxic conditions) and removes electrons from the contaminants in the presence of an electron sink (oxic conditions) . The remediation protein extracted from soil or sediment as described herein is abiotic in that neither microbial activity nor metabolism is required. Furthermore, with the exception of the dehalogenase enzyme described in more detail below, which is oxygen sensitive, the remediation protein does not require strict anaerobic conditions for activity. Treatment of the remediation protein with protease K, a digestive enzyme that hydrolyses peptide bonds in polypeptide chains, decreases the ability of the remediation protein to degrade contaminants thereby demonstrating that the active component of the remediuation protein is a protein. Isolation and Characterization of the Nitroorganic Reducing Agent An enzyme specific for reduction of nitroorganic compounds was isolated from the semi-purified enzyme extract described above by the following procedure. Charge Void volume fractions from the Sephadex™ G-25sf column containing proteins having a molecular weight greater than 5000 daltons were collected and pooled. The pooled fractions were concentrated with a Zetaprep™ 15 disk (Pharmacia, Piscataway, NJ) , adjusted to pH 8.6 with a 100 mM Tris buffer, and subjected to low pressure ion exchange fractionation by passage through a QAE or DEAE Sepharose™ anion exchange column. The mobile phase was 100 mM Tris buffer, pH 8.6. Under these conditions, fractions having a neutral or positive charge at pH 8.6 passed through the column into the void volume while fractions having a negative

charge were retained on the column. Fractions were collected and assayed for protein concentration and the ability to reduce the nitroorganic compounds TNT and 4-chlorobenzonitrile as described above. The highest levels of activity for reducing both TNT and 4-chlorobenzonitrile were found in the fractions having a neutral or positive charge at pH 8.6 (void volume fractions) as shown in Figure 1. Size The fractions having high reducing activity were pooled, precipitated to reduce the volume, resuspended in 50 mm Tris, at pH 7, and subjected to fractionation by size exclusion chromatography by passage through a Sepharose™ CL-6B column. The mobile phase was 50 mm Tris, pH 7. Fractions were once again assayed for protein content and the ability to reduce the nitroorganic compounds TNT and 4-chlorobenzonitrile. The molecular weight range for each fraction was determined by comparing the elution time of the active fractions to the elution time of proteins standards having known molecular weights. Fractions containing molecules having a molecular weight between 300,000 and 400,000 daltons, between 49,000 and 80,000 daltons, and between 18,500 and 27,000 daltons exhibited the highest levels of TNT reducing activity as shown in Figure 2, while fractions containing molecules having a molecular weight of approximately 600,000 daltons exhibited the highest levels of 4-chlorobenzonitrile reducing activity and molecules having a molecular weight of approximately 400,000 to 600,000 daltons had the highest levels of 1,2,3,4-tetrachloroethene (PCE) reducing activity. Hvdrophobicitv

The fractions having the largest molecular weight molecules with TNT-reducing activity were

pooled, precipitated, reconstituted in 1.7 M ammonium sulfate, pH 7 and passed through a Phenyl Sepharose™ CL4B column for separation and fractionation on the basis of hydrophobicity. Molecules were eluted with a linear gradient of 1.7 M ammonium sulfate to 1.7 M ammonium phosphate, pH 7. The gradient was begun after five fractions were collected. Under these conditions, hydrophilic components pass through the column quickly while hydrophobic components are retained and are eluted as the concentration of ammonium sulfate in the mobile phase is reduced. Fractions were assayed for protein content and the ability to reduce TNT. The highest TNT reducing activity was observed in the hydrophobic fractions, with maximal reducing activity contained in fraction 32, as shown in Figure 3.

Metal-binding Ability

The active fractions were pooled, precipitated, reconstituted in 1 M potassium chloride and passed through a Zn Iminodiacetic acid-Sepharose™ CL-6B Fast Flow column. In this column, any material that does not bind zinc is eluted early. The sample was loaded onto the column in 1 M potassium chloride in 50 mM K ₂ P0 ₄ , washed with five void volumes of 50 mM potassium phosphate buffer, pH 7, containing 1 M KC1 and eluted with a linear gradient from 1 M potassium chloride to 2 M ammonium sulfate in phosphate buffer, pH 7. As shown in Figure 4, fractions containing TNT- reducing activity are found in fractions 16-18 and bind zinc to some extent, whereas fractions containing 4-chlorobenzonitrile-reducing activity are found in fractions 6-8 and bind zinc to a lesser extent. A metals analysis of the active fractions revealed that the TNT-reducing enzyme

contains iron whereas the 4-chlorobenzonitrile reducing enzyme does not contain any metals. Enzymatic Digestion

Treatment of an aliquot from pooled fractions 16-18 from Figure 4, having TNT-reducing or fractions 6-8, having 4-chlorobenzonitrile-reducing activity, with increasingly larger concentrations of pronase E, protease K or subtilisin caused inhibition of TNT-reducing activity, as shown in Figure 5. Treatment with the proteases chymotrypsin and trypsin, known to cleave proteins only at aromatic amino acids, caused no inhibition of TNT reducing activity. Therefore, the TNT reducing enzyme has few available aromatic amino acids. The activity of the nitrilase was inhibited by both trypsin and chymotrypsin. Temperature Effects

Figure 6 shows the effect of elevated temperature on low, intermediate, and high molecular weight fractions eluted from the

SepharoseTM CL-6B size exclusion column. (The molecular weights of these fractions are shown in Figure 2.) Elevation of temperature had little effect on the high molecular weight fraction (fraction 4) , moderate effect on the intermediate molecular weight fraction (fraction 13) and a greater effect on the low molecular weight fraction (fraction 19) . Kinetics The kinetics of the purified TNT-reducing agent and the purified nitrile-reducing agent, isolated as pooled fractions 16-18 from the Zn-Iminodiacetic acid-SepharoseTM CL-6B column, respectively, are set forth below in Table 2.

Table 2. Reducing Agents Kinetics

Compound (Concentration) Half-life

TNT (1.5 X 10-6) 15 minutes

TNT (3.0 X 10-6) 30 minutes

TNT (7.5 X 10-6) > 5 hours nitrobenzene (1.0 X 10-6) 28 hours

Molecular Weight

The size of the TNT-reducing enzyme, as determined by polyacrylamide gel electrophoresis, is approximately 316,000 daltons and is believed to contain six subunits having a molecular weight of approximately 19,000, two subunits having a molecular weight of approximately 37,000, two subunits having a molecular weight of approximately 66,000, and at least two iron molecules. The TNT- reducing enzyme exhibits a dumbbell shape when viewed by scanning tunnelling microscopy.

The size of the 4-chlorobenzonitrile-reducing enzyme has a molecular weight of approximately 600,000 and is believed to be composed of four subunits each having a molecular weight of approximately 150,000.

Subsequent experiments, set forth below in Example 6, confirmed and futher revealed that the nitroreductase has a native size of 316 kD; subunits of 66, 37 and 19 kD; contains four sulfur and, most likely, four iron molecules; a temperature optimum of 20-37°C; a pH optimum of 6.5 to 8.5; general protease sensitivity, but no sensitivity to trypsin or chymotrypsin; is inhibited by zinc, lead and low pH; uses NADPH and cannot be replaced by NADH; and the reaction rate is faster when bound to solids.

Isolation and Characterization of the Halogenated Hydrocarbon Remediation Protein

An enzyme specific for remediation of halogenated hydrocarbons, or a dehalogenase, was isolated from the semi-purified enzyme extract described above by the procedure described above for the isolation of the nitroorganic reducing agent with the exception that all the buffers were kept anaerobic. Sequential dehalogenation of a halogenated hydrocarbon with the dehalogenase results in the formation of the dehalogenated hydrocarbon. The final product appears to be carbon dioxide, a compound that is acceptable to the environment. Charge

As described above, void volume fractions from the Sephadex™ G-25sf column containing proteins having a molecular weight greater than 5000 daltons were collected, pooled, concentrated and subjected to low pressure ion exchange fractionation by passage through a QAE or DEAE Sepharose™ anion exchange column so that fractions having a neutral or positive charge at pH 8.6 passed through the column into the void volume while fractions having a negative charge were retained on the column.

Fractions were collected and assayed for protein concentration and the ability to reduce the halogenated hydrocarbon tetrabromoethene (TBE) as described above. The highest levels of activity for reducing TBE were found in the fractions having a neutral or positive charge at pH 8.6 (void volume fractions) as shown in Figure 1. Size The fractions having high reducing activity were pooled, fractionated by size exclusion chromatography by passage through a Sepharose™ CL-6B column as described above, and once again

assayed for protein content and the ability to reduce the halogenated hydrocarbon TBE. The molecular weight range for each fraction was determined as describe above. Fractions containing molecules having a molecular weight of approximately 293,000 daltons exhibited the highest levels of TBE reducing activity.

Metal-binding Ability

A metals analysis of the active fractions revealed that the dehalogenase enzyme contains copper and iron.

Temperature and pH

The purified dehalogenase exhibited maximal activity at a temperature between approximately 10 and 50°C and at a pH between approximately 3 and 7. The maximal activity for HCA dehalogenation was observed at a temperature between approximately 10 and 50°C, as shown in Figure 18, and at a pH between approximately 3.8 and 6.0, as shown in Figure 19, whereas the maximal activity for PBE dehalogenation was observed at a temperature between approximately 25 and 50°C, as shown in Figure 18, and at a pH between approximately 5.0 and 7.0, as shown in Figure 19. Kinetics

The kinetics for the reduction of the following halogenated alkanes and alkenes: tetrabromoethene (PBE) , hexachloroethane (HCA) , tetrachloroethene (PCE) and trichloroethene (TCE) , by the purified dehalogenase, using microgram quantities of protein in and near saturated solutions of the halogenated hydrocarbons, are set forth below in Table 3.

Table 3. Dehalogenase Remediation Protein Kinetics

Compound (Concentration) Half-life

PBE ( 30 μM) 1100 minutes

HCA ( 30 μM) 200 minutes

PCE ( 60 μM) 200 minutes

PCE (100 μM) 600 minutes

TCE ( 60 μM) 200 minutes

A graph showing a reduction in the concentration of TCE by the dehalogenase over time is shown in Figure 17. A graph showing reduction of HCA by the dehalogenase over time is shown in Figure 20. The Michaelis-Menton kinetics for reduction of PBE are shown in Figure 21.

It was observed that ascorbic acid functions as a co-factor for the dehalogenase. In addition, competition experiments revealed that the dehalogenase preferentially reduces alkenes over alkanes.

Molecular Weight

The size of the dehalogenase, as determined by polyacrylamide gel electrophoresis, is approximately 293,000 daltons and is believed to contain two subunits having a molecular weight of approximately 90,000 daltons, two subunits having a molecular weight of approximately 39,000 daltons, two subunits having a molecular weight of approximately 28,000 daltons, and one or more copper and iron molecules.

Subsequent experiments confirmed and further revealed that the dehalogenase has the following physical characteristics: a native size of 293 kD; subunits of 90, 39, and 28 kD; one copper molecule per mole and as many as 48 iron molecules per mole; a temperature optimum of 10-50°C (HCA) and 25-50°C (PBE) ; ability to degrade both alkenes and alkanes in mixtures, preferably alkenes; ascorbic acid is a cofactor; and sensitivity to oxygen.

Isolation and Characterization of the Nitrilase Remediation Protein

An enzyme specific for degradation of nitriles, such as benzonitrile and acetonitrile was isolated from the semi-purified enzyme extract described above by the procedure described above for the isolation of the nitroorganic reducing agent and as set forth below in more detail in Example 6.

The nitrilase has the following characteristics: a native size of 660 kD; a subunit size of 150 kD; no detectable iron, zinc, molybdenum or copper molecules, optimum temperature of 15 to 45°C, sensitivity to trypsin, chymotrypsin, and general proteases; and no cofactors were required for activity.

Oxidation of Reduced Contaminants

Contaminants that have been reduced by the reducing agent as described above are oxidized to environmentally safe compounds by the addition of oxygen. Oxygen is preferably added by bubbling air into the incubation mixture. Exposure of the reduced contaminants to oxygen can also be achieved, for contaminated soil or sediment that has been flooded with water, by simply removing the water by evaporation or other methods known to those skilled in the art.

The remediation method can be carried out in a variety of reactors including batch reactors. Alternatively, the contaminated soil can be remediated in situ without removing the soil from the ground. Classes of Compounds Degraded

The remediation method can be used to degrade a wide variety of environmental contaminants including, but not limited to, nitroorganic compounds, halogenated organic compounds, cyano compounds, anisoles, and metals. It will be

understood by those skilled in the art that the remediation method may be used to degrade other contaminants that are similarly reduced and oxidized. a) Nitroorganic Compounds

The remediation method provided herein is particularly useful in reducing nitroorganic compounds contaminating soil, sediment, water or aquifer materials. Nitroorganic compounds are defined herein as nitroaromatics and nitroaliphatics and specifically include munitions such as trinitrotoluene (TNT) , hexahydro-1,3,5-trinitro-l,3,5-triazine (RDX™) , octahydro-1,3,5,7-tetranitro-l,3,5,7-tetraazocine (HMX™) , N-methyl-N-2,4,6-tetra-nitroaniline

(Tetryl™) , nitrocellulose, and munition processing wastewater such as the TNT manufacturing by-product known as "red water". The described remediation method is also useful for remediation of nitroorganic pesticide contaminants such as methyl parathion and 2-(sec-butyl)-4,6-dinitrophenol, also known as Dinoseb™. In addition, the remediation method is useful for remediating nitrobenzenes, benzonitriles such as 4-chlorobenzonitrile, Disperse Blue 79™, Disperse Red 5™, and azo compounds such as azobenzene.

The term "azo compound" as used herein refers to a compound containing a -N=N- moiety. Disperse Blue 79™, Disperse Red 5™, and azo-compounds, are industrial chemicals.

The chemical structures for Disperse Blue 79™, Disperse Red 5™, Parathion™, Dinoseb™, RDX™, HMX™, and TNT are shown in Figure 7.

Figure 8 shows chemical formulas for reduction of several nitroaromatic compounds by reducing agents in soil to the corresponding anilines in an anaerobic environment. Oxidation of anilines

produces catechols which are degraded to carbon dioxide and the corresponding acetates, as shown in Figure 10. b) Other Compounds The method provided herein is useful for remediation of halogenated hydrocarbons contaminating soil, sediment, water or aquifer materials. Halogenated hydrocarbons are defined herein as halogenated organic compounds or solvents such as hexachloroethane (HCA) , tetrachloroethene (PCE) , tetrabromoethene (PBE) , trichloroethene (TCE) and trichloroethylene, halogenated pesticides and other industrial chemicals such as halogenated aromatics and pentachlorophenol (PCP) . The method provided herein is useful for remediating soil, sediment, water or aquifer materials contaminated by cyano compounds such as benzonitrile, acetonitrile, and other industrial chemicals; anisoles such as anisole, dyes and pesticides containing methoxy moieties; and metals such as chromium and arsenic, especially K ₂ Cr0 ₄ . Immobilized Enzymes for Use in Remediation

It will be understood by those skilled in the art that the reducing enzymes may be bound to a solid phase, sush as glass beads or sand, to facilitate administration of the enzymes to a contaminated site for in situ remediation.

A major portion of the reducing enzymes isolated from sediment are glycoproteins. The carbohydrate portion of the enzyme can be used to bind the protein to a solid phase or matrix, such as glass beads, sand, or aquifer material. The carbohydrate can be converted to an aldehyde by the use of a periodate in a pH of 8.5 to 9.25. The matrix can be coated with a polymer such as poly-lysine, then covalently bound to the modified glycoprotein by the addition of sodium cyanoborohydrate or sodium

borohydrate in the same high pH buffer in accordance with the methods described by Hermanson et al., IMMOBILIZED AFFINITY LIGAND TECHNIQUES, Academic Press, San Diego, CA. The activity of the enzyme dehalogenase can be used for up to two months.

A diamine, such as histidine or imidazole or a compound such as tryptophan can be conjugated to the enzyme in a similar manner as described above for use in a flow-through system or if the solid matrix is not easily treated with poly-lysine. The binding or release of the enzyme is pH dependent. The diamine can be designed according to the pH of the site undergoing remediation. The pH of the active dehalogenase protein is 2.5 to 6. Therefore, the binding needs to be effective at those pH values.

The remediation method will be further understood with reference to the following non-limiting examples.

Example 1. In Vitro Reduction of para-substituted nitrobenzenes by Six Anaerobic Sediments

A series of para-substituted nitrobenzenes were shown to undergo abiotic reduction to the corresponding anilines in several anaerobic sediment samples.

Materials

Nitrobenzene, 4-nitrotoluene, 4-nitroanisole, 4-chloronitrobenzene, 4-bromonitrobenzene, 4-nitrobenzonitrile, 1,4-dinitrobenzene, 4-toluidine, 4-anisidine, 4-chloroaniline, 4-bromoani1ine, 4-aminoacetophenone, 4-aminobenzonitrile, 4-nitroaniline, catechol, nitrosobenzene (all at least 97% pure; Aldrich Chemical Co. , Inc. , Milwaukee, Wl) , 4-nitroacetophenone (97% pure; Pfaltz and Bauer, Inc., Waterbury, CT) , aniline (99.5% pure; Fluka

Chemical Corp. , Hauppauge, NY) were used without further purification. Phenylhydroxylamine, m.p. 80-82°C, was synthesized by reducing nitrobenzene with zinc in accordance with the method of Kamm, 0., Org. Syn . 4:57-58 (1924) and recrystallized from hexane. The identity and purity of these compounds was confirmed by HPLC and gas spectrometry-mass spectrometry (GC-MS) using a Finnigan MAT (Finnigan, San Jose, CA) automated gas chromatograph/EI mass spectrometer. Acetonitrile (Burdick and Jackson, Musegon, MI) , sodium hydroxide (Fisher Scientific Co., Pittsburgh, PA), nitric acid and hydrochloric acid (J. T. Baker Chemical Co., Phillipsburg, NJ) were of high purity.

Sediment

Anaerobic sediments were collected from several different sites. Three of these sediments were collected near Athens, Georgia from a slow moving river (the Oconee River) , a slow moving high sediment load stream (Beaver Dam) , and a small pond (Bar H pond) . A fourth sediment was collected from a bog near Gaston, Illinois (Morgan's Muck) . A fifth sediment was collected from an uncontaminated aquifer near Quincy, Florida (Florida Aquifer) .

The sixth sample was collected from a peat bog near Bilthoven, The Netherlands (Loosdrechtsche Plassen) . The sediments and associated water, except the Florida aquifer sample, were collected by scooping up the first 5 to 10 cm of bottom sediment into 1 liter canning jars at a depth of 30 to 50 cm below the water surface. The jars were capped before being brought to the surface. The samples were sieved through a 1 mm sieve in the laboratory to remove debris and stored in a glove box in a nitrogen atmosphere until used. The aquifer sample was collected using a hollow stem

auger with a split spoon sampler in a fresh drilled well 15 to 20 meters deep. The sediment was covered with groundwater, transported to the laboratory and stored in a glove box under a nitrogen atmosphere until used.

Sediment Eh readings were taken in the glove box with a Markson 1202 combination platinum electrode that was placed in the sediment. (Markson, Houston, TX) The electrode was calibrated against a standard ferrous-ferric poised standard (Light et al. , 1972) . If the electrode did not read to within ± 10 mV of Eh=750 mV, the electrode was cleaned with 10% HC1 (v/v) followed by 10% H ₂ 0 ₂ (v/v) . The measurement was taken when the Eh reading stabilized (approximately 20 min) . The fraction organic carbon, F _∞ , (w/w) was measured using air dried sediment by coulometric titration using an automated instrument (Dohrman, Santa Clara, CA) in accordance with the method of Lee, CM. and D.L. Macalady, Inter. J. Environ . Anal . Chem . 35:2219-225 (1988).

Sediment pH was measured in the glove box with a portable Corning pH/TempMeter 4 equipped with a Ross combination electrode (Orion Research, Inc. , Boston, MA) . After calibrating the pH meter with pH 4.00, 7.00, and 10.00 buffers (Fisher Scientific Co. , Pittsburgh, PA) , the pH was measured by inserting the electrode into a stirred sediment slurry. Stirring was stopped and the pH was recorded when the reading stabilized (approximately 10 min) .

Liquid Chromatography

A Waters 501 chromatographic pump equipped with a Kratos 757 variable wavelength UV-visible detector (Kratos Analytical Instruments, Ramsey, NJ) and a Rheodyne 7125 injector using a 20 μl sample loop (Rheodyne, Inc., Cotati, CA) was used

for reverse phase HPLC. Typically, the column was a pH stable Hamilton PRP-1 column, 250 mm long x 4.1 mm i.d., 10 μm particle size (Hamilton Co., Reno, NV) . The analytical column was protected with a cartridge guard column containing a PRP-1 cartridge, 10 mm x 4.6 mm i.d. , 10 μm particle size (Alltech Associates, Inc., State College, PA). Typically the mobile phase was acetonitrile:water (60:40, v/v) at pH 12 (adjusted with 10 N NaOH) , and the flow rate was 1.0 ml/min at 1500 psi.

These conditions were used for the analysis of both disappearance of the parent compound and product formation. The mobile phase for catechol analyses was acetonitrile:water (20:80, v/v) at pH 1.75 (adjusted with concentrated HN0 ₃ ) .

Detector wavelength varied with the compound of interest: nitrobenzene, 262 nm; 4-nitrotoluene, 272 nm; 4-nitroanisole, 310 nm; 4-chloronitrobenzene, 278 nm; 4-bromonitrobenzene, 278 nm; 4-nitroacetophenone, 262 nm; 4-nitrobenzonitrile, 272 nm; 1,4-dinitrobenzene, 262 nm; aniline, 233 nm; 4-toluidine, 235 nm; 4-anisidine, 232 nm; 4-chloroaniline, 240 nm; 4-bromoaniline, 240 nm; 4-aminoacetophenone, 230 nm; 4-aminobenzonitrile, 275 nm; 4-nitroaniline, 220 nm; catechol, 262 nm; nitrosobenzene, 320 nm; phenylhydroxylamine, 262 nm. Product formation was confirmed by GC/MS and quantitated by HPLC through comparison with known standards. The GC/MS system used for product formation was a Hewlett-Packard 5890 GC with a 15 m HP-1 0.2 mm capillary column with 0.33 μm film thickness connected to a Hewlett-Packard 5970 mass selective detector (Hewlett Packard Co. , Palo Alto, CA) . Distribution Coefficient

The soil to water ratio was measured by weighing triplicate aliquots of sediment-water slurry, then

evaporating at 110°C overnight and determining the dry weight. The distribution coefficient (K _d ) of the compounds were determined in conjunction with most kinetic experiments. This was accomplished by centrifuging spiked sediment-water samples after vortexing 1 min and removing a 3.0 ml aliquot of the aqueous phase. The aqueous phase was extracted with 1.0 ml acetonitrile. The remaining sediment phase was extracted with 1.0 ml acetonitrile and centrifuged. Both extracts were analyzed via HPLC. Corrections were made for the water remaining in the sediment phase. Kinetic Procedures Kinetic experiments were conducted by a batch method in which 5 ml of a stirred sediment-water slurry was pipetted into a series of 10 ml screw-capped test tubes. The tubes were spiked with 50 μl of a 1.0 X 10 ^~3 M solution of the desired nitrobenzene derivative in acetonitrile or methanol. The test tubes were capped with screw caps fitted with teflon or n-butyl rubber lined septa, removed from the glove box and vortexed for 10 seconds. All sediment sample manipulations except for the initial sieving were conducted within the glove box. The test tubes were incubated at the desired temperature (± 1°C) in a water or oil bath with gentle shaking. Kinetic studies conducted above 100°C were carried out in sealed glass ampoules. Periodically a test tube was sacrificed for analysis by adding 1.0 ml acetonitrile then vortexing 1 min. The tube was centrifuged (15 minutes at 4000 rpm) and the supernatant analyzed by HPLC. Typical sample recoveries were in excess of 95% (w/w) . To examine the role of pH, the pH of sediments was altered by the dropwise addition of either 10 N NaOH or concentrated HCl to within 0.5 pH units of the

desired value and allowed to equilibrate 24 hours. The pH was measured and adjusted again as necessary. The samples were not used until the desired pH had stabilized for 24 hours. Results

The eight para-substituted nitrobenzene compounds were reduced by the general stepwise reduction mechanisms shown in Figure 8. Nitrobenzene gave aniline as the stable product with phenylhydroxylamine and nitrosobenzene as reactive intermediates.

As shown in Table 4, the para-substituted nitrobenzene compounds were reduced when the pH of the sediments was between 5.3 and 10.55. However, at pH 3.85 and 2.50, no reduction of the para-substituted nitrobenzene compounds could be detected within experimental error. Additionally, sediment flocculation and agglutinization of the solids was observed at this low pH. At pH 7, K _d was 3.4 and Eh was 105 mV. The organic carbon fraction was 0.0166.

The disappearance rate constant for nitrobenzene correlated best with the organic carbon content of the solid phase as shown in Figure 11.

Table 4. Disappearance rate constants for nitrobenzene reduction in Beaver Dam sediment over pH range 2.5-10.6

Sediment/H ₂ 0

EH ratio ^.(lO-^- ¹ ) ^* corfl0 ⁴ m- ^, ) ^b

2.5 0.0288 no observed rxn 0.25 ± 0.025

3.85 0.0281 no observed rxn 0.12 ± 0.012

5.3 0.0433 4.0 ± 0.21 4.6 ± 0.24

7.0 0.0305 3.8 ± 0.18 4.2 ± 0.20

7.75 0.0314 7.7 ± 0.70 8.5 ± 0.77

9.15 0.0244 8.6 ± 0.57 9.3 ± 0.62

10.55 0.0520 6.6 ± 1.0 7.8 ± 1.2

* Observed first-order disappearance rate constant and standard deviation. ^b Calculated maximum rate assuming a 10% analytical method error.

Example 2. In Vitro Reduction of Nitrobenzene Compounds by Four Anaerobic Sediments

Nitrobenzene, 4-ethylnitrobenzene, 4-(n- butyl)nitrobenzene and 4-(n-octyl)nitrobenzene in anaerobic sediment samples were reduced to the corresponding anilines. Chemicals and Sediments Nitrobenzene (99+%) , 4-ethylnitrobenzene (99+%) , 4-ethylaniline (99+%), 4-(n-butyl)aniline (97%), and 4-(n-octyl)aniline (99%) from Aldrich Chemical Co., Milwaukee, WI, and reagent grade aniline from the J. T. Baker Chemical Co., Phillipsburg, NJ, were used without further purification. Formaldehyde (37%) solution, with 12% Methanol, from J.T. Baker Chemical Co.), hydrogen peroxide (30% solution from J.T. Baker Chemical Co.), mercuric chloride (Fisher Scientific Co., Pittsburgh, PA) , and isopropanol (spectrograde, Burdick and Jackson, Musegon, MI) were used as chemical sterilants. Sodium azide (Fisher Scientific Co.), toluene (Spectrograde, Burdick and Jackson) , and m-cresol (99+%) (Aldrich Chemical Co. Inc., Milwaukee, WI) were used as metabolic

inhibitors. All solvents used were of high purity grade from Burdick and Jackson.

4-(n-Butyl)nitrobenzene and 4-(n- octyl)nitrobenzene were synthesized by slowly adding a 3.0 M methylene chloride solution (2.5 ml) of the corresponding aniline to a stirred 0.7 M methylene chloride solution (45 ml) of 3-chloroperoxybenzoic acid (tech., 80-85%, from Aldrich Chemical Co. Inc.). After refluxing for one hour and cooling, the mixture was washed with 1.0 N NaOH, then 0.1 N HC1, then water, and then the organic layer was dried over sodium sulfate. The methylene chloride was evaporated under a nitrogen stream, leaving a yellow-orange liquid. NMR and IR spectra showed no remaining amino compound, and both nitro compounds gave a single peak via liquid chromatography. Yields for 4-(n- butyl)- and 4-(n-octyl)nitrobenzenes were 77 and 62%, respectively. Thioglycollate indicator was used to check anaerobic sterility. The thioglycollate indicator agar was made by bringing thioglycollate medium (14.5 g, Difco Laboratories, Detroit, MI), Bacto-Agar™ (5 g, Difco Laboratories), and 0.1% resazurin indicator (5 ml, resazurin from Sigma

Chemical Co., St. Louis, MO) to a boil in 500 ml of water, with subsequent cooling. The resazurin indicator turns the nutrient medium red if conditions become aerobic. Tryptone Glucose Extract Agar (Difco Laboratories) was used to assay for aerobic sterility. Sediments

Anaerobic sediment samples from four bodies of water near Athens, Georgia were used. Three of these (known as Hickory Hill, Memorial Park, and

Bar-H) were samples from lakes; the fourth (Beaver Dam) was from a stagnant, slow moving stream.

Sediments were collected by scooping up samples in 1 quart canning jars at a depth of 1-2 feet below the water surface. The jars were capped, brought back to the laboratory, passed thorough a 1 mm sieve, and stored in a glove box under a nitrogen atmosphere until used. Sediments were used within two weeks of collection. Liguid Chromatography A Tracor 950 chromatographic pump equipped with a Tracor 970A variable wavelength detector (Tracor Instruments, Austin, TX) and a Rheodyne 7120 injector with a 50 μl sample loop (Rheodyne, Inc., Cotati, CA) was used for liquid chromatography. The column was a Micromeritics Microsil™ C-18, 25 cm length x 4.6 mm I.D., 5 μm particle size

(Micromeritics Instrument Co., Norcross, GA) . A guard column (2 mm I.D. x 7 cm length) was filled first with a small amount of Microsil™ 5 μm packing and then with Whatman 30-38 μm pellicular C-18 guard column packing (Whatman SA, France) . For

4-(n-octyl)nitrobenzene, the solvent was 70:15:15 acetonitrile:tetrahydrofuran:water. The elution time was 4.3 minutes behind the solvent front (flow rate 1.5 ml/min) . For all other compounds, the solvent was 70:30 acetonitrile:water. Nitrobenzene, 4-ethylnitrobenzene, and 4-(n-butyl)nitrobenzene eluted 2.3, 3.0 and 5.5 minutes behind the solvent front, respectively. In each case, the corresponding aniline eluted slightly earlier than its nitro analog. The wavelength used was 280 nm. Kinetic Experiments

The pH was determined before each experiment using an Orion 91.62 probe (Orion Research, Inc, Boston, MA) by gently agitating the probe in the solid phase at the bottom of the jar. In all four sediment samples, the pH was maintained between 6.8 and 7.1. Sediment Eh readings were also taken

before each experiment by placing a Markson 1202 combination platinum electrode (Markson, Houston, TX) in the sediment jar overnight to equilibrate before taking readings. The Eh generally read between -170 and -230 mv.

Aliquots (5 ml) of the desired sediment, drawn while stirring the sediment with a glass rod in the glove box, were transferred to a series of 15 ml screw cap test tubes. The tubes were capped with Hungate (open top) screw caps (fitted with teflon lined septa) and brought outside of the glove box. The tubes were spiked with 5 μl of a tetrahydrofuran standard (10 ² M) of the desired nitroaromatic compound. The tubes were briefly vortexed. At selected time intervals, one tube from the series was sacrificed by quenching with acetonitrile (2 ml) and vortexing for 30 seconds. The remaining sample tubes were inverted four times at this point. For nitrobenzene, 4-ethylnitrobenzene, and 4-(n-butyl)nitrobenzene, the tubes were then centrifuged (tabletop centrifuge, 2500 rpm for 20 minutes) and the supernatant (5 ml) was filtered through a 925 mm Millipore™ filter (Millipore Corp., Bedford, MA). The sample was then ready for analysis. For 4-(n- octyl)nitrobenzene, hexane (4 ml) was added to the quenched sample, and it was vortexed again for 1 minute. An aliquot (3 ml) of the hexane layer was removed, evaporated under a nitrogen stream, and redissolved into acetonitrile (5 ml) . The sample was then ready for analysis.

For runs with added chemical sterilant or inhibitor, the appropriate amount of chemical was added to the sediment 16 hours before spiking with nitrobenzene. For kinetic runs without the sand and silt fractions, the sediment was allowed to settle for 1.5 hours under nitrogen after stirring

before aliquots were drawn. For kinetic runs with sediment associated water only, sediment was centrifuged as above, and the resulting supernatant was used for kinetic runs. Recoveries of the nitroaromatic compounds 4 hours after spiking from twice-autoclaved sediment (220°C, 20 psi x 20 m) were 84 ± 3% for nitrobenzene, 4-ethylnitrobenzene, and 4-(n-butyl)nitrobenzene, and 100 ± 5% for 4-(n- octy1)nitrobenzene. Sediment/Water Ratios

Triplicate aliquots (5 ml) of sediment were transferred to tared test tubes. After reweighing, the samples were centrifuged as above, the supernatant removed, and the pellet dried in an oven overnight at 95°C. After cooling, the dried samples were weighed again. Ratios were calculated by dividing the dry weight of the sediment by the weight of the aqueous phase. Distribution coefficients Distribution coefficients (K _d ) for the four nitroaromatics were determined at 3 minutes, 1 hour, 3 hours, and 4 hours after spiking in twice autoclaved sediment (as above) . Samples were periodically inverted. For nitrobenzene and 4-ethylnitrobenzene, the spiked sample was centrifuged (as above) after the desired incubation time, and supernatant (3.5 ml) was removed and filtered as described above. To the remaining supernatant and pellet, acetonitrile (2 ml) was added, and the sample was vortexed for 30 seconds. The sample was then centrifuged again, and the supernatant (2-2.5 ml) was filtered. After correcting for the supernatant contribution to the pellet extract, the K _d (6.25 μg compound/g dry solid)/ (μg compound/g supernatant) was calculated using the sediment/water ratios calculated for the experiment. Recoveries were quantitative for both

compounds. For 4-(n-octyl)nitrobenzene, sediment diluted 10-fold with supernatant from another sample jar (obtained by centrifugation) was used. After the desired incubation time, the sample was centrifuged, and 4 ml of supernatant was removed and centrifuged again. A portion of this second supernatant (3 ml) was removed and extracted with hexane (2 ml) . An aliquot of the hexane layer (1.5 ml) was then evaporated under a nitrogen stream, and the residue redissolved in acetonitrile (1 ml) before analysis. To the pellet and remaining supernatant, acetonitrile (2 ml) and hexane (4 ml) were successively added, vortexing for 30 seconds and 1 minute afterwards, respectively. An aliquot of the hexane layer (3 ml) was then evaporated and the residue was redissolved in acetonitrile (5 ml) for analysis. The K _d s were calculated after correcting for the sample handling steps. The recovery was 86 ± 4% for 4-(n-octyl)nitrobenzene. Diluted sediment was also used to determine the K _d for (n-butyl)nitrobenzene. The K _d was determined in a similar manner as for 4-(n-octyl)nitrobenzene, except that acetonitrile alone (4 ml) was sufficient to extract the pellet, and after a second centrifugation, an aliquot of the extract (3 ml) was filtered and analyzed directly. Recoveries were only 60 ± 6% for 4-(n-butyl)nitrobenzene, but this method was found to give the best recoveries of the extraction methods attempted. K _d were found to remain nearly constant with time for all of the above compounds.

Aerobic and Anaerobic Sterility Tests Aerobic sterility was determined by applying 0.1 ml of treated sediment to sterilized, hydrated tryptone glucose extract agar spread on a Petri dish and watching for growth within 5 days. Anaerobic sterility was determined by stabbing

treated sediment into 5 ml plugs of sterilized thioglycollate indicator agar under nitrogen in 15 ml test tubes. Sterility was indicated by the retention of the pink oxygenated band at the top of the plug, the absence of gas formation, and no visible growth in the anaerobic portion of the agar over a five day period. Polarographv The reduction potential of the four nitroaromatics was measured in aqueous solution (10 mM) at pH 6.8 (0.1 M phosphate) using square wave polarography. The instrument was an EG & F Princeton Applied Research Corp. Digital Polarograph 82 (EG & F Instruments, Ltd., United Kingdom) used in the HMDE mode, with a medium drop size. The purge time was 4 minutes (helium) . The scan range was -100 to -600 mV (relative to Ag/AgCl) ; pulse size, -20 mV; step size, -5 mV; drop settle delay, 400 ms; sweep delay, 0 m; preintegration time, 2000 μs; integration time 12,562 s; and current gain 256. Results

As shown in Table 5, nitrobenzene was readily degraded in all samples of nitroaromatic anaerobic sediment gathered from the four different water bodies. The half-lives for the reduction of nitrobenzenes in anaerobic sediment samples are on the order of a few hours as shown in Table 6. A graph showing the reduction of nitrobenzene to aniline over a two hour period of time is shown in Figure 9. Nitrobenzene in sediment which had been autoclaved exhibited a reduced rate of reduction to aniline. Chemical sterilization with formaldehyde had no effect on the rate of reduction to aniline. Therefore, the reducing component is heat labile but not labile to formaldehyde when bound to sediment.

Table 5. Reduction of nitrobenzene in four anaerobic sediments

Sediment %0M" EEhh(fmmvv)) Ep KKa„ tt _w ^(fmmiinn )) rr ² _

Beaver Dam 5.6 ±0.3 -170 0.112 ± 0.016 3.9 ± 0.8 142 ± 51 0.980

Bar-H 2.2 ±1.0 -170 0.060 ± 0.006 2.6 ± 0.5 21 0.999

Hickory 1.8 ±1.0 -230 0.052 ±0.002 2.6 ±0.4 142 0.999 Hill

Memorial 4.3 ± 0.2 -250 0.038 ±0.003 9.0 ±1.0 120 0.991 Park

"Determined by heating dry sediment at 425°C for 20 hours, corrected for water of adhesion (obtained by heating dry sediment at 70°C for 3 days)

Table 6. Pseudo-first-order disappearance rate constants for reduction of nitrobenzene and three 4-substituted nitrobenzenes

Compound t _1/3 (min) nitrobenzene 53

4-ethylnitrobenzene 74

4-(n-butyl)nitrobenzene 120

4-(n-octyl)nitrobenzene 1140

Example 3. Kinetics of Reduction of 15 Halogenated Hydrocarbons in Four Anoxic Sediment Samples Sediment-water slurries were collected from the ponds known as Vechten Pond, Bilthoven; Breukelveen, and Loosdrechtse Plassen and the slow-moving stream known as Dommel, all in The Netherlands. Samples were collected by scooping the first 5-10 cm of bottom sediment into glass jars. The jars were completely filled with sediment and water and capped under the water surface. Sediment samples were stored at 20°C until used for experiments. Prior to use, the samples were sieved through a 1 mm wire sieve to remove debris. The sediment to water ratio (g/g) of samples was determined by placing 10 ml aliquots of the thoroughly mixed sediment-water sample in weighed, open 50 ml jars. Consequently the water was evaporated during one day at 80°C and the jars were reweighed. Each determination was repeated five times.

Kinetics experiments were performed using a batch method in which sediment aliquots were distributed into a series of test tubes and spiked with a known concentration of a halogenated hydrocarbon under a nitrogen atmosphere. All halogenated hydrocarbons were purchased from Aldrich Chemical Co., Milwaukee, WI. A tube was sacrificed for analysis of the concentration of chemical in the sample at specific time intervals during incubation. For each experiment, 10 ml aliquots of sediment were placed in 20 ml test tubes. A stock solution of each halogenated compound was made in methanol (Burdick and Jackson, Musegon, MI) such that a 10 μl addition of chemical into 10 ml sediment gave the desired initial experimental concentration. Sample tubes were spiked with 10 μl halogenated hydrocarbon. After

vortexing, the tubes were incubated at 22°C and were periodically mixed. At specific time intervals, 2 ml of acetonitrile (Burdick and Jackson) was added to the tubes to quench the reaction. Tubes were vortexed for 1 minute, the sediments extracted with 4 ml cyclohexane (Burdick and Jackson) and vortexed again for 2 minutes. The cyclohexane layer was recovered from the tubes after centrifugation at 3500-4000 rpm for fifteen minutes, placed in a clean tube and stored at -30°C.

Cyclohexane extracts were analyzed using a Carlo Erba (Milan, Italy) 4160 gas chromatograph equipped with an electron-capture detector and a Hewlett Packard (Avondale, PA) 3392 integrator. The column used was a fused silica open tubular column with CP-sil-5CB as the stationary phase (25 m x 0.32 mm) . The relative concentrations of the compounds were calculated by comparing peak areas of samples at given times against the peak area of the zero time sample.

A Metrohm (Herisau, Switzerland) pH meter was used for both the pH and Eh measurements. Eh values were measured using a platinum Ag:AgCl reference electrode. All Eh values are reported versus SHE. All pH and Eh measurements were performed under a nitrogen atmosphere.

GC-MS analyses were performed using a Hewlett Packard model 5890A gas chromatograph interfaced with a Finnigan (San Jose, CA) 4500 quadrupole mass spectrometer. The column was a fused silica capillary column with CP Sil-5 as the stationary phase.

The NMR spectra were recorded at 200 MHz on a Bruker (Karlsruhe, Germany) AC 200 NMR spectrometer interfaced with an ASPECT 3000 computer. The spectra were recorded using CDC1 ₃ (deuterated

chloroform) as the solvent with TMS (tetra-methyl silane) as the internal standard.

The results of the kinetics studies for 15 halogenated hydrocarbons are shown in Table 7. The Eh and pH did not change appreciably during the experiment. Below an Eh value of -50 mV, the disappearance rate constant was essentially independent of the Eh. The rate of disappearance of all the compounds that reacted was first-order through at least two half-lives. The halogenated hydrocarbons exhibited a wide range of reactivity with the shortest half-life of about ten minutes for tetraiodoethene to no detectable reaction after 90 days of incubation for perflurodecaline, perfluorohexane and DDT. A calculated maximum rate constant and minimum half-life can be arrived at by assuming a 10% experimental error in the analysis of the compound after 90 days reaction time. All of the halogenated hydrocarbons were stable to hydrolysis under the redox reactions employed in this example.

Table 7. Kinetic data and reaction parameters for reduction of halogenated hydrocarbons in sediment samples under anaerobic conditions

Initial Rate Half

Comoound Cone. Constant life rmol/lϊ k (min-1) (hour)

Tetraiodoethene ¹ 9.60E-06 9.22E-02 0.1

Hexachloroethane ² 7.32E-08 2.47E-02 0.5

Hexachloroethane ³ 7.32E-08 7.62E-03 1.5

1,2-Dibromo-1,2- dichloroethane ^1* 1.21E-06 5.44E-03 2.1

1,2-Dibromo-l,2- dichloroethane ^1* 1.21E-06 4.54E-03 2.5

Hexachloroethane ¹ 7.32E-08 2.33E-03 5.0

Hexachloroethane ⁴ 7.32E-08 1.22E-03 9.5

Carbontetrachloride ^{: 2} 1.6E-06 1.16E-03 10.0

2,3-Dibromobutane ^1* 1.19E-06 1.10E-03 10.5

2,3-Dibromobutane ^1* 1.19E-06 9.15E-04 12.6

1,2-Dibromoethane ¹ 9.28E-07 7.13E-04 16.2

Hexachloro- cyclohexane ¹ 5.70E-07 1.51E-04 76.5

Tetrachloroethene ¹ 9.79E-08 4.05E-05 285.5

Iodobenzene ¹ 2.2E-06 2.55E-05 453.0

Hexachloro- cyclohexane ¹ 5.09E-07 2.36E-05 489.4

DDT ¹ 1.44E-06 NR ~

Perfluorodeca1ine ¹ 1.44E-06 NR —

Perfluorohexane ² 2.01E-06 NR -—.

¹ Sediment obtained from Vechten Pond: pH 7.2, Eh -139 mV, Organic Carbon 6.0%.

² Sediment obtained from Breukelveen: pH 7.6, Eh -145 mV, Organic Carbon 29%.

³ Sediment obtained from Dommel: pH 7.5, Eh -155 mV, Organic Carbon 0.53%

⁴ Sediment obtained from Loosdrechtse Plassen: pH 7.7, Eh -128 mV, Organic Carbon 32.6%.

NR = no reactivity observed during incubation; * = Diastereomer

Example 4. In vitro Remediation of Hexachloroethane in 18 sediment or Aquifer Samples

The sorption-corrected rate constants for the reduction of the halogenated hydrocarbon hexachloroethane (Aldrich Chemical Co., Milwaukee, WI) by remediation proteins present in 18 different sediment, soil and aquifer samples were correlated with organic carbon content.

Rate constants for reduction of hexachloroethane were determined generally as described in Example 3.

Organic carbon content was determined by the method of Lee and Macalady, J-nter. J. Environ . Anal. Chem . 35:219-225 (1988).

Correlation of Rate Constants with Organic Carbon Content

Disappearance rate constants for the reduction of hexachloroethane by remediation proteins in 18 different sediment, soil and aquifer samples are listed in Table 8, in the order of decreasing rate constants (k^,) . The data in Table 8 suggest a relationship between the rate constants and organic carbon content of the samples.

Table 8. Measured (k _obβ ) and sorption corrected (k _∞rr ) disappearance rate constants for hexachloroethane in selected sediments and aquifer samples.

Sediment

SEDIMENT Fraction cone. log k _ob ,* log k _corr *

SOURCE OC (g.g- ¹ ) min ^"1 min ^"1

BarH 0.022 0.08 -1.301 -1.081

BarH 0.022 0.09 -1.455 -0.248

HickoryH 0.018 0.11 -1.585 -0.519

Breukelveen 0.29 0.045 -1.607 0.407

BeaverD 0.056 0.2 -1.699 -0.570

MemorP 0.043 0.055 -1.721 -0.841

BarH 0.022 0.075 -1.721 -0.586

Loosdr. 0.33 0.050 -2.118 -0.018

Plassen

Vechten 0.06 0.087 -2.632 -1.019

Pond

EPA-B1 0.009 0.1 -2.745 -1.847

Dommel 0.0053 0.469 -2.915 -1.613

EPA-13 0.03 0.1 -3.089 -1.708

EPA-11 0.015 0.1 -3.104 -2.007

EPA-6 0.0072 0.1 -3.350 -2.535

Lula, aq 0.000065 0.161 -4.267 -4.233

Blythville, 0.00012 0.142 -4.359 -4.306 aq

Blythville, 0.00012 0.613 -4.361 -4.167 aq

Lula, aq 0.000065 0.689 -4.521 -4.393

* observed rate constants

# sorption corrected rate constants: k _& - kob» (1+ P * k _d ) aq = aquifer

Sorption Corrected Rate Constants

The sorption corrected rate constant (k^) in which k _ob , is corrected for the fraction of the compound sorbed is given by the following equation, in which P denotes the sediment concentration (g.g ¹ ):

^• "•obs (1 + P * K _d ) (4)

The k _COTr values are included in Table 8 and were calculated using a log K _^ , (octanol-water partition coefficient) value of 4.61. Linear regression

analyses showed that this correction for sorption increases the correlation between organic carbon content and the disappearance rate constant.

Although there are other factors that contribute to the reduction of hexachlorethane, organic carbon content accounts for about 91 % of the variance of the data.

Example 5. Nitroreduction of 2,4,6- Trinitrotoluene (TNT) with Crude Enzyme Extract

The effects of initial TNT concentration on the reaction order in anaerobic sediment samples, the reaction of TNT with crude proteins extracted from high organic carbon content sediments, and the effect of iron powder on TNT degradation in flooded soil were analyzed.

To monitor the disappearance of TNT and formation of its products in anaerobic sediment samples, TNT and its reduced products were extracted from water/sediment samples with acetonitrile (Burdick and Jackson, Musegon, MI) and analyzed by HPLC equipped with a UV detector as follows. The solvent was 30% water (pH 10) and 70% acetonitrile at a flow rate of 0.7 ml/min.

Absorbance was detected at a wavelength of 238 nm. A representative HPLC chromatogram for TNT and its reduced products is shown in Figure 12. The dinitro and monoamino toluene isomers (Peak 2) eluted at same retention time, however, these isomers can be separated and identified by GC/MS. The two diamino, mononitrotoluene isomers (Peak 3) also co-eluted. Triaminotoluene (Peak 4) eluted with the solvent front. TNT (Peak 1) and standards for identification of its reduced products were obtained from the U.S. Army Research, Development, and Engineering Center, Picatinny Arsenal, NJ.

The dependence of initial concentration of TNT on reaction order was investigated. The initial

concentrations of TNT used were 0.25 ppm, 2.5 ppm and 125 ppm. The saturated concentration of TNT is about 125 ppm in water.

Sediments were collected from a stagnant, slow moving stream near Athens, Georgia. These sediments were passed through a 1 mm sieve. Aliquots of 5 ml of sediment slurries were transferred to a series of 15-ml screw cap test tubes and kept under anaerobic condition. The sediment/water ratio was 0.12, pH was 6.5 and Eh value was -368 mv (vs. Ag/AgCl) . A certain amount of concentrated TNT solutions were spiked into sediment samples to achieve three concentrations of TNT. At given time intervals, one tube of sample from the series was extracted by adding 1 ml of acetonitrile, vortex-mixing for 1 min and centrifuging. The supernatant was transfer centrifuged again and analyzed. Figure 13 shows percentage of TNT remaining in the water/sediment sample as a function of time for three different concentrations of TNT. The rate of change of TNT concentration is a constant when the concentration of TNT is 125 ppm or greater. Beyond this concentration, the reaction proceeds at a rate independent of both concentrations of TNT and sediment. Therefore, at high TNT concentration, the reaction rate is zero order. When the concentration of TNT was decreased to 2.5 ppm, the concentration of TNT decreased exponentially with time. Plotting ln(c/c0) against time by using least-square regression method, a straight line was obtained with the value of r ² equal to 0.99. This reaction of TNT is a first order reaction and half- life is 90 min. With 0.25 ppm TNT, the reduction rate was fast. After a certain elapsed time, the concentration change slowed down. A plot of 1/c as a function of time and regression analysis showed

that the TNT reduction in water/sediment system appeared to follow a second order reaction. This result assumed that both TNT and sediment activity have the same initial stoichiometric concentration. Consequently, the results indicated that the reaction order of TNT in the water/sediment system depends on TNT initial concentration.

The reducing activity of a protein extract isolated from a high organic carbon sediment was analyzed after introduction into either an aqueous solution or a low organic carbon content system such as aquifer material. One ppm TNT solution was mixed with 0.5 g/ml aquifer sample as the control. One ppm TNT solution was mixed with 15 μg/ml of the TNT-reducing protein extract isolated from fraction 18 of Figure 2 as described above and incubated for 24 hours. The purpose of incubation was to let the substrate, TNT, associate with the denatured subunits to recover the original active form of the enzyme. This procedure was followed by the

"Microbial Metabolism of Aromatic Nitriles" study described by David B. Harper, Biochem . J. 167:685- 692 (1977) . After the incubation period, the incubated sample was spiked with concentrated TNT solution so that the final TNT concentration was 1 ppm.

To determine whether the protein extract would bind to aquifer material and retain activity in water, 15 μg/ml of protein extract was incubated with 0.5 g/ml aquifer sample for 24 hours before the TNT solution was added. The aquifer material had been collected from Columbus Air Force Base in Georgia. The organic carbon content in this aquifer was 0.027%. Figure 14 shows the kinetics of TNT degradation in the three systems described above. When Ln(c/c0) was plotted as a function of time, the

regression analysis showed that the reaction of TNT follows first-order kinetics in all three systems. A comparison of these three results indicates that the protein extract bound to sediment enhances the rate of disappearance of TNT. Half-life of TNT in the combined aquifer material and protein extract system is smaller than that in protein extract system alone.

The effects of the addition of iron to the system was analyzed. 1.6 ppm TNT in distilled water was combined with 1% (w/v) iron powder under anaerobic conditions. As shown in Figure 15, a plot of ln(c/c0) as a function of time using the regression method gave a straight line indicating that the TNT reduction is a first order process with a half-life of 1.14 days. The TNT reduction rate was increased by shaking the reactor to enhance the mass transfer. It should be noted that TNT reduced products were not observed in this study, most likely because the reaction rate of TNT reduced products on the metal surface was extremely fast.

A bench scale experiment was designed to investigate the redox degradation of TNT in contaminated soils. TNT contaminated soils were sampled from Alabama Army Ammunition Plant near Childersburg, Alabama. 100 g of contaminated soil was flooded by water to a volume of 900 ml in a quart jar. At given time intervals, aliquots of 1- 1.5 ml were taken from the reaction mixture and centrifuged. The supernatant was decanted and the sediment was extracted with acetonitrile. Both supernatant and sediment extracts were analyzed. Figure 16 shows TNT concentration remaining in supernatant and solid phase as a function of time. The initial concentration of TNT in contaminated soil was 6000 ppm based on dry soil. In this

study, the concentration of TNT remaining in the solid phase based on wet soil was determined. The concentration of TNT in both aqueous and solid phases were almost constant during 48 day period. Also, Eh measured in the glove box was very positive. At day 48, 9 grams of iron powder was added to the reaction system. Although the concentration of TNT in solid phase did not change significantly, both the concentration of TNT in supernatant and the Eh decreased with time. When the reaction jar was inverted and shaken vigorously, both TNT in the supernatant and solid phases decreased dramatically. Therefore mass transfer enhances the degradation of TNT.

Example 6. Identification of Nitroreductase, Nitrilase, and Dehalogenase from Plant Sources

This series of experiments was conducted to locate sources of nitroreductase, nitrilase and dehalogenase from sediments and soils. The purified enzymes were used to obtain the monoclonal antibodies. Field immunoassays were developed using enzyme linked immunosorbant assays (ELISA) . Using the field immunoassays, plant sources of the proteins were located around ponds and streams. Enzymes were then isolated from the plants and their activities compared with that of the sediment or soil activities and whole plant activities. The general approach described in this experiment, using immunospecific assays, can be used to identify, assess the distribution of, and identify the origin of additional environmental redox components. Assays

TNT was used as the substrate in an activity assay for the isolation of the nitroreductase. The normal procedure was to add 20 μl of an enzyme to

be analyzed to one milliliter of substrate, incubate, then analyze. If an inhibitor was being tested, it was added first to the enzyme alone. If temperature effects were being tested, the enzyme mixture was pre-incubated at that temperature for 30 minutes prior to the addition of substrate.

Because little, if any, reducing activity for nitroaromatic compounds could be found in the aqueous phase of sediment samples, the solubilization of the components was approached as if the components were hydrophobic or highly charged. Isolation

A series of extracting solutions including high salt, urea, detergents, and glycerol were evaluated for efficiency in extracting nitroreductase activity from the solids. All of these reagents could solubilize some activity from reactive sediment. However, the most reproducible reagent used to extract the sediment was glycerol, alone or in combination with buffers and high salt concentration. The glycerol extraction reagent also conveyed the most stability to the activity. The glycerol extract was then concentrated by ammonium sulfate precipitation.

The procedure for isolating and characterizing the reducing enzymes involved extraction, followed by precipitation, chromatographic separation by ion exchange chromatography, hydrophobic interactions, sizing, and zinc affinity. With the first extraction, the activity was fractionated on a sizing column that included components of 5000 daltons or less. The nitrate reduction activity eluted in the void volume, implying that the active component had a molecular weight greater than 5000. The component was further purified by several column chromatography steps. The activity was

followed through a series of columns that separate according to different characteristics: charge, hydrophobicity, size and affinity for metals. At the end of these isolation procedures, the component was checked by either gel electrophoresis or capillary electrophoresis to determine purity. A single band under native conditions was the criteria for purity.

To determine if the component was a protein, a portion of the resulting fractions was treated with one to four units of three general proteases. A loss of reductase activity with increasing concentrations of added protease was observed, indicating that the active component was a protein. A more detailed description of the isolation and characterization of a dehalogenase is set forth below in Example 7. Characterization

The component of the sediment that catalyzed the reduction of TNT has a native size of 316 kD, subunits of 66, 37 and 19 kD, contains four sulfur and 4 iron molecules (initial experiments indicated the presence of only 2 iron molecules per protein, however, the ESR signal did not show a signal for a two iron:four sulfur complex, therefore, most likely four iron molecules are present) , a temperature optimum of 20 to 37°C, a pH optimum of 6.5 to 8.5, general protease sensitivity, but no sensitivity to trypsin or chymotrypsin, is inhibited by zinc, lead and low pH, uses NADPH and cannot be replaced by NADH, and the reaction rate is faster when bound to solids.

Figure 22 shows the kinetics of degradation of TNT as analyzed by the Lineweaver-Burke procedure (Wong, J. T-F, KINETICS OF ENZYME MECHANISMS Academic Press, New York, NY pp. 1-37, 1975). This procedure uses initial velocities and is used to

estimate the Michaelis constant, K,,,. This is the concentration of substrate which gives half-maximal velocity. The other derived constant is a maximal rate, V^. Also shown in Figure 22 is the effect of adding NADPH to the enzyme reaction. NADH does not replace NADPH. For the enzyme alone, V equals 1.42 x 10" ⁷ moles TNT/min/μg enzyme. For the enzyme plus NADPH, V equals 2.5 x 10" ⁷ moles TNT/min/μg enzyme. The effect of initial concentration of TNT on incubation in sediment, or with protein extract was investigated. In sediment, zero-order reaction kinetics were observed at 1.25 x 10 ^"3 M TNT, but first-order kinetics were found between 1 x 10 ^"5 and 1 x 10" ⁶ M. With the purified enzyme, the range was much smaller. The reaction occurs very slowly at 7.5 x 10" ⁶ M. However, the half-life of a 1.5 x 10"* M solution is 20 minutes. Monoclonal Antibodies Using microgram quantities of the purified protein (antigen) , antibodies were produced in mice and the mouse speen cells fused with myeloma cells to produce clones that contained only one kind of antibody. Eight different monoclonal antibodies against the nitrate reductase isolated from sediment were obtained using known procedures described by Dunbar and Skinner, "Preparation of Monoclonal Antibodies" in METHODS IN ENZYMO OGY, vol. 182, Guide to Protein Purification, ed. by M.P. Deutscher, Academic Press, Inc., San Diego, CA, pp. 663-670 (1990) . Five different monoclonal antibodies were produced to the nitrilase and seven to the dehalogenase. A polyclonal antibody, obtained using the known procedures described by Dunbar and Schwoebel, "Preparation of Polyclonal

Antibodies" in METHODS IN ENZYMOLOGY, vol. 182, Guide to Protein Purification, ed. by M.P. Deutscher,

Academic Press, Inc., San Diego, CA, pp. 663-670 (1990) , and subsequently the monoclonal antibodies, were used to identify potential sources of the enzyme. The polyclonal antibody results are shown below in Table 9.

Table 9. Concentration of Nitroreductase in Aquatic Plants Detected by Polyclonal Antibody

Plant ug/gram plant

Stonewort 23

Soft Rush Roots 13.5

Soft Rush Leaves 10.0

Waterwort Leaves 22

Sheep Sorrel 2

Oak Leaves 2

Immunoassav

The immunoassays in these experiments utilized the binding specificity of the mouse-produced antibody (primary antibody) for its specific antigen (isolated sediment protein) . The antigen was attached to a solid phase, and the antibody selectively allowed to bind to the primary antibody. A labelled antibody (secondary) was then added that binds to the primary antibody in accordance with the method of Harlow and Lane, ANTIBODIES: A LABORATORY MANUAL. Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory, NY pp. 139-245 (1988) . For these experiments, the label was an enzyme with a high turnover number such as horseradish peroxidase or alkaline phosphatase. These enzymes were attached to a secondary antibody and the enzyme mediated production of nitrophenol or a derivative of tetramethyl benzidine that was detected spectrophotometrically. In these experiments, the protein of interest could be detected at the nanogram level.

With this enzyme linked immunosorbant assay, it is possible to search for environmental sources of

the enzyme. Field samples of aquatic weeds and plants around ponds and streams where nitroreductase had been observed in sediments and soils were collected. These possible sources of the enzymes were extracted with 20% glycerol in pH 7 phosphate buffer. The plants were weighed out, frozen in liquid nitrogen in a mortar, ground to a dry powder, then 2 ml of glycerol mixture was added per gram of plant. The mixture was continued to be ground until there were no obvious pieces. The mixture was scraped into microcentrifuge tubes, and centrifuged. Sediments of interest were weighed, and extracted by inversion, then centrifuged as above. A portion of each supernatant was bound to Immunlon™ 4 ELISA plates along with purified standard proteins using pH 9.5, 100 mM borate. The quantity of reaction of each source was compared to the standard curve of the protein of interest. The primary antibody was a mouse polyclonal or monoclonal developed against either the sediment nitrate reductase, nitrilase or dehalogenase. The second antibody was either goat anti-mouse conjugated with alkaline phosphatase or horse radish peroxidase. The presence of the nitrate reductase was more reproducible with the use of horse radish peroxidase as the indicator protein. Field Assay

When preliminary data indicated that aquatic plants had a high concentration of a nitrate reductase that would cross-react with the isolated monoclonal antibodies generated against the sediment enzyme, all plants which had roots or stems in water were tested for activity using the methods of Oaks and Hirel, "Nitrogen metabolism in roots", Ann . Rev. Plant Physiol . 36:345-365 (1985). In preliminary results using western blots,

proteins isolated which showed a response to the antibodies were slightly different from sediment proteins, which may be due to the presence of proteases or the sediment enzymes are processed after binding to particles. This would give a different molecular weight to sediment proteins than to plant proteins. Nitroreductase Results

Kinetic data using whole plants as sources of the enzyme were compared with those for the sediment. The estimated range of K,,, and V,, _^ are 5 x 10" ⁵ to 7 x 10" ⁷ moles and 1 x 10" ⁷ and 3.3 x 10" ⁶ moles/min/gm, respectively. Above the K,,, concentrations, the degradation of TNT is zero- order. However, the reactions do occur. The reducing activities of the enzymes of plants collected in the natural environment are highly regulated. (Jonri and Hendriques, "Nitrate reductase from the moss Funaria. Proceedings of the Phytochemicla Society of Europe-29." BRYOPHYTES: THEIR CHEMISTRY AND CHEMICAL TAXONOMY, eds. H.D. Zinsmeister and R. Mues, Oxford Science Publications, Clarendon Press, Oxford, pp. 265-273 (1990) . The regulation of this function is plant specific. Some plants require light and nitrate, some only require nitrate to activate the genes for TNT reduction. Nitrilase Results

The isolation of the nitrilase from sediment was similar to that of the nitrate reductase above. The enzyme has the following characteristics: a native size of 660 kD; a subunit size of 150 kD; no detectable iron, zinc, molybdenum or copper molecules, optimum temperature of 15°C to 45°C, sensitivity to trypsin, chymotrypsin, and general proteases; and no cofactors were required. As shown in Figure 23, nitrilase activity can be separated from both dehalogenase and nitrate

reductase activity on a single column using metal affinity as the basis for separation and measuring nitrilase activity by the ability to reduce benzonitrile. The protein used for the production of monoclonal antibodies was passed through four chromatography columns and exhibited a single electrophoretic band on a native polyacrylamide gel.

The use of the polyclonal antibody of the nitrilase against extracts of plants showed more selectivity than the antibodies against nitrate reductase. There was no cross-reactivity between nitrate reductase and nitrilase antibodies. As shown below in Table 10, nitrilase activity is not found in all plants or sediments. The nitrilase activity is most likely due to the production of indole-3 acetic acid from indole-3-acetonitrile. Only the hornwort (stonewort) or oak leaves have substantial activity.

Table 10. Concentration of Nitrilase in Aquatic Plants Detected by Polyclonal Antibody

Plant ug/gram plant

Stonewort 16

Soft Rush Roots 5

Soft Rush Leaves 3

Waterwort Leaves 4

Sheep Sorrel 3.2

Oak Leaves 24

Dehalogenase Results

The isolation procedure for the dehalogenase was the same as for the nitrate reductase except that, because the dehalogenase activity is oxygen sensitive, all buffers, columns, collection tubes and transfers were kept under an argon atmosphere. Buffers were autoclaved, and while hot, were gassed with argon. Column fractions were kept in 15 ml tubes sealed with butyl rubber stoppers (covered

with aluminum foil) under argon or 150 ml milk bottles sealed with screw caps having Teflon™ liners.

The original substrate used during isolation was perbromoethene. Volatile substrates were made fresh every day from saturated stocks. Autoclaved water in sealed serum bottles were gassed with argon, then the saturated halogenated hydrocarbon was added through the seal with a glass syringe. Samples were analyzed for enzyme activity and controls were dispensed from the same bottle at the same time. Due to the volatile nature of the substrates, most often six samples and six controls were used at every time point. The dehalogenase has the following physical characteristics: a native size of 293 kD; subunits of 90, 39, and 28 kD; one copper molecule per mole and as many as 48 iron molecules per mole; a temperature optimum of 10°C to 50°C (HCA) and 25°C to 50°C (PBE) ; ability to degrade both alkenes and alkanes in mixtures, preferably alkenes; ascorbic acid is a cofactor; and sensitivity to oxygen.

This enzyme is not the only dehalogenase present in sediment. During the isolation procedure using the hydrophobic interaction column, several peaks of activity were observed. The activity lacking nitrate reducing activity was chosen for this set of experiments to produce an antibody that would have minimal cross reactivity with the nitrate reductase antibody. As described above, the enzyme will dehalogenate both alkenes and alkanes and appears to survive the reaction; more substrate can subsequently be added to form more product.

The Lineweaver-Burke plots of the rates versus substrate concentrations for PBE and HCA are given in Figure 24a and 24b. For PBE, the K _m is 10 μM

and the V,,^ is 2 x 10' ⁸ moles/min/μg. For HCA, the K _m is 125 μM and the V,^ is 4.1 x 10" ⁸ moles/min/μg.

The enzyme contains both copper and iron. One copper molecule and 48 iron molecules per mole enzyme is predicted. An enzyme system containing both metals in a model reducing system using copper nitrate and ascorbic acid exhibits reduction of halogenated hydrocarbons by two electrons at a time as shown in Figure 25. The polyclonal antibody produced by injection of the dehalogenase into mice is specific. Seven monoclonal antibodies have been produced and fail to exhibit cross-reactivity with either nitrilase or nitrate reductase. Conclusions

The nitroreductase, nitrilase and dehalogenase enzymes are large, ulti-subunit reducing proteins, stable for long periods of time in sediments and possibly more stable and more active when bound to a solid surface, such as the solid surfaces in sediment that contain metals. Except for the nitrilase, the enzymes have a metal coordination center necessary for activity. The nitroreductase originally isolated appeared to include two iron molecules per mole of protein. However, it contained four sulfurs and exhibited a preliminary ESR spectra that would not predict a two-iron center. Therefore, the nitroreductase most likely has a four-iron center. Iron must be important in the active site of the enzyme because it is inhibited by zinc, copper and lead and can be removed by procedures such as dialysis. The data indicates that the nitroreductase is of plant origin. The nitrilase is a very large enzyme, most likely also of plant origin. It does not contain any metals in the active site and is sensitive to

protease, including trypsin and chymotrypsin. It must bind to pores in the sediment, or across a surface. It is faster in solution rather than bound to sediment. The dehalogenase contains at least copper in the active site and requires acid conditions for activity. The oxygen sensitivity is most likely related to the presence of the copper molecule. The dehalogenase is most likely also of plant origin.

Example 7. Isolation and Partial Characterization of a Dehalogenase from Sediment

This experiment describes, in detail, the isolation and partial characterization of a dehalogenase from sediment. Extraction

Sediment samples having a high organic content were collected anaerobically, then a 1:1 ratio of argon-saturated water containing 20% glycerol and 0.5 M KCl was used to extract the sediment sample. The slurry was mixed in argon-gassed fruit jars, the supernatant poured into pregassed 250 ml centrifuge bottles, and centrifuged at 7000 rpm for 20 minutes in a Sorval™ GSA rotor at 6°C. The supernatant was decanted into clean 250 ml centrifuge bottles with gassing, capped and recentrifuged to remove any sediment particles. A clear yellow supernatant was formed. Sufficient ammonium sulfate was added to make the solution 50% saturated. The mixture was stirred gently in a refrigerator at least one hour prior to centrifugation. The supernatant was saved, made up to 85% ammonium sulfate while stirring in a reduced atmosphere, and incubated in a refrigerator overnight. The precipitate resulting from the centrifugation of the solution contained the dehalogenase activity.

All subsequent buffers, columns and analysis were performed anaerobically, usually with boiled, argon-gassed solutions and argon-gassing of all columns and fractions resulting from the fractionation. Analysis

Portions of the extracted samples were added to 1.8 ml autosampler vials containing the halogenated hydrocarbons to be analyzed. Approximately 20 to 50 μl of an enzyme extract was added in a reducing atmosphere to 1 to 3 ppm of the solvent. The solvents were diluted from saturated stock solutions into boiled, argon-gassed water. Six samples were generally made for each time point. Each sample containing an enzyme was compared to its control made at the same time from the same stock.

Short chain Halogenated hydrocarbons were analyzed by gas chromatography. A Hewlett Packard 5890 II™ equipped with an ECD detector and an autosampler was used to inject 0.5 μl on a 15 m, 0.53 μm i.d. column coated with DB-1 (3 μm) . Both temperature and pressure were programmed. Hexachloroethane (HCA) eluted at 4.5 minutes, perchloroethene (PCE) at 3.18 minutes, and trichloroethene (TCE) at 2.29 minutes. Isolation of Dehalogenase

A portion of the ammonium sulfate precipitate was added to a Sephadex G-25™ (2.5 x 50 cm) column to remove the salts. The activity eluted in the void volume. The void volume was added to a 5 x 50 cm Sepharose CL-6B™ column. The activity eluted in the 500,000 to 100,000 dalton range when compared to standard proteins. The sample was loaded in 20 mM phosphate buffer, pH 7, and eluted with the same buffer. A 250 ml aliquot could be loaded to clean the sediment sample and, subsequently, a 50 ml

aliquot was loaded to determine relative size. Fraction size was either 50 ml or 100 ml.

The active fractions were made up to 1.7 M ammonium sulfate in phosphate buffer, pH 7, and loaded onto either a gravity-fed phenyl sepharose column (1 x 35 cm) or a BioCAD™ phenyl column (1.6 ml) . The column was loaded and washed in 3 M ammonium sulfate, pH 7. A decreasing gradient of ammonium sulfate was added, then the column was washed with buffer at pH 7.

The dehalogenase activity adheres tightly to the hydrophobic column eluting at about forty column volumes, or ten to fifteen column volumes after washing with buffer alone. The active fractions were adjusted to pH 8 with phosphate, then loaded onto a DEAE PRODUCTIV™ column. Once the entire sample from the phenyl column was loaded, the dehalogenase was eluted by raising the pH to 9. The dehalogenase eluted in the first three fractions. These fractions were added onto a zinc- imminodiacetic acid column (either a 0.8 x 15 cm gravity fed column or a 1.6 ml BioCAD column) in 0.5 M NaCl, 2.5 mM imidazole, pH 7. The column was washed with this buffer, then an increasing gradient of imidazole in NaCl was added until a concentration of 25 mM imidazole was reached. The column was washed for five column volumes, then the pH was dropped to pH 3. The dehalogenase was eluted soon after the eluant reached pH 3. The active fractions were desalted using the Sephadex G-25™ column and stored in 50 mM ascorbic acid, pH 5, or 20 mM phosphate buffer, pH 7, in a refrigerator. Production of Monoclonal Antibodies Sufficient purified protein was supplied to the University of Georgia Monoclonal Antibody Facility, Athens, Georgia, to produce antibodies to the

protein. Three mice were injected with 100 μg per mouse and six weeks later 50 μg per mouse was administered in a second injection.

The mouse having the best immune response to the protein was sacrificed, the spleen removed and minced, and single spleen cells fused to cancer cells to produce a hybridoma containing both the cancer cell and the mouse spleen cell by established methods. All hybrid cells that grew in tissue culture were tested for an antibody response to the purified dehalogenase by the following procedure.

The protein (antigen) was attached to an ELISA plate (either an Immunolon 4™ or a PVP plate) by adding 1 μg of antigen in 50 μl of 50 mM Na ₂ B ₄ 0 ₇ (borax) , adjusted to pH 8.5 with boric acid and preferably refrigerated overnight at 10°C or mixed at 37°C for one hour. The inclusion of 10 mM potasium periodate to the antigen prior to addition of the borate improves the reproducibility of the binding. The periodate produces aldehydes on the carbohydrate sidechains, allowing for improved binding to the plastic solid phase.

The unbound sites were blocked by incubation for one hour with 50 μl of either gelatin (1%) in phosphate buffered saline or nonfat dry milk (5%) in phosphate buffered saline at room temperature. The plates were washed five times in phosphate buffered saline, then 50 μl of serum from each colony to be tested was added and incubated for one hour at room temperature with gentle mixing. The plates were washed, and 50 μl of a 1:500 dilution of a goat anti-mouse antibody, conjugated with alkaline phosphatase or horseradish peroxidase, was added and incubated for one hour. The plates were washed and the substrate of the reporter enzyme added in the buffer of the reporter enzyme. For

alkaline phosphatase, 1 mM MgCl ₂ , 96 ml diethanolamine per liter, pH 9.6, and 1 mg/ml of para-nitro phenyl phosphate were added. (Sigma Chemical Co., St. Louis, MO) The plates were allowed to react for thirty minutes, then read at an absorbance of A405 on an ELISA plate reader. Each serum sample was tested against both the antigen as well as to the blocker. Plate to plate variability was assayed by the inclusion of positive controls, usually two wells of purified nitrate reductase assayed with a previously purified monoclonal antibody. Any serum sample having a high background against the plastic or blocker was discarded. For analyzing the position of the dehalogenase in column fractions, 20 μl of a column fraction was added to the unmodified plate, then 30 μl borate buffer was added, incubated for one hour, and blocked as described above. Characterization of Dehalogenase Size

A portion of the active fraction from a zinc- metal affinity column was analyzed for absorbance (A280-A340) of the subunits after boiling in sodium dodecyl sulfate on a capillary electrophoresis system. The protein showed peaks of approximately 300,000 daltons, 89,125 daltons, 63,000 daltons, 39,000 daltons and 28,000 daltons. There were two peaks of smaller size, possibly 12,000 and 7,000 daltons. The separation of the purified enzyme by sizing on a Sephadex CL-6B™ column showed activity with perbromoethene (PBE) for the following molecular weights 300,000, 131,000, 90,000, 60,000, 39,000 and 28,000. There was no activity at the lower molecular weights. When sizing was performed as the first step, the activity eluted only as a high molecular weight peak.

Metals

A sample of the purified dehalogenase was supplied for metal analysis by graphite furnace AA. The sample was spun through a 10,000 molecular weight cutoff filter prior to analysis, then washed and resuspended in distilled water. The sample contained 110 μg/ml protein and contained 119 μg/liter copper. This calculates to less than one mole of copper per mole of protein. The sample contained 48 irons per mole of protein assuming a 300,000 dalton molecular weight and 110 μg/ml protein.

Temperature

The substrates, hexachloroethane (HCE) and perbromoethene (PBE) , were prepared anaerobically in 125 ml serum bottles and kept on ice. Autosampler vials of 1.8 ml were gassed with argon, and 20 μl, equal to 1 μg of purified enzyme, was added. The cold substrate was added, the tops sealed and samples placed in incubators at temperatures ranging from 10°C to 75°C. The samples were incubated for 24 hours prior to analysis by GC-ECD. The samples having the most degradation was set to 1.0 and all other samples were compared to it.

ES

The two substrates, hexachloroethane (HCA) and perbromoethene (PBE) , were prepared anaerobically in distilled water and enzyme added as described above. The substrate was added to a final concentration of 0.1 M potassium phosphate buffer adjusted with phosphoric acid or potassium hydroxide to the pH value given. The substrates were added to fill the vials, then incubated at room temperature before analysis one day later by GC-ECD. All samples were normalized to the

condition that gave the best degradation as described above.

Reaction Kinetics

Substrates were prepared as described above, enzyme added, immediately capped, and reaction time initiated from the time of enzyme addition. Possible Electron Sources

Reactions were performed as described above, the electron source added, then the enzyme. Reaction kinetics were compared to enzyme reaction that did not contain a possible electron source. Results

The enzyme is a multi-subunit enzyme containing one copper and as many as 48 irons per mole. Both the size exclusion chromatography and the denaturing capillary electrophoresis data indicated that the enzyme has a native size of approximately 300,000 daltons and contains subunits of 28 kilodaltons, 39 kilodaltons, and 89 kilodaltons, most likely in a 2:2:2 ratio.

Temperature and pH optimum were determined for the dehalogenase for an alkene (perbromoethene) as well as an alkane (hexachloroethane) . The pH range for both substrates was nearly the same. The alkane (hexachloroethane) prefers a lower pH (pH 4- 5) , but will work at pH values up to 8 (pH 3-8) . The enzymatic activity using the alkene (perbromoethene) had a similar pH range (pH 3-9) , but the optimum was pH 6-7. The temperature activity using the alkane

(hexachloroethane) showed a broad temperature range of 10°C to 60°C, but still had activity at 70°C. The activity using the alkene (perbromoethene) showed a range of 25°C to 68°C, with an optimum at 37°C to 56°C.

The rate of degradation of hexachloroethane by the dehalogenase was twice as fast as the rate of

degradation of perbromoethene, and the K,„ of the enzyme using hexachloroethane as the substrate was 12-fold greater than using perbromoethene.

The half-lives of the individual reactions of the rate of degradation of hexachloroethane or trichloroethene alone are in the order of 100 to 200 minutes, whereas the half-lives of trichloroethene and perchloroethene in sediment were on the order of 200 days. When alkanes and alkenes were mixed, the hexachloroethane concentration did not vary over the time of the experiment (less than 250 minutes) , and the perchloroethene was degraded with a half- life of approximately 200 minutes. Ascorbic acid accelerates the reaction of perchloroethene to other unknown products.

A comparison of the binding of seven monoclonal antibodies specific for the dehalogenase with the extract of ten plants is shown below in Table 11. The antibodies designated 1F9.H12, 2F6.H4, 10E2.E1, 6F1.G5, 4F4.C2, 4D6.D3, and 11H11.C10 are monoclonal antibodies.

These results provide some indication of the presence of the enzyme in the plant. The highest response observed was with oak leaves, dogwood leaves, or the weed quakegrass. However, parrot feather showed a consistent response as did stonewort. These latter two plants can grow in submerged solutions, therefore they were chosen to determine whether halogenated hydrocarbons could be degraded by incubation with an intact plant system.

Table 11. Dehalogenase concentrations in selected plants determined by monoclonal antibody assay

Antibody

Plant 1F9.H12 2F6.H4 10E2.E1 6F1.G5 4F4.C2 4D6.D3 11H11.C10 polvclonal soft rush 8 32 16 4.4 12 4 13 hornwort . 8 28 8.4 7 58 4 parrot feather 42 22 15 10 3.6 20 13 26 compacta holly 30 11 70 7.3 124 73 5 teathumb 42 96 20 22 16 182 110 22 stonewort 10 60 64 80 34 106 60 16 spirogyra 11 64 32 47 55 12 52 17 water oak leaves 208 40 36 29 185 122 170 112

S dogwood leaves 148 32 64 116 182 216 96 110 quakegrass 33 80 80 245 115 388 34 88

The results shown in Figures 26a and 26b show the degradation of hexachloroethane and perchloroethene by four plants incubated together with the substrate. Within the first 60 to 90 minutes, as much as half of the halogenated hydrocarbon (at 3 ppm) either binds to the plant or is degraded. The products formed are similar to those formed in the enzyme system.

Example 8. Remediation of TNT Contaminated Soils Using Proteins Extracted from Sediment and Hornwort

This purpose of this experiment was to enhance the degradation of trinitrotoluene (TNT) in contaminated soil samples collected from six munitions sites. TNT was reduced to the corresponding amino compounds using proteins extracted from a local sediment and the aquatic weed, the hornwort, as remediators.

A nitroreductase was purified, a monoclonal antibody specific for the nitroreductase prepared, and immunoassays conducted as described above in Example 6. Field sampling and analysis indicated that the common aquatic plant, the hornwort, had the highest activity of the nitrate reductase protein complexes. Studies with TNT and hornwort indicated that the plant had a large capacity to reduce TNT to the triamino compound. Products other than the corresponding amino compounds were not apparent in the analysis.

These studies were conducted in two parts. The first was to react TNT contaminated soils with protein isolated from sediment to give the reduced amino compounds. The second part was to react TNT contaminated soil samples using a source of the protein, the hornwort aquatic plant.

Initially, samples were analyzed by high pressure liquid chromatography. Subsequently, samples were

analyzed by a capillary electrophoresis analytical method.

Materials and Methods

Soil samples were provided by the U.S. Army Corp of Engineers and were labelled Radford, Lone Star, Iowa,

Sub-Base, McAlester and Hastings. Treatments of these soils with protein extract and hornwort are described as follows.

Part l 1) Controls - 5.0 g soil from each sample was weighed into individual screw cap test tubes and 9 ml water was added to each tube.

2) Remediation - approximately 195 g of each soil sample was transferred into a 0.5 liter reactor and mixed with 350 ml of protein extract (approximately 1 mg/ml total protein) . The samples were then placed in a glove box (nitrogen environment) and shaken on a shaker at low speed.

Part 2 Hornwort as a remediator for soil samples

Samples of contaminated soil, hornwort, and distilled water were mixed in Erlenmeyer flasks under aerobic conditions. The preparation methods for soils with high and low concentrations of contamination are described below. For the more concentrated soil samples, larger quantities of hornwort were used. Low concentration contamination

1) Controls - 10.0 g sub-samples of each sample of Radford, Lone Star, Iowa and Sub-Base were weighed into individual Erlenmeyer flasks and 50 ml water was added.

2) 10.0 g sub-samples of each sample of Radford, Lone Star, Iowa and Sub-Base were weighed into individual Erlenmeyer flasks and 50 ml water and 10 g of intact hornwort plant was added to each flask.

High concentration contamination

3) Controls - 10.0 g each of McAlester and Hastings soils were transferred into individual Erlenmeyer flasks and mixed with 100 ml water. 4) 10.0 g each of McAlester and Hastings soils were transferred into individual Erlenmeyer flasks and mixed with 100 ml water and 30.0 g of intact hornwort.

Characterization of Contaminated Soils

Soil pH, total organic carbon (TOC) , and metal content of each soil sample were determined with an Accument™ Microprocessor ISE/pH meter, a CD-85A high temperature TOC analyzer and a Perkin-Elmer™ ICP- emission spectrometer, respectively. Analytical Analysis HPLC

Initial analysis was carried out on a Waters™ 740 HPLC equipped with a variable wavelength UV detector. Conditions for HPLC analysis were as follows: PRP-1 250 x 4.1 mm column; 50 μl injection volume; 35% H ₂ 0, adjusted to pH 10 with NaOH, 65% acetonitrile mobile phase; 1 ml/min flow rate, UV detector wavelength set at 238 nm.

MECE

Subsequent analysis was performed using micellar electrokinetic capillary electrophoresis (MECE) , a Spectra Phoresis 1000™ system with a UV detector. Conditions for MECE analysis are as follows: fused silica 70 cm x 75 μm (i.d.) column; 25 mM sodium dodecyl sulfate (SDS) in 2.5 mM borax buffer (Ph 8.56) mobile phase; hydrodynamic injection type, usually 2 seconds (5.37 nL/s) ; 20 dV running voltage; 220 nm UV wavelength.

Remediation

Protein Extract as a Remediator

At given time intervals, the pH and Eh of treated samples were determined. Both control and treated samples were centrifuged and the supernatants were decanted. The sediment was extracted with acetonitrile. In general, for contaminated soils containing a low TNT concentration, a ratio of 1:3 (g- sed./ml-H ₂ 0) was used, while for high levels of TNT contamination a ratio of 1:10 (g-sed./ml-H ₂ 0) was used. Both supernatant and sediment extract, with or without dilution, were analyzed for TNT and its reduced products by an HPLC (UV detector) . Aquatic Weed Hornwort as a Remediator At given time intervals, both supernatant and sediment extract of treated soils were analyzed with the same method as above. At day 43, each supernatant and sediment extract of treated soils was analyzed by micellular electrokinetic capillary electrophoresis (MECE) (UV detector) . Data Analysis

Calculations for TNT and its reduced product concentrations were based on the peak areas from the HPLC chromatograms and the MECE electropherograms. At given time intervals, two aliquots of each soil slurry were collected. Both supernatant and sediment extract from each soil slurry aliquot were analyzed using HPLC with MECE. The results from HPLC or MECE were then averaged. Because the relative standard deviations of area for HPLC and MECE previously determined are less than 2%, there was only one injection per sample.

In the HPLC analysis, the inability to separate both dinitromonoaminotoluene isomers and both diamino- mononitrotoluene isomers was observed. Also,

triaminotoluene eluted with the solvent front, preventing quantitation of the formation and disappearance of triaminotoluene. The separation of all amino products could be obtained with MECE. In the MECE analysis, triaminotoluene could be readily detected. In calculating the concentration of the products, it was assumed that both dinitro- monoaminotoluene isomers and both diamino- mononitrotoluene isomers have the same extinction coefficients.

Results and Discussion

The soil characteristics of the samples are listed in Table 12. The pH values are the pH of a 1:1 soil- water (g/ml) slurry. The pHs of the soil slurries with the exception of Lone Star were pH 6 to 7 and optimal for remediation. No steps were taken to adjust the pH of any of the slurries, even Lone Star, which had a pH of 4.8. Organic carbon content was 1 to 2% with the exception of Lone Star, which was only 0.35%.

Table 12. Character!*itics of Contaminated soil

Samples

Soil Source EH TOC Metal Content

( % ) (ppm)

Fe Pb Zn

Radford 7.6 1.9 206 <0.5 1.3

Lone Star 4.8 0.35 53.6 <0.08 0.67

Iowa 7.8 0.13 95.8 <0.25 0.2

Sub-Base 6.0 1.2 106 <0.05 0.26

McAlester 6.7 1.2 106 <0.07 0.39

Hastings 7.2 2.05 112 <0.08 0.66

Protein Extract as a Remediator

The protein additive was obtained by extraction of sediments with high organic carbon content from local sources as described above in Example 6. In this

experiment, only crude protein with high activity was used.

For all samples, the pH appeared to increase with time while the Ehs, with the exception of the Lone Star sample, all went negative with time. In the supernatant of low TNT concentration soils, the aqueous phase concentration decreased to below the limit of detectin for Radford, Iowa, and Sub-Base. However, it should be noted that the initial concentration of TNT in Radford was below the detection limit. For the highly contaminated samples, there was no apparent attenuation of TNT in the supernatant. For the sediments, there was no apparent decrease in TNT concentrations. This is attributed to the high concentrations of TNT and the short incubation time. Bioreactor

After the protein remediation step, the samples were returned to WES for biological remediation in a biological reactor and then were returned for analysis of TNT and potential products. Compared with the results achieved above with the sediment, TNT concentrations were attenuated. For the high concentration samples (McAlester and Hastings) , dinitro-monoaminotoluene concentration increased significantly. It was not know whether the attenuation was due to continued activity of the protein or microbial activity in the bioreactor. Aquatic Weed Hornwort as a Remediator In a time course study, hornwort collected from Beaver Dam, a small watershed pond near Athens, Georgia, was used as a remediator. The results are summarized as follows. The concentration of TNT and its reduction products in the supernatant and in the solid phase decreased significantly by day 43. For

four of the samples, Radford, Lone Star, Iowa, and Sub-Base, there was no detectable TNT in the supernatants even at time zero and no detectable TNT in the solids at day 43. Apparently, the nitroreductase activity is so high in the supernatant that the TNT is transformed as fast as it can be transferred to the aqueous phase. in the fifth sample (McAlester, high TNT concentration) , the supernatant concentration was reduced from approximately 70 ppm to 1 ppm and the solids concentration was reduced from about 8723 ppm to 11 ppm. The dinitroamino and diaminonitro isomers were degraded as well. Triaminotoluene could not be detected at day 43 when analyzed by MECE in the Hastings samples treated with hornwort. The limit of detection was established at about 0.01 ppm. It is believed, based on other studies with substituted anilines, that the triaminotoluene is rapidly oxidized by autooxidation or microbially to catechols. Catechols are known to be readily microbially oxidized to innoxious compounds.

One of the bio-treated TNT contaminant soil samples (Hastings) was mixed with hornwort under aerobic conditions. The initial concentrations of dinitro- monoaminotoluene in this treated soil were 62 and 3910 ppm in the supernatant and the sediment, respectively. After three weeks, only 5.5 ppm and 358 ppm of dinitro-monoaminotoluene were found in the supernatant and solid phase, respectively. Conclusions

The isolated nitrate reductase protein has the potential to remediate soils contaminated with low concentrations of TNT. It was effective in reducing four of the six soil samples. Hornwort has the potential to remediate soils contaminated with high

concentrations of TNT and was effective in reducing all six soil samples. Hornwort treatment should be particularly suited for in situ remediation. Also, hornwort is effective in indirectly controlling the pH of the soil/water system and eliminating poisoning by lead and zinc, thus making it applicable to soils with a wide range of physical and chemical characteristics.

Example 9. Degradation of Hexachloroethane by Enzyme- coated Sand

This purpose of this experiment was to demonstrate degradation of a halogenated hydrocarbon with the isolated dehalogenase when the dehalogenase was bound to glass beads or sand. Procedure

A 0.1 M poly D-lysine solution (Sigma Chemical Co., St. Louis, MO) was prepared in 0.1 M borate, pH 8.5, combined with glass beads so that the beads were covered completely, and incubated at room temperature for 30 minutes to one hour. The poly D-lysine solution was decanted, beads poured into shallow glass trays, and dried in a low temperature oven (57°C) until dry. The resulting beads stuck together and were white in color. 10 mM KI0 ₄ (up to 5mg/ml = 0.04 M) was added to a volume of enzyme (a glycoprotein) in pH 7 buffer and incubated for one hour at room temperature in a dark bottle due to the light sensitivity of the reaction. The activated enzyme was added to the dried, coated beads and incubated for one hour at room temperature. 0.1 M sodium cyanoborohydrate or sodium borohydrate (pH 8.5 to 9.5) was added to complete the covalent bonding of the protein to the lysine groups. The beads were washed with pH 4.5 ascorbic acid three to five times and analyzed for activity. Reactivation of

inactive protein can be achieved by adding 10 mM CuS0 ₄ in pH 4.5 ascorbic acid mixture and incubating for one hour. Reactivation may take up to two days.

Beads were washed with approximately 6-10 column volumes with gentle mixing until all copper was removed with ascorbic acid.

The results of degradation of hexachloroethane with enzyme-coated beads, or sand are shown in Figure 27. The results indicate that, when bound to glass beads or sand, the dehalogenase retains the ability to degrade halogenated hydrocarbons.

Modifications and variations of the contaminant remediation method and composition will be obvious to those skilled in the art from the foregoing description. Such modifications and variations are intended to come within the scope of the appended claims.