Wet cleaning" for the aquatic environment

Degradation of TCE and EDB using the CYP2E1 protein in transgenic duckweed

Clean water is essential to the life of all organisms. Because the environment contains toxins and xenobiotics, many organisms have evolved enzymes to degrade such harmful species to less damaging forms. In mammals, cytochrome P450s serve as the primary interface for the metabolism of drugs and other xenobiotics. Human cytochrome P450 2EI (CYP2E1) is one the most conserved xenobiotic metabolizers in mammals1, and it is known for its broad substrate specificity. While it is biologically essential, CYP2E1 is not perfect—due to its mechanism of action; it is hypothesized to uncouple from it accessory proteins, thus allowing reactive intermediates to migrate out of its active site, leading to the damage of tissues2.

Herein we propose a plan to exploit the astounding abilities of CYP2E1. First and foremost, we will remove this protein from its natural environment of the human liver and transform it into common duckweed. Then we will harness it as the central player in a constructed wetland designed to remove xenobiotics directly from contaminated water, rather than the blood stream of a poisoned individual. We will show how human CYP2E1 behaves in the plant system and attempt to correct its known weaknesses.

Specific Aims

Aim 1: Structural characterization and manipulation of human CYP2E1 in complex with plant NADPH-dependent P450 reductase and cytochrome b5

We will use indirect enzyme-linked immunosorbent assays (ELISAs) to determine if human CYP2E1 can bind to plant homologs of NADPH-dependent P450 reductase (NPR) and cytochrome b5 (Cyt b5), then use bimolecular fluorescence complementation (BiFC) to determine the basal interaction of these proteins within Arabidopsis thaliana. Because we plan on enhancing the interaction of human CYP2E1 with the plant accessory proteins, we will first determine if conformational allostery exists within human CYP2E1 and if its association with plant NPR and Cyt b5 affects the substrate-binding pocket. Contact Rearrangement Network (CRN) Analysis will be used to measure changes in the contact between pairs of associating protein crystal structures. Because we hypothesize that we can generate an enzyme that does not uncouple as easily by tightening the interactions between CYP2E1 and its redox partners, we will perform site directed mutagenesis of the amino acids at the interaction interfaces in order to further promote interaction, taking care to avoid the amino acids that are allosterically coupled to the active site. We will perform BiFC assays to determine if the mutants generated interact better. This series of experiments will result in the generation of an enhanced version of CYP2E1, which we will call cytochrome p450 2E1 enhanced or 2E1en.

Aim 2: Transformation of CYP2E1 into duckweed and the characterization of enhanced uptake and metabolism of two common environmental contaminants of the generated system.

We will determine an expression system that best suits 2E1en. Agrobacterium-mediated transformation is commonly used in duckweed3, and it has been used to transform human CYP2E1 into other plants4, so we will try this system first. We will verify the transformation of duckweed with 2E1en via Northern and Western blots with antibodies that are specific for CYP2E1 and recognize the parts of the protein that have not been manipulated. We will then assay the ability of the 2E1en-transformed duckweed to remove two known substrates of human CYP2E1, trichloroethylene (TCE) and ethylene dibromide (EDB), from water and degrade them to less harmful compounds. We will assay for these degradation products as a measure of success.

Aim 3: Analysis of the performance of CYP2E1-transformed duckweed in constructed wetlands.

A constructed wetland (CW) analog will be built in a greenhouse to allow for control of environmental factors and systematic determination of physical CW parameters including hydraulic retention time, hydraulic loading rate, and organic loading rate. 2E1en-transformed duckweed will be grown in synthetic wastewater doped with ecologically-relevant levels of TCE and EDB. An experimental CW will then be built at a research site to study environmental effects both on and of the CW. Here, the water in the CW will be supplemented with industrial effluent, contaminated groundwater, or secondary wastewater effluent. TCE and EDB uptake and metabolism will be quantified in both greenhouse and research site CWs.

Background and Significance

Water polluted with TCE and EDB

One class of environmental pollutant is volatile organic compounds (VOCs). They have a high vapor pressure and relatively low solubility. As a result, they can easily transfer between air, soil, and water, which increases their ability to affect the environment and human health. TCE and EDB are two examples of VOCs. The EPA sets maximum containment levels (MCL) for pollutants, and the MCL defines the concentration of pollutant that does not affect human health. The MCL for TCE is 5 µg/L, and the MCL for EDB is 50 ng/L. Both pollutants are found in groundwater and domestic wells at levels above these concentrations5. Well water is especially at risk of high pollutant levels because it is not regulated or treated like other public water6. Finding more efficient ways to remove TCE and EDB from drinking water would improve human health.

TCE primarily enters the water supply from waste that is generated at factories that degrease metals7. Chronic TCE exposure can lead to infertility8, liver damage, and elevated risk for cancer9. EDB is a gasoline additive usually used in aviation fluid for its "anti-knock" properties, meaning it prevents premature ignition. It enters the water supply through gasoline spills, chemical waste, and from its previous use as a pesticide. It is suspected to harm the stomach, reproductive system, and kidneys, as well as increase the risk of cancer10.

TCE is broken down by CYP2E1 in humans. The byproducts of this degradation include S-(1,2-dichorovinyl)glutathione (1,2-DCVG) and trichloroethano-glucoronide (TCOH-Gluc), both of which are excreted through urination. Two other potential metabolites are trichloroethanol (TCA) and dichloroethanol (DCA). These two metabolites are largely responsible for the liver damage and increased hepatocarcinogenicty caused by TCE exposure9.

TCA interacts with proliferation-activated receptor a (PPARa), and as a result, peroxisome proliferation (PP) occurs. PP leads to reactive oxygen species that can damage DNA and proteins. Another way that these metabolites harm the liver is by depleting the stores of S-adenosymethionine (SAM), which is needed for proper DNA methylation. Loss of methylation due to TCA and DCA exposure leads to overexpression of c-Myc and activating-protein 1 (AP1). Both of these proteins are used in cell cycle control, and their overexpression can cause increased cellular proliferation. Reactive oxygen species and hypomethylation are mechanisms for liver damage and tumor formation9.

To address the problem of TCE and EDB in drinking water, we propose to use transgenic phytoremediation technology. Phytoremediation is the use of plants to degrade pollutants and toxins found in the environment. Compared to other methods of environmental remediation, it is cheaper, better at preserving topsoil, and easily monitored. Traditional methods of environmental remediation can have prohibitively high costs, so contaminated sites are often abandoned. These sites are known as brownfields, and over 500,000 of them exist in the United States11.

Microbes have been used as bioremediators, but they often need additional carbon and energy sources to survive in contaminated locations. They also sometimes need additional cofactors in order to degrade the environmental toxins at the site, and these cofactors can include toxins such as phenol. (reviewed in Fingerman 2005) Unlike microbes, plants receive their energy from the sun and nutrients from soil, making them more adaptable. On average, phytoremediation costs ten times less than traditional, mechanical means of remediation11.

Plants can inherently metabolize and sequester pollutants, but they can be enhanced using transgenics. In that case, proteins from other species are added to the plant, allowing it to then metabolize or sequester new substrates12. Many mammalian proteins that metabolize organic chemicals and heavy metals have been added to plants13. Examples include glutathione S-transferases14-16  and cytochrome P450 proteins4,17,18, both of which metabolize xenobiotics.

Many different species of plants have been used in phytoremediation. In our case, we will be using species of duckweed. These are small, free-floating aquatic plants that reproduce by budding and have a doubling time of less than one day. They consist of fronds and roots, both of which absorb water and the nutrients within water. They grow easily in many climates, making them useful for phytoremediation. Lemna gibba and Spirodela punctata are species of duckweed that are best at overexpressing introduced proteins19. Duckweed is widely used in wastewater treatment, showing it can survive in highly polluted water3,19,20.  We will introduce an enhanced version of the human protein CYP2E1 to duckweed species and use the duckweed in polluted water to degrade harmful xenobiotics.

Introduction to the Cytochrome P450 Superfamily

Cytochrome P450 is superfamily of monooxygenase enzymes that use molecular oxygen (O2) to oxidize substrates and generate the reduced product of water.  They possess a critical heme group that absorbs at 450nm, which was the basis for the superfamily name21.  These enzymes usually require the presence of a reducing agent (NADH or NADPH) that will alternatively become oxidized (to NAD+ or NADP+) through the transfer of electrons during the enzymatic reaction. Cytochrome P450s (CYP450s) are primarily involved in steroid hormone biosynthesis within humans and other organisms21,22. 

CYP450 monooxygenases are present within mitochondria, and their catalytic site is exposed to the matrix where it participates in oxidizing specific substrates.  The basic reaction catalyzed by mitochondrial CYP450s is hydroxylation of substrates using NADPH as a reducing agent and molecular oxygen (Fig. 1). The reaction requires that an electron be carried from NADPH to the CYP450, where the O2 molecule is held in place by the iron atom found within the heme.

Heme is a common protein prosthetic group in which an iron ion center is coordinated by a porphyrin ring (Fig. 2). Heme prosthetic groups are found in many proteins that require the coordination of oxygen atoms, such as hemoglobin.  The electron flow involved in CYP450 reactions requires both a flavoprotein and an iron-sulfur (Fe-S) protein.  Flavoproteins are electron transport proteins that contain a nucleic acid derivative of riboflavin, such as flavin adenonine dinucleotide (FAD).  Similarly, Fe-S proteins are proteins that contain iron-sulfur clusters that facilitate the transfer of electrons along the mitochondrial membrane. Flavoproteins and Fe-S proteins are abundant within mitochondrial membranes across species21.

Figure 1 Blue arrows indicate electron flow between proteins located in the mitochondria in order for the enzymatic reaction of CYP450 to take place. First, NADPH transfers two electrons to the FAD protein, thus reducing it to FADH2. These electrons are then transferred one at a time to the Fe-S protein.  From there, single electrons are passed to the CYP450 enzyme, which couples molecular oxygen (O2) and the substrate (R-H) with the electron to form H2O and R-OH as products. (Modified from Fig. 19-36 of Nelson & Cox, 2008)

Figure 2 Organic structure of the heterocyclic porphyrin ring that coordinates an iron ion (Fe3+ or Fe2+). This structure is a crucial prosthetic group in CYP450 proteins that is required for the coordination of an oxygen molecule (O2) for the hydroxylation of substrates. (Generated using ChemDraw).

A large subgroup of the CYP450 superfamily is specialized for a wide array of xenobiotic metabolism and is subsequently localized on the endoplasmic reticulum of human liver (hepatocytic) cells21. This specialized group of heme monooxygenase enzymes functions similarly to non-hepatocyte CYP450s through oxidation of various substrates. Hepatocyte CYP450s are specialized for the degradation of hydrophobic xenobiotics.  Xenobiotics are synthetic chemical compounds that are not usually found in nature but can pose mutagenic and carcinogenic threats on cells if not deactivated21,23. CYP450s of the liver partake in hydroxylation of xenobiotics in order to decrease the compound's hydrophobicity so they can be easily excreted by the kidneys. CYP450s act on both exogenous (xenobiotic) and endogenous (steroid hormone) substrates within the human body. Because of the wide spectrum of substrates that can be targeted by CYP450s for degradation, we were interested in finding a CYP450 enzyme that could degrade harmful materials found in aquatic environments.

Generally, CYP450s are highly specific for the substrates on which they act, so many different kinds of CYP450s have evolved in mammalian and plant species21.  In humans alone, there exists at least 57 different, but highly related, CYP450s that are found in the endoplasmic reticulum of various tissues, including the liver22. Only in the liver are CYP450s thought to have a role in xenobiotic metabolism, while in extrahepatic tissues it is thought that they play a role in hormone biosynthesis22,23.  All of the human CYP450s differ in the size and shape of the active site, which is indicative of substrate specialization by these enzymes23. About seven CYP450s account for approximately 98% of the drug metabolism in humans, as shown in Fig. 3. This proposal specifically focuses on CYP450 2E1 (CYP2E1) because it has a broad substrate specificity and experimentally, CYP2E1 has been successfully transformed into plants to enhance xenobiotic metabolism4,17,18.

Figure 3 Seven CYP450s account for approximately 98% of the drug metabolism in humans.  (Graph generated by from data found in Ortix de Montellano, 2005).

Function and Structure of Cytochrome P450 2E1

CYP2E1 is known to successfully degrade trichloroethylene (TCE), which is known as "one of the most widespread environmental pollutants in the industrialized world"4, and has been used in successful phytoremediation experiments4,17,18. Additionally, CYP2E1 functions on several other substrates that predominantly have molecular weights of 70-100 grams/mole, such as those shown in Table 121-23.  Many of these compounds are suspected carcinogens and their efficient degradation within the human body is crucial for survival. CYP2E1 functions to detoxify substrates through oxidation in order to convert these mostly hydrophobic substrates to hydrophilic compounds that can then be excreted via the kidneys22. We are interested in the enzymatic activity of CYP2E1 because of the wide array of substrates that it acts on. CYP2E1 has the potential to eliminate many carcinogens present in the environment due to human activities. For this project, we will be specifically looking at the enzymatic activity of CYP2E1 degrading TCE and EDB found in the environment.

The structure of CYP2E1 was determined by X-ray crystallography using two inhibitors: indazole (INZ) and 4-methylpyrazole (4MP) (Fig. 4). The Protein Data Bank codes for the CYP2E1 with the indazole inhibitor (CYP2E1-INZ) and CYP2E1 with the 4-methylpyrazole inhibitor (CYP2E1-4MP) structures are 3E6I and 3E4E, respectively23. The CYP2E1-INZ consists of amino acids 31-137 and 140-493, while CYP2E1-4MP contains amino acids 31-137 and 140-493. The missing amino acids are part of a loop motif that contributed difficulty in crystallizing the structure23. Residues 1-30 are part of the N-terminal transmembrane helix and

also inhibited crystallization of the protein. Both structures show that CYP2E1 appears to be comprised of an a+ß domain with no repetitive pattern23.   Absorption spectra results showed CYP2E1 in a mixed spin state, indicating that water is bound to the sixth coordination site of the heme prosthetic group. It was also determined that, like other CYP450s, CYP2E1 has 12 helices (11 a-helices and one 310 helix), four ß-sheets, and multiple regions of random coils23 (Fig. 4). The most abundant types of helices are a-helices, which involve 3.6 amino acids per turn with a 1.5Å rise per residue and a helical radius of 2.3 Å.  These are the most common forms of helices present within proteins because they maximize the hydrogen-bonding potential of the amino acids involved in the helix. The 310 helix, however, has 3.0 amino acids per turn with a 2Å rise per residue and a helical radius of 1.9Å. Because the hydrogen-bonding of a 310­ helix is not maximized, it usually is only present for 1-2 turns, unlike a-helices which can exist for 2-20 turns. Thus, some steric hindrance exists for a 310 helix, but this additional, unfavorable energy is likely compensated for in another aspect of the protein, such as in burying the hydrophobic regions of the protein. Several ß-meander motifs are also present, in which antiparallel ß-strands are linked by a hairpin loop.

Figure 4 Structure of CYP2E1 with indazole (INZ) inhibitor [purple] (A) and 4-methylpyrazole (4MP) inhibitor [orange] (B). This shows that within CYP2E1, there are 12 helices (11 a-helices and one 310 helix) [red], four ß-sheets [blue], and several regions of random coils [green]. (Pictures generated using Swiss PdbViewer 4.0.1, Protein Data Bank codes 3E6I and 3E4E for INZ and 4MP, respectively.)

The percentages of favorable, allowed, and disallowed amino acids, based on the Ramachandran plots, are shown in Table 2 for CYP2E1-INZ and CYP2E1-4MP. Although it seems contradictory that amino acid residues are present in the "disallowed region" of space, it is likely that the binding energy of favorable interactions between other residues makes up for the energy required for interactions in the disallowed region. Because the two protein complexes with different inhibitors (INZ and 4MP) are nearly identical23, as shown by Table 2, we will only focus the analysis of the structure of CYP2E1 on the 4MP inhibitor because it has the lower percentage of amino acids in the "disallowed region," so it is probably more indicative of the true structure of the protein.

The site volume of the active site of CYP2E1-4MP is 189.5 Å3, which is consistent with small (~70-100g/mol) molecular weight substrates23. Due to its small active site volume, 

Figure 5 Amino acids present within or near the active site of CYP2E1. (A) Threonine 303 hydrogen-bonding to the 4MP inhibitor [orange] on the heme prosthetic group [magenta] of the CYP2E1 protein, as well as with the carbonyl group of Ala299. (B) Hydrophobic residues that border the active site, including Phe106, Ile115, Phe116, Val364, Leu368, and Phe478. (C) The five phenylalanines that create a hydrophobic "roof" above the active site; Phe106, Phe116, Phe207, Phe298, Phe478. (D) The electrostatic potential of the molecular surface of CYP2E1-4MP, showing the predominant positively charged [blue] surface of the protein, where it is believed to coordinate the negatively-charged region of NADPH-cytochrome P-450 reductase and cytochrome b5. Positively charged areas are shown in red and neutral areas are white. For the amino acids shown in A, B, and C, red indicates oxygen, blue indicates nitrogen, white indicates carbon, and all hydrogens are not shown. (Pictures generated using Swiss PdbViewer 4.0.1., Protein Data Bank code 3E4E)

CYP2E1 is known as a "molecular sieve" and is particularly good for small substrates23. Ala299 and Thr303 are part of the same a-helix that lines the active site. The 4MP inhibitor hydrogen bonds to hydroxide of Thr303, which in turn hydrogen bonds to the carbonyl group of Ala299, suggesting that these residues are important for catalysis (Fig. 5a). The 4MP inhibitor also coordinates with the sixth position of heme, consistent with spectral experimental results (Fig. 5a). On the other site of the active site, the a-helix residues Phe106, Ile115, and Phe116 create a border along with Val364, Leu368, and Phe478 (Fig. 5b). Through mutation studies of residues Val364 and Leu368 and Phe478, it has been determined that these residues are required

for CYP2E1 specificity and have been found to coordinate molecules in the active site.  Five phenylalanines, Phe106, Phe116, Phe207, Phe298, and Phe478, create a roof above the active site, which drastically reduces its size (Fig. 5c). Overall, the active site is small, globular, and mostly hydrophobic. Thr303 is the only polar residue present in the active site and is highly conserved across homologous proteins23. The predominantly nonpolar active site is ideal for interaction with small nonpolar (and some slightly polar) uncharged substrates. It is possible that other isozymes might act on small substrates, as CYP2E1 does, with a higher maximum velocity (Vmax­).  However, CYP2E1 is the highest affinity enzyme for small substrates, for which it frequently averages a low Michaelis-Menten constant (KM). This is yet another reason that CYP2E1 was chosen for our phytoremediation of TCE and EDB in our proposal.

Mechanism and Accessory Proteins

CYP2E1 activity requires two additional membrane proteins: NPR and Cyt b524,25. The interaction surfaces between human CYP2E1 and both NPR and Cyt b5 have been determined. These interactions stem mostly from a positively-charged "bowl"-like region on one side of CYP2E1 (Fig. 5d) which is hypothesized to interact with negatively charged residues of both NPR and Cytb523. Although other groups have reported successful cooperation between transformed mammalian CYP2E1 and plant NPR and Cyt b54,18, no enzymatic reaction scheme studies have been performed using transgenic plants expressing mammalian CYP2E1.
The first detectable product in the CYP2E1-mediated degradation of TCE is chloral, which is immediately hydrated to form chloral hydrate9.  It is thought that before chloral forms, a TCE-O-P450 reactive epoxide intermediate forms, and that this species is toxic because it can react with DNA and proteins2.  Chloral hydrate is further metabolized by an alcohol dehydrogenase to trichloroethanol (TCOH) or by an aldehyde dehydrogenase to trichloroacetic acid (TCA)26,27. Other groups have detected TCE metabolites in transgenic human CYP2E1 plant systems, including TCOH18. Tobacco plants that express CYP2E1 are not able to degrade TCA28, and TCA is not used as a metabolite to "read out" CYP2E1 activity in transgenic systems. It is thus possible that TCA does not form in the transgenic mammalian/plant system.

Transgenic tobacco that expresses mammalian CYP2E1 takes up TCE through its roots, but TCE accumulates in the leaves instead.  After a few days, the level of TCE in the leaves decreases.  This observation suggests that either the final steps of metabolism take place in leaves, or metabolism is fast in the stems and roots and slow in the leaves18.

In the reaction mechanism carried out by human CYP2E1 on EDB, EDB is first oxidized to bromoacetaldehyde18. Following oxidation, it is further metabolized to several metabolites, including bromoacetic acid, thiodiacetic acid, and a mercapturic acid derivative29. In plants transgenically expressing mammalian CYP2E1, it has been shown that EDB is broken down into a bromoacetaldehyde intermediate and a free bromide ion18. The free bromide ion is what is assayed for to confirm degradation of EDB by the transgenic plant.
The presence of cytochrome b5 has been shown to strongly induce CYP2E1 reactivity, 12-fold for p-nitrophenol, 25-fold for acetaminophen, 67-fold for 7-ethoxycoumarin, and 270-fold for aniline23. This increased reactivity is observed because Cyt b5 is involved in electron transfer, specifically it delivers the second electron to the reactive intermediate in the reaction mechanism and stabilizes CYP2E1. This hypothesis is supported by the observation that if apo cytochrome b5, which does not contain a heme and therefore cannot participate in electron transfer, is used instead of Cyt b5, then CYP2E1 loses its metabolism activity23. The protein-protein interaction between CYP2E1 and cytochrome b5 involves two cross-links, one between Lys 428 in CYP2E1 and Asp 53 in Cyt b5, and the other involving Lys 434 in CYP2E1 and Glu 56 in Cyt b530. Based on X-ray crystallography, the hypothesized interaction surface consists of the CYP2E1 ß-bulge and two a helices23.
In the instances in which human CYP2E1 has been introduced into plant systems4,17,18, it has been observed that the human protein works without its needed human accessory proteins. Currently the field collectively assumes that human CYP2E1 is able to co-opt the plant versions of Cyt b5 and NPR, but no one has characterized that protein-protein interaction. Genes encoding oxidoreductases in plants show high sequence similarity to animal oxidoreductases31. We have confirmed the sequence similarity between the human and plant copies of these genes using the NCBI Conserved Domain and BLAST. Because no duckweed species have a completely determined genome sequence, we instead used Arabidopsis thaliana as a reference. Both humans and A. thaliana have Cytochrome b5 (accession # for human, AAA63169). A. thaliana has five isoforms of Cyt b5, and their sequence similarity is 35%, meaning these sequences are fairly similar and that similarity is likely due to homology32. These proteins all share a Cytochrome b5-like Heme/Steroid binding domain that coordinates a heme group33. Human NADPH-dependent P450 reductase (NPR, accession # AAB21814) is most similar to A. thaliana P450 reductase 1 (accession # X66016), with a sequence similarity of 37%. These two proteins have great sequence similarity32. They also have the same three conserved domains: CYPOR, FMN red, and CysJ31. The CYPOR domain donates two electrons from NADH to the heme group in CYP2E1, thus allowing proper function of CYP2E133.  Due to the presence of conserved domains and sequence similarity, it is highly likely that these proteins are homologous, and that they play the same role in humans and A. thaliana. However, the interaction between human CYP2E1 and plant Cyt b5 and NPR has yet to be studied in a biochemical fashion.

Although CYP2E1 has particularly broad substrate specificity it is very much an imperfect enzyme. Due to its mechanism, reactive intermediates can migrate out of its active site2. This tendency is evidenced both by the association of CYP2E1 with hepatotoxicity in humans34  and the blackening of non-vein leaf tissue to transgenic poplars expressing CYP2E1 after exposure to vinyl chloride17. Although a structural basis for this characteristic has not been definitively determined, it has been hypothesized that CYP2E1 has a propensity to uncouple, thus allowing the substrate to migrate away from the active site and leaving an activated oxygen in the active site which can react with other species, leading to cellular damage and apoptosis35.

Our contribution to the field

To date, phytoremediation has been shown to be effective at degrading hazardous materials and to be cost-effective.  It is still not widely used though.  One reason is that plants are slower than other means, and plants can produce hazardous byproducts while degrading pollutants.  CYP2E1 has been introduced to multiple plant species, yet its interactions with the plant homologs of the accessory proteins are yet unknown4,17,18.

In our system, we will address these concerns.  First, because duckweed is a fast-growing aquatic plant, it will be able to quickly and efficiently degrade xenobiotics in a constructed wetland.  Secondly, by tightening interactions between CYP2E1 and it accessory proteins, hazardous byproducts will be reduced.  Finally, this will be the first study to characterize how human CYP2E1 is able to use plant proteins for its proper function.

Overview of Project

Phytoremediation describes the removal of pollutants and toxins from soil and/or groundwater by plants.  A specific type of phytoremediation, rhizofiltration, refers to the removal of contaminants from water by the root systems of plants.  Plants possess a number of mechanisms for metabolizing and sequestering pollutants; however, the rate of contaminant removal using natural plants has been found to be insufficient in many systems36.  The utilization of transgenics to add proteins found in other organisms to plants has resulted in an explosion of potential uses for phytoremediation12.  There are many examples of mammalian proteins responsible for the metabolism of organic chemicals and heavy metals being introduced into plants in the literature.  The plant is then transformed into a bioreactor designed to degrade the harmful substance13.  Herein we propose a novel use of such transgenic technology, in which a protein that degrades xenobiotics will be expressed in an aquatic plant that absorbs toxins through its root system.  Trichloroethylene (TCE) and ethylene dibromide (EDB) are two examples of xenobiotics that are common water and soil pollutants.  They are known to induce numerous harmful effects within the central nervous system37 and to be carcinogenic in mammalian studies38, respectively, and we will focus our initial system on the removal of these two compounds.

We propose using human CYP2E1, a member of the cytochrome p450 (CYP450) family, as our protein of interest.  Human CYP2E1 acts on several small organic substrates, such as trichloroethylene (TCE), ethylene dibromide (EDB), benzene, vinyl chloride, chloroform, styrene, carbon tetrachloride, 1,2-dichloropropane, and several more4,18.  The structure, mechanism, and biochemical properties of CYP2E1 proteins are well known22,23,39,40.  However, although human CYP2E1has been expressed in plants4,17,18, a biochemical characterization of the transgenic human-plant protein interaction has not been performed.  Such information is particularly important, as human CYP2E1 requires the presence of two other proteins, NPR and Cyt b5, for proper function.  Therefore, as our initial aim we propose to characterize the interaction of human CYP2E1 with plant accessory proteins, build an allosteric network, and then improve this interaction via site directed mutagenesis of CYP2E1, resulting in more efficient degradation of TCE and EDB.  We will refer to our generated version of CYP2E1 and cytochrome P450 2E1 enhanced or 2E1en. 

To implement our system, 2E1en must be expressed in plants capable of rhizofiltration, and these plants must be placed in an optimal location where pollutant levels are high. We will use duckweed, a plant which is commonly used at wastewater treatment plants in part because it can survive in highly polluted water3,19,20.  Mammalian CYP2E1 has already been overexpressed in several plant species, including Nicotiana tabacum (tobacco) and Atropa belladonna (deadly nightshade)4,11,17,18.  Duckweed is commonly used in genetic studies because it produces a large amount of protein relative to its biomass, and it overexpresses foreign DNA well19,41.  Lemna gibba and Spirodela punctata are the species of duckweed that best overexpress introduced proteins19, so we will use those species to generate our phytoremediators.  As done in previous genetic studies using duckweed, we will transform the 2E1en gene into Agrobacterium tumefaciens, a bacterium that can randomly insert DNA into the host plant genome.  We will select for plants that added the 2E1en gene to their genome without any adverse affects to the duckweed3.

Once we engineer duckweed plants to overexpress 2E1en, we will need to position them near the pollution source.  To do so we will use constructed wetlands (CWs), which are relatively inexpensive systems that have been used widely to enhance phytoremediation via removal, detoxification, and/or stabilization of pollutants in wastewater42,43.  The structure and composition of these systems is analogous to that of natural wetlands: CWs typically consist of shallow troughs of gravel in which climate- and geographically-appropriate plants are grown44.  Contaminated water is permeated through these systems to facilitate interaction between the plants' root systems and the nutrients, metals, and toxic organics in the water45  The proposed system will utilize a continuous flow, free water surface (FWS) wetland wherein small artificial ponds will be created and populated with transgenically-modified duckweed41.  After validating the efficacy of the proposed biological system through preliminary greenhouse studies, the constructed wetland will be carefully engineered to optimize its functionality and minimize the negative impact on the surrounding ecosystem.

The ultimate goal of this project is to develop an effective pilot study where a cytochrome P450 enzyme is overexpressed within an aquatic plant in order to degrade harmful xenobiotic compounds, such as TCE and EDB, within a constructed wetland system for water purification.  If this pilot succeeds, further experiments could focus on characterizing 2E1en's ability to degrade other known substrates of mammalian CYP2E1.  We hypothesize that ultimately our system will be able to degrade many more harmful xenobiotics than just TCE and EDB.  Whether or not this characterization proves successful, this pilot study, particularly the biochemical characterization, will aid in additional projects designed to express other cytochrome enzymes in order to degrade additional harmful xenobiotic compounds found in the environment.  In the long run in terms of implementation, the aesthetic pleasure of our constructed wetland and the low cost of construction and maintenance should make it particularly marketable. 

Although this system has astounding potential, there are several issues to be considered regarding unexpected impacts that this project could have on the environment.  For example, we do not yet know if the byproducts of 2E1en metabolism will harm the duckweed plants or other organisms within the environment.  However, we will gain information about this aspect of the system through the controlled experiments designed in Aim 2.  It is most likely that if harmful byproducts are produced they will react within the duckweed, potentially killing the organism.  Therefore, over time periodic replacement of the duckweed may be necessary.  To dispose of the contaminated duckweed the plants will be pulverized and pressed using a commercial screw press to extract any harmful chemicals46.  This process should lower the concentration of contaminants so much so that the remaining plant matter should be safe to compost, and the contained chemicals can be disposed of properly.  If it turns out that the harmful byproducts are minimal, the plant matter can merely be autoclaved prior to composting46.

Experimental Design and Methods

Aim 1: Characterization and improvement of the interactions between human CYP2E1 and plant NADPH-dependent P450 reductase and cytochrome b5

The purpose of Aim 1 is to characterize the interaction of human CYP2E1 with plant accessory proteins, build an allosteric network, and then potentially improve this interaction, resulting in more efficient degradation of TCE and EDB. Because this aim has three purposes, they will be covered separately.

Characterizing protein-protein interactions

To begin characterizing the interaction of human CYP2E1 with plant cytochrome b5 (Cyt b5) and NADPH-dependent P450 reductase (NPR), we will use indirect enzyme-linked immunosorbent assays (ELISAs) to measure in vitro binding. Purified human CYP2E1 will be attached to a polystyrene plate, and biotinylated copies of plant NPR and Cyt b5 will be added. Unbound NPR and Cyt b5 will then be washed off the plate. Streptavidin-peroxidase will be used to measure the binding of these proteins. This method has been used for the human version of all three proteins47, showing that the human copies all physically interact. Next, we will investigate in vivo interactions of these proteins using bimolecular fluorescence complementation (BiFC) in A. thaliana. We are using A. thaliana because no species of duckweed has been sequenced, so this experiment will be used as a proof of concept. In BiFC, yellow fluorescent protein (YFP) is divided into two segments, neither of which is fluorescent. Each segment is then attached to a protein of interest. If the two proteins can interact, the two segments of YFP will fuse and fluoresce. This assay has similarities to FRET, but is simpler to perform. It has also been used to characterize interactions between human copies of CYP2E1, Cyt b5, and NPR48. After these experiments, it will be known if human CYP2E1 can bind to the plant homologs of its accessory proteins, both in vitro and in an in vivo system.

In the indirect ELISA assay, protein binding is indicated by fluorescence at 405 nm. If the biotinylated protein binds to the original protein, then it will fluoresce after adding streptavidin-peroxidase and peroxidase substrate. We expect that human CYP2E1 will bind to plant copies of NPR and Cyt b5 in vitro, and then fluorescence will be detected. Fluorescence should also be detected in the BiFC assay. Human CYP2E1 is expected to interact with NPR and Cyt b5 in A. thaliana, so the two segments of YFP will fuse and fluoresce yellow.

A potential pitfall is that while human CYP2E1 does interact with the plant homologs of the accessory proteins, the interaction may no longer be direct and therefore could not be detected by ELISA. The proteins would still presumably be close enough to perform their roles of electron transfer without actually binding to CYP2E1. In that case, the BiFC assay should detect this indirect interaction. Another pitfall is that protein interaction may not occur. In that case, it is clear that human CYP2E1 is co-opting some protein in the Cytochrome P450 machinery because CYP2E1 is functional in plants. In that instance, we will use co-immunoprecipitation and mass spectrometry to determine which proteins bind to human CYP2E1 in plants.

The methods for the ELISA assay are modified from Ozalp 2005. The first step is to obtain purified proteins from human CYP2E1 and A. thaliana Cyt b5 and NPR. To do so, plasmids will be created that contain the gene in question, a six-histidine tag, and that are artificially inducible by IPTG. Escherichia coli will be transformed with this plasmid, and the gene will be induced to form protein. After protein synthesis, the protein is purified using a nickel column. The human CYP2E1 protein will be immobilized to a poly-styrene plate and excess protein will be washed away. A. thaliana Cyt b5 and NPR will then be biotinylated by incubating them each with succinimidyl-6'-(biotinamido)-6-hexanamidohexanoate. Biotinylated Cyt b5 is added to one well of the poly-styrene plate, and biotinylated NPR is added to another. They are allowed to incubate for an hour at room temperature, then any unbound protein is washed away.   To measure protein-protein interaction, a streptavidin-peroxidase solution is added to the proteins, and this solution incubates for an hour. If Cyt b5 and/or NPR bound to the CYP2E1, then the streptavidin will bind to the biotin marker on those proteins. Unbound streptavidin-peroxidase is then washed off using PBS. Hydrogen peroxide is then added. The reaction is terminated using SDS. Protein binding is detected by measuring the plate at 405 nm. If binding occurred, then the sample will fluoresce at that wavelength.

The methods for the BiFC assay are adapted from Ozalp 2005 and Walter 2004. The transformation vectors used contain either the YFP N-terminus (residues 1-155, YN) or YFP C-terminus (156-239, YC). They also contain the constitutively expressed 35S promoter. Human CYP2E1 will be cloned into the YN vector, while either A. thaliana Cyt b5 or NPR will be cloned into the YC vector. These vectors will be transfected into A. thaliana cultured cells49. These cells are in protoplast form, and their cell walls are digested before transfection50. At 12-18 hours post-transfection, the cells will be assayed by epifluorescence microscopy49. The plant cells are excited at 514 nm, and fluorescence is detected at 535-545 nm. If the two proteins interact, then the YFP fragments will recombine, and fluorescence will be detected48.

Determination of the CYP2E1 Allostery through Contact Rearrangement Network Analysis

When an effector binds a protein at an allosteric site, protein function often changes due to changes throughout the three-dimensional protein structure, even at residues far from the interaction site, which is known as an allosteric network51-53. In this study, it will be necessary to determine if conformational allostery exists within human CYP2E1 and if its association with plant NRP and Cyt b5 affects the substrate-binding pocket. Contact Rearrangement Network (CRN) Analysis will be used to measure changes in the contact between pairs of associating protein crystal structures. This method of allosteric network formulation uses direct, model-free analysis of crystal structures54.  CRN does not account for large-scale motions in the quaternary structure, and the degree and closeness of allostery observed using CRN Analysis of the CYP2E1 protein will be between residues within the protein.  CRN does not describe the protein/accessory protein/substrate interaction54. The metrics derived from CRN analysis refer to internal signaling that occurs within the CYP2E1 protein through tertiary structure rearrangement as a result of the proteins interactions with accessory proteins or binding substrate.

A potential pitfall is that CRN Analysis cannot be used to identify allostery within the CYP2E1 protein.  If so, Normal Mode Analysis could be used. This technique is more computational and requires lower-resolution structural information. If there is insufficient CYP2E1 structural data, Statistical Coupling Analysis, which uses evolutionary residue sequence information, could be performed to predict which specific residues are allosterically coupled54.

To construct a CRN, CYP2E1, NPR, and Cyt b5 will be purified. Then, the CYP2E1 will be crystallized in four conformations: with Cyt b5 and NPR (active state, A1), without Cytb5 and NPR (inactive state, I1), with TCE (active state, A2), and without TCE (inactive state, I2).  It may be necessary to complex CYP2E1 with a substrate analog for crystallization (Journal of Biological Chemistry 2004). The crystal structures will be analyzed, focusing on two residues, i and j, that are at least one atom-atom distance (i.e., <5 Å) apart. The rearrangement factor between the crystal structures for all active and inactive states (A1, I1, A2, and I2) will be determined. This calculation takes into consideration the number of atomic contacts between residues i and j as well as the number of atomic contacts shared between i and j. The metrics degree and closeness will be calculated as described by Daily, et al. and used to order residues by functional importance within the CRNs54. The discovery of an allosteric network in CYP2E1 and subsequent ranking of critical residues will provide insight that will facilitate the rational design of the 2E1en engineered protein. We hypothesize that through tightening the interactions between human CYP2E1 and the plant accessory proteins, we can improve the efficiency of this enzyme in the duckweed system. In choosing a mutant of CYP2E1, it will be important to consider the results of the CRN analysis. It will be necessary to select a mutant of CYP2E1 that has enhanced interactions with these accessory proteins without adversely affecting the effectiveness of the substrate binding pocket.

Improving upon the protein-protein interactions

The interaction surfaces between human CYP2E1 and both NPR and Cyt b5 stem mostly from a positively-charged "bowl"-like region on one side of CYP2E1 (Fig. 5d) which is hypothesized to interact with negatively charged residues of both NPR and Cyt b523. We hypothesize that this interaction is most likely suboptimal and that this suboptimal interaction could be particularly detrimental to the success of our transgenic constructed wetland. Even in its natural environment, human CYP2E1 has a propensity to uncouple from Cyt b5 and NPR. Therefore, we propose a method of site directed mutagenesis designed to increase the charge of the positively-charged "bowl"-like region of CYP2E1, thus making it more difficult for the enzyme to uncouple from its redox partners.

To find mutant CYP2E1 proteins that associate more strongly with Cyt b5 and NPR we will perform a technique similar to alanine scanning mutagenesis; however, instead of using the amino acid alanine, we will use lysine because it is positively charged. We will systemically replace any residues surrounding and within the positively charged bowl region that are not positively charged with lysine. It is important to note that we will be able to avoid residues that are allosterically coupled to the active site based on the CRN data. Following the generation of  a number of potential CYP2E1 mutants, we will test their ability to associate with Cyt b5 and NPR via our BiFC assay (described above). Because, theoretically, uncoupling will be minimized, mutants that associate more readily with Cyt b5 and NPR should show more intense YFP fluorescence than we will observe in our baseline system characterized in the first part of Aim 1. Based on the results of the BiFC assays, we will determine which positions when mutated to lysine increase the association of CYP2E1 with its redox partners. We will use this information to generate combinatorial mutants and following increased characterization via the BiFC assay will be able to ascertain the combinatorial mutant that most optimally associates with Cyt b5 and NPR, we will call this mutant cytochrome p450 2E1 enhanced or 2E1en.
Although we will have an allosteric map of CYP2E1 to aid in the avoidance of residues that may have a global effect on the protein fold or the conformation of the active site, it is still possible that mutagenesis will result in unexpected effects on global protein structure. We will address this potential pitfall in Aim 2 with the in vitro characterization of the activity of 2E1en. This problem could also be addressed by performing a CRN analysis for 2E1en. Although time consuming, the knowledge gained from this type of a study would be particularly useful in describing mechanism if 2E1en is determined to be an optimally functioning enzyme in the transgenic plant system. Finally, because we will be performing multiple rounds of mutagenesis it is possible that unrelated mutations will occur in our DNA sequence over time. Therefore, we will be careful to periodically have the entire construct that we are working on sequenced to make sure that no unplanned changes have occurred.

For the site directed mutagenesis of CYP2E1 we will use the QuikChange Site-Directed Mutagenesis Kit (Stratagene). As mentioned in the beginning of Aim 1, we will have plasmids containing the CYP2E1 gene. Our primers for mutagenesis will be designed following the guidelines described by Stratagene and, due to the redundancy of the genetic code, care will be taken to generate mutants by changing the least number of codons as possible. Following thermal cycling of the template and primer DNAs, the reaction mixtures will be Dpn1 digested and amplified. The Dpn-1 treated DNA will be transformed into XL-1 Blue SuperCompetent Cells, which are optimized for the expression of mutant proteins. After antibiotic selection on agar plates, surviving colonies will be sequenced to determine if they contain the desired mutation. For the functional BiFC assay the constructs will be transformed into bacteria that are optimized for protein expression rather than mutagenesis. The BiFC assay and CRN analysis will be performed as described above.

Aim 2: Transformation 2E1en into duckweed and the characterization of enhanced uptake and metabolism of two environmental contaminants by the generated system.

As this aim is two-fold we will discuss both parts separately. The method we will use for the transformation of duckweed with 2E1en is Agrobacterium-mediated transformation. In nature, Agrobacterium can cause tumors, called crown galls, at plant wound sites. This natural ability of the Agrobacterium to introduce the genetic information for the crown gall (tumor) formation into the plant genome is harnessed for transformation. Agrobacterium-mediated transformation is one of the most commonly used methods for transporting new genes into plant cells and for ensuring their stable integration into the genome55. Also, Agrobacterium-mediated transformation is commonly used in duckweed3 and has been used by other groups to transform human CYP2E1 into plants4.
Based on the citations above, we anticipate this form of transformation to be successful in our system. We will verify successful transformation via Northern and Western blots to detect human CYP2E1 mRNA and protein, respectively. The antibodies we use will be from sections of CYP2E1 not modified in making 2E1en.

The most probable pitfall that we may encounter is that the 2E1en expression is not robust enough, so the plants will not optimally degrade xenobiotics. If expression is too low, we will try other methods of transformation including particle bombardment in which tungsten or gold particles are coated with DNA and propelled at high speed into the plant tissue, electroporation in which high voltage is used to migrate the DNA into plant cells, or silicon carbide fibers, in which a mixture of the fibers, DNA, and plant material is vortexed, allowing the DNA to gain access to the inside of the cel56. If after these methods are tried and the expression is still too low, we do recognize that other plant species that perhaps express foreign DNA better could be used in place of duckweed in our constructed wetland.

The methods for this part of Aim 2 have been adapted from Yamamoto et. al., 2001. The plasmid will be constructed by creating a DNA construct with a CMV35S constitutively expressed promoter and human CYP2E1en DNA. The construct will be cloned into a GUS site on the pBI121 vector containing a kanamycin resistance cassette. Electroporation will be used to introduce the plasmid into Agrobacterium and successful clones will be selected by treating the Agrobacterium with kanamycin. To transform the duckweed, we will grow up plasmid containing Agrobacterium and submerge the plant nodule within it, allowing bacteria to enter the plant. To select for positive plants we will stain with GUS, and will generate full plants from those with positive nodules.
Following transformation, in the second half of Aim 2, we will assay for increased uptake and metabolism of trichloroethylene (TCE) and ethylene dibromide (EDB) in duckweed transformed with 2E1en. For a successful constructed wetland, the transformed duckweed must be able to remove TCE and EDB from the water system prior to metabolism. Following uptake, the transformed duckweed must be able to successfully metabolize TCE and EDB and hence the degradation products should be detected at levels above controls. Although further parameters must be considered for the whole system, these very controlled experiments will serve as the first step toward understanding the capabilities of our newly generated system. We will subject the transformed and control duckweed lines to a hydroponic solution supplemented with TCE or EDB in an air-tight environment for approximately one week. Uptake will be easily assayed by measuring the amount of TCE or EDB in the medium before and after the experiment. To explore TCE metabolism we will analyze the plant tissue for amounts of TCE and its metabolites (chloral and TCOH) and for EDB metabolism we will measure the amount of halide ion released into the solution to determine activity of the transformed enzyme.
To date, unmodified CYP2E1 has been introduced into and shown to enhance the xenobiotic degrading ability of tobacco18, hairy root cultures of Atropa belladonna4, and poplar trees17. In each case, the plant's ability to degrade xenobiotics increased significantly, and we expect to observe effects just as good if not better in our enhanced system. It has been described that following one week of growth in a hydroponic solution containing TCE, CYP2E1 transgenic poplar cuttings remove between 51-91% of the TCE in solution, compared to the less than 3% removed by control plants, with the best performing line removing TCE from solution 53 times faster than controls17. Also, CYP2E1 transformed tobacco cuttings removed 35% more EDB supplemented in their growth medium than control cuttings18. TCE metabolism in CYP2E1 transgenic poplars was more than 100-fold larger than controls for some lines, with an average 45-fold enhancement17. In transgenic tobacco, TCE metabolism was elevated 642-fold in roots, 171-fold in stems, and 140-fold in leaves in best expressing line17 and Atropa belladonna hairy root cultures metabolized 5-10 times more TCE than control root cultures4.

Although our enhanced system is designed to not uncouple as easily as the unmodified CYP2E1 we believe that the most likely potential pitfall is toxicity to the plants due to the release of reactive metabolites as other groups in the field have observed this problem. For example, after 7 days of exposure to vinyl chloride, transgenic CYP2E1 poplars showed significant blackening of the non-vein leaf tissue, most likely due to reactive metabolites chloroethylene and 2-chloroacetaldehyde, which bind to DNA17. However, it is highly unlikely that environmental levels of vinyl chloride could ever approach the concentrations used in this controlled experiment17. Therefore, if toxicity is observed the first step will be to determine if our plants could ever be exposed to the concentration of xenobiotic we are using in vitro.

The methods for the second half of Aim 2 have been adapted from Doty et al., 2002 and 2007. Prior to exposure to pollutants plants or plant parts (both transformed with 2E1en and control) will be surface sterilized and grown for 4-6 weeks. The plants will be transferred to sealed flasks with sidearms for addition of chemicals and removal of media for sampling. After equilibration, plants will be dosed with appropriate amount of TCE or EDB. To extract TCE and its byproducts, five days later plants will be harvested, flash frozen, and ground. This extract will be analyzed for amounts of TCE and its metabolites via gas chromatography. To quantify TCE uptake, samples of medium will be collected 15min after TCE exposure and after 5 days. The medium will be concentrated, extracted with hexane and sodium chloride, and analyzed via gas chromatography to quantify the amount of TCE. The quantification of EDB uptake will be performed in the same manner as TCE uptake quantification only EDB will be assayed for. To determine EDB metabolism one assays for the bromide ion, to do so samples of hydroponic growth medium will be collected, filtered, and run on a ion chromatograph to determine bromide ion concentration.

Aim 3: Analysis of the performance of 2E1en-transformed duckweed in constructed wetlands for TCE and EDB phytoremediation

A constructed wetland (CW) containing 2E1en-transformed duckweed will be used to facilitate phytoremediation of TCE and EDB. To assess the feasibility of such a system, a well-controlled greenhouse CW, whose design will be adapted from Ran et al., 2004 and Cheng et al., 2002, will first be explored. This CW will initially be fed with mineral nutrient-enriched water for fertilization45. The chemistry of the system will encourage duckweed growth and allow for the optimization of the CW physical parameters including hydraulic retention time, hydraulic loading rate, and organic loading rate44. Next, 2E1en-transformed duckweed will be grown in artificial wastewater dosed with the xenobiotics of interest45. This experimental setup will allow for the analysis of duckweed growth in ecologically-relevant conditions, as well as 'in situ' uptake, metabolism, and degradation products of the xenobiotics TCE and EDB. Finally, a constructed wetland will be built at a research site. Here, CW water will be supplemented with industrial effluent, local contaminated groundwater, or wastewater secondary effluent. Important considerations at this stage include investigation of environmental effects both on and of the CW and cost/benefit analysis of system implementation.

Duckweed has been used effectively for phytoremediation in constructed wetlands because of its high growth rate, metabolic activity, and resistance to contaminated water44.  These plants, like all helophytes, do not significantly affect a constructed wetland's hydraulic parameters, which are critical for effective phytoremediation to occur41. The properties of duckweed-populated constructed wetland allow for CO2 and O2exchange directly with the atmosphere with no need for external supply27,44. It is expected that duckweed growth in contaminated constructed wetlands will be only marginally reduced since helophytes in aquaculture CW systems generally have better resistance to the extreme rhizosphere conditions including acidic or alkaline pH, salinity, toxic components such as phenols, tensides, biocides, and heavy metals43. In fact, duckweed grown in wastewater effluent was found to double in frond numbers, and therefore in area covered, every four days. Therefore, greenhouse CW experiments should proceed for approximately one month to allow for sufficient analysis41.

As in any phytoremediation system, the effects of the constructed wetland on the local ecosystem must be considered. Fortunately, CWs have limited negative environmental effects as they are designed to emulate natural wetlands44. TCE and EDB metabolite-laden duckweed plants will be harvested periodically and, depending on the nature of the degradation products, these plants may be able to be used as a source of biomass. While harvesting may be required once a week in warmer climates44, duckweed is relatively cold tolerant and can be grown practically at temperatures as low as 7°C57. Duckweed has very little structural fiber and is therefore, essentially all metabolically active cells, so metabolism rates should be high58.

It is possible that high levels of TCE, EDB, or their metabolites could adversely affect the growth of duckweed. If this is the case, it will be necessary to tune the physical CW parameters to lower hydraulic and/or organic loading rate. Additionally, the rate of TCE/EDB metabolism may be too low to be effective. Again, physical parameters of the CW could be changed (lower flow rates, more surface area, etc) to increase hydraulic retention times59.

The methods used in this aim have been adapted from Cheng, S et al., 2002. CW populations of 2E1en-transformed duckweed will be supplemented with mineral nutrient-enriched water every 8 hours - the duration of this phase of conditioning depends upon the rate of duckweed growth. CW populations of 2E1en-transformed duckweed will then be supplemented with artificial wastewater and ecologically-relevant concentrations of TCE and EDB will be added upstream of the CW. Similarly, this length of this phase will depend upon duckweed growth and efficiency of uptake and metabolism of xenobiotics, as determined through weekly-harvested duckweed samples. The methods used to evaluate the research site CW were adapted from the US Environmental Protection Agency guidelines for constructed wetlands. Briefly, water flowing through CWs containing 2E1en-transformed duckweed will be supplemented with industrial effluent, local contaminated groundwater, or wastewater secondary effluent. Weekly harvesting and analysis will be used to determine potential toxicity of TCE, EDB, and byproducts on duckweed as discussed in Aim 2. Finally, TCE and EDB uptake will be quantified in both greenhouse and research site CWs.


With this project, we hope to develop a pilot study in which TCE and EDB are degraded in an inexpensive and sustainable water-filtration system.  Further work on our project will hopefully develop into systems for degrading other xenobiotic materials present in industrial, ground, and waste water42.  This project will contribute to saving the world by eliminating TCE from freshwater systems.  TCE is a common industrial solvent and contaminant of hazardous waste sites, groundwater, and drinking water.  It has been identified as a central nervous system depressant, a hepatotoxin, and a potential carcinogen7.  One striking example of the dangers of TCE was seen at Camp Lejeune, an air force base, between the 1960s and 1980s.  Breast cancer was found in 20-40 servicemen and their sons after extended TCE exposure. 

TCE is still found in ground and well water above EPA mandated levels in the United States.  Additionally, industrialization is increasing worldwide and this problem will only continue to manifest into further problems if not addressed early enough.  Increased TCE levels will provide a sustainability challenge throughout the world if this toxin is not safely degraded.

Our system efficiently and cheaply degrades TCE, and the development of systems that degrade other xenobiotics will support the survival of humans and all other species on Earth.  Pure, nontoxic drinking water is required for human health and development, as well as for the survival of wildlife.


1.Ioannides, C. & Lewis, D.F.V. Cytochromes P450 in the bioactivation of chemicals. Curr Top Med Chem 4, 1767-1788(2004).

2.Anders, M.W. & Jakobson, I. Biotransformation of halogenated solvents. Scand J Work Environ Health 11 Suppl 1, 23-32(1985).

3.Yamamoto, Y. et al. Genetic transformation of duckweed Lemna gibba and Lemna minor. In Vitro Cellular & Developmental Biology - Plant 37, 349-353(2001).

4.Banerjee, S. et al. Expression of functional mammalian P450 2E1 in hairy root cultures. Biotechnol. Bioeng 77, 462-466(2002).

5.Rowe, B.L. et al. Occurrence and potential human-health relevance of volatile organic compounds in drinking water from domestic wells in the United States. Environ. Health Perspect 115, 1539-1546(2007).

6.U.S. E.P.A. Water on tap: What you need to know. (Office of Water, United States Environmental Protection Agency: Washington, D. C., 2003).at <<http://www.epa.gov/safewater/wot/pdfs/book_waterontap_full.pdf>

7.US DHHS Trichloroethylene Toxicity. (2007).at <http://www.atsdr.cdc.gov/csem/tce/>

8.Xu, H. et al. Exposure to trichloroethylene and its metabolites causes impairment of sperm fertilizing ability in mice. Toxicol. Sci 82, 590-597(2004).

9.Jollow, D.J. et al. Trichloroethylene risk assessment: a review and commentary. Crit. Rev. Toxicol 39, 782-797(2009).

10.U.S. E.P.A. Basic information about ethylene dibromide in drinking water. (2009).at <<http://www.epa.gov/safewater/contaminants/basicinformation/ethylene-dibromide.html>>

11.Doty, S.L. Enhancing phytoremediation through the use of transgenics and endophytes. New Phytol 179, 318-333(2008).

12.Dowling, D.N. & Doty, S.L. Improving phytoremediation through biotechnology. Curr. Opin. Biotechnol 20, 204-206(2009).

13.Abhilash, P.C., Jamil, S. & Singh, N. Transgenic plants for enhanced biodegradation and phytoremediation of organic xenobiotics. Biotechnol. Adv 27, 474-488(2009).

14.Noctor, G. et al. Synthesis of Glutathione in Leaves of Transgenic Poplar Overexpressing [gamma]-Glutamylcysteine Synthetase. Plant Physiol 112, 1071-1078(1996).

15.Karavangeli, M. et al. Development of transgenic tobacco plants overexpressing maize glutathione S-transferase I for chloroacetanilide herbicides phytoremediation. Biomol. Eng 22, 121-128(2005).

16.Flocco, C.G., Lindblom, S.D. & Smits, E.A.H.P. Overexpression of enzymes involved in glutathione synthesis enhances tolerance to organic pollutants in Brassica juncea. Int J Phytoremediation 6, 289-304(2004).

17.Doty, S.L. et al. Enhanced phytoremediation of volatile environmental pollutants with transgenic trees. Proc. Natl. Acad. Sci. U.S.A 104, 16816-16821(2007).

18.Doty, S.L. et al. Enhanced metabolism of halogenated hydrocarbons in transgenic plants containing mammalian cytochrome P450 2E1. Proc. Natl. Acad. Sci. U.S.A 97, 6287-6291(2000).

19.Stomp, A. The duckweeds: a valuable plant for biomanufacturing. Biotechnol Annu Rev 11, 69-99(2005).

20.Zirschky, J. & Reed, S.C. The Use of Duckweed for Wastewater Treatment. Journal (Water Pollution Control Federation) 60, 1253-1258(1988).

21.Nelson, D.L., Lehninger, A.L. & Cox, M.M. Lehninger principles of biochemistry. (W.H. Freeman: New York, 2008).

22.Ortiz de Montellano, P. Cytochrome P450 : structure, mechanism, and biochemistry. (Plenum Press: New York, 2005).

23.Porubsky, P.R., Meneely, K.M. & Scott, E.E. Structures of human cytochrome P-450 2E1. Insights into the binding of inhibitors and both small molecular weight and fatty acid substrates. J. Biol. Chem 283, 33698-33707(2008).

24.Omata, Y. et al. Specificity of the cytochrome P-450 interaction with cytochrome b5. FEBS Lett 346, 241-245(1994).

25.Voznesensky, A.I. & Schenkman, J.B. Quantitative analyses of electrostatic interactions between NADPH-cytochrome P450 reductase and cytochrome P450 enzymes. J. Biol. Chem 269, 15724-15731(1994).

26.Dekant, W., Koob, M. & Henschler, D. Metabolism of trichloroethene--in vivo and in vitro evidence for activation by glutathione conjugation. Chem. Biol. Interact 73, 89-101(1990).

27.Green, M., Friedler, E. & Safrai, I. Enhancing nitrification in vertical flow constructed wetland utilizing a passive air pump. Water Research 32, 3513-3520(1998).

28.James, C.A. et al. Degradation of low molecular weight volatile organic compounds by plants genetically modified with mammalian cytochrome P450 2E1. Environ. Sci. Technol 42, 289-293(2008).

29.van Bladeren, P.J. et al. The influence of disulfiram and other inhibitors of oxidative metabolism on the formation of 2-hydroxyethyl-mercapturic acid from 1,2-dibromoethane by the rat. Biochem. Pharmacol 30, 2983-2987(1981).

30.Gao, Q. et al. Identification of the interactions between cytochrome P450 2E1 and cytochrome b5 by mass spectrometry and site-directed mutagenesis. J. Biol. Chem 281, 20404-20417(2006).

31.Lesot, A.Plant Physiology and Biochemistry 33, 751-757(1995).

32.NCBI BLAST: Basic Local Alignment Search Tool. National Center for Biotechnology Information at <http://blast.ncbi.nlm.nih.gov/Blast.cgi>

33.NCBI Conserved Domain Search. National Center for Biotechnology Information at <http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi>

34.Smith, S.V. et al. Rapid conformational dynamics of cytochrome P450 2E1 in a natural biological membrane environment. Biochemistry 45, 15617-15623(2006).

35.Porter, T.D. The roles of cytochrome b5 in cytochrome P450 reactions. J. Biochem. Mol. Toxicol 16, 311-316(2002).

36.Goel, A. et al. Plant cell biodegradation of a xenobiotic nitrate ester, nitroglycerin. Nat. Biotechnol 15, 174-177(1997).

37.Costa, A.K., Katz, I.D. & Ivanetich, K.M. Trichloroethylene: its interaction with hepatic microsomal cytochrome P-450 in vitro. Biochem. Pharmacol 29, 433-439(1980).

38.Ploemen, J.P. et al. The use of human in vitro metabolic parameters to explore the risk assessment of hazardous compounds: the case of ethylene dibromide. Toxicol. Appl. Pharmacol 143, 56-69(1997).

39.Boysen, G. et al. Identification of covalent modifications in P450 2E1 by 1,2-epoxy-3-butene in vitro. Chem. Biol. Interact 166, 170-175(2007).

40.Rendic, S. Summary of information on human CYP enzymes: human P450 metabolism data. Drug Metab. Rev 34, 83-448(2002).

41.Ran, N., Agami, M. & Oron, G. A pilot study of constructed wetlands using duckweed (Lemna gibba L.) for treatment of domestic primary effluent in Israel. Water Res 38, 2240-2247(2004).

42.Cheng, S. et al. Xenobiotics removal from polluted water by a multifunctional constructed wetland. Chemosphere 48, 415-418(2002).

43.Stottmeister, U. et al. Effects of plants and microorganisms in constructed wetlands for wastewater treatment. Biotechnol. Adv 22, 93-117(2003).

44.U.S. E.P.A. Constructed wetlands treatment of municipal wastewaters. (United States Environmental Protection Agency: 2000).

45.Cheng, S. et al. Efficiency of constructed wetlands in decontamination of water polluted by heavy metals. Ecological Engineering 18, 317-325(2002).

46.Elless , M. Transgenic Citrate-Producing Plants for Lead Phytoremediation. (2003).

47.Shimada, T., Mernaugh, R.L. & Guengerich, F.P. Interactions of mammalian cytochrome P450, NADPH-cytochrome P450 reductase, and cytochrome b(5) enzymes. Arch. Biochem. Biophys 435, 207-216(2005).

48.Ozalp, C., Szczesna-Skorupa, E. & Kemper, B. Bimolecular fluorescence complementation analysis of cytochrome p450 2c2, 2e1, and NADPH-cytochrome p450 reductase molecular interactions in living cells. Drug Metab. Dispos 33, 1382-1390(2005).

49.Walter, M. et al. Visualization of protein interactions in living plant cells using bimolecular fluorescence complementation. Plant J 40, 428-438(2004).

50.Merkle, T. et al. A plant in vitro system for the nuclear import of proteins. Plant J 10, 1177-1186(1996).

51.Holt, J.M. & Ackers, G.K. The pathway of allosteric control as revealed by hemoglobin intermediate states. FASEB J 9, 210-218(1995).

52.Lee, J. et al. Surface sites for engineering allosteric control in proteins. Science 322, 438-442(2008).

53.Lockless, S.W. & Ranganathan, R. Evolutionarily conserved pathways of energetic connectivity in protein families. Science 286, 295-299(1999).

54.Daily, M.D., Upadhyaya, T.J. & Gray, J.J. Contact rearrangements form coupled networks from local motions in allosteric proteins. Proteins 71, 455-466(2008).

55.Tzfira, T. & Citovsky, V. Agrobacterium : from biology to biotechnology. (Springer: New York ;;London, 2008).

56.Slater, A., Scott, N.W. & Fowler, M.R. Plant biotechnology : the genetic manipulation of plants. (Oxford University Press: Oxford ;;New York, 2003).

57.Leslie, M. Water hyacinth wastewater. (United States Environmental Protection Agency: 1983).

58.Reed, S.C., Middlebrooks, E.J. & Crites, R.W. Natural systems for waste management and treatment. (McGraw-Hill: New York, 1988).

59.Platzer, C. Design recommendations for subsurface flow constructed wetlands for nitrification and denitrification. Water Science and Technology 40, 257-263(1999).