GENETIC ENGINEERING TO ENHANCE MERCURY PHYTOREMEDIATION

Summary

Most phytoremediation studies utilize merA or merB genes to modify plants via the nuclear or chloroplast genome, expressing organomercurial lyase and/or mercuric ion reductase in the cytoplasm, endoplasmic reticulum or within plastids. Several plant species including Arabidopsis, tobacco, poplar, rice, Eastern cottonwood, peanut, salt marsh grass and Chlorella have been transformed with these genes. Transgenic plants grew exceedingly well in soil contaminated with organic (~400 μM PMA) or inorganic mercury (~500 μM HgCl₂), accumulating Hg in roots surpassing the concentration in soil (~2000 μg/g). However, none of these plants were tested in the field to demonstrate real potential of this approach. Availability of metal transporters, translocators, chelators and the ability to express membrane proteins could further enhance mercury phytoremediation capabilities.

Phytoremediation is the use of plants to clean up contaminated environments. It is a cost-effective, environmentally friendly approach with great advantages for large-scale clean up of contaminated sites [1, 2, 3]. Current remediation methods for heavy metal contamination are environmentally invasive, expensive and inefficient, especially for large scale clean up [4]. However, plants have limited capabilities for treating high levels of heavy metal contamination.

Mercury (Hg), one of the most toxic pollutants threatening our health and ecosystems, has been introduced from natural and anthropogenic sources. Recent estimates calculate the annual global emissions between 4,800–8,300 tons per year [5]. In the United States alone coal burning power plants emit about 48 tons of mercury annually [6]; the combined estimates for Asia and Africa surpass the 1,500 tons [7*]. The cost of remediation of each pound of mercury from the environment using current technologies is in the range of tens of thousands of dollars. Therefore, finding alternate remediation approaches is an urgent need.

Mercury is usually released in the metal or ionic form, accumulating in sediments where it becomes methylated by anaerobic sulfate-reducing bacteria to produce methyl mercury, a highly toxic organomercurial compound. The toxicity of mercury greatly depends on the form present in the environment. Inorganic mercury forms are usually less harmful than organic forms, in part because they bind strongly to soil components that reduce their availability and absorption. On the other hand, organomercurials are highly toxic because of their hydrophobicity, which facilitates their movement across cell membranes and accumulation in membrane bound organelles, inhibiting essential oxidative and photosynthetic pathways. The toxic effects of methyl mercury are further magnified at higher trophic levels due to increased accumulation in tissues [8,9]. Organomercurials are potent neurotoxins with over 90% absorption into the blood stream from the intestinal track, while mercury salts and elemental mercury show reduced absorptions of less than 10% and 0.1%, respectively [10]. In plants, ionic mercury tends to affect the plasma membrane where it damages membrane transporters such as aquaporins, leading to nutrient and water disruption [11]. Organomercurials rapidly localize to plastids where they accumulate and disrupt important metabolic functions including electron transport, oxygen evolution [12], photophosphorylation, Hill reaction, chlorophyll content and chlorophyll fluorescence [13,14].

Plants cannot successfully detoxify or interconvert mercury to more benign forms. Genetic engineering can integrate genes from other organisms to enhance phytoremediation capabilities in plants. This article reviews our current knowledge of transgenic plant systems for Hg phytoremediation, provides insight on mechanistic aspects and directions for further studies.

MerAB Engineered via the Nuclear Genome

A well characterized phytoremediation system is the use of the bacterial merA (mercuric ion reductase) and merB (organomercurial lyase) genes to genetically engineer plants for the remediation of Hg [15–26]. Organomercurial lyase facilitates protonolysis of organic-Hg to Hg²⁺ while mercuric ion reductase reduces Hg²⁺ to Hg⁰, which is volatilized from plants. Initial attempts to use the native bacterial merA gene in nuclear transgenic plants failed to express the mercuric ion reductase [24*,25]. When the coding sequence was modified to plant preferred codons, transgenic Arabidopsis thaliana, yellow poplar, peanut, and Eastern cottonwood plants were resistant to levels up to 25 –100 μM HgCl₂ [24*–26]. Kim et al, [22*] demonstrated that root-specific expression of the merA gene in Arabidopsis thaliana provided resistance up to 80 μM HgCl₂ (Table 1). Heaton et al, [27] showed that the root of merA transgenic plants grew directly into Hg “hotspots” while the wild type grew away from the source. These results indicated that root protection was important for phytoremediation and that root is the main organ affected by Hg. One drawback of this finding was the implication that Hg was being detoxified to Hg⁰ in the root where it may become volatilized, though Hg⁰ could also be transported through the xylem to the leaves where it is transpired.

Table 1.

Advances in transgenic systems for phytoremediation of mercury

Species	Gene	Expression Compartment	Level of Resistance	Mode of Resistance	Special Future	Year	Ref.
A. thaliana	merA	Cytoplasm	100 μM HgCl2	Hg(II) to Hg(0)	First report of merA in plants	1996	25
Yellow Poplar (L. tulipifera)	merA	Cytoplasm	50 μM HgCl2	Hg(II) to Hg(0)	merA transformed into a tree species	1998	24*
A. thaliana	merB	Cytoplasm	2 μM PMA	PMA to Hg(II)	First report of merB in plants	1999	17
A. thaliana	merA/B	Cytoplasm	5–10 μM organic- Hg	PMA to Hg(II) to Hg(0)	First report of merA and merB in plants	2000	16**
Tobacco (N. tabacum)	merA	Cytoplasm	50 μM HgCl2	Hg(II) to Hg(0)	merA transformed into tobacco	2001	21
Tobacco	merA/B	Plastid	400 μM PMA	PMA to Hg(II) to Hg(0)	First chloroplast phytoremediation system	2003	29**
Tobacco	merB	Endoplasmic reticulum (ER)	5 μM PMA	PMA to Hg(II)	MerB targeted to the ER	2003	15
Rice (O. sativa)	merA	Cytoplasm	250 μM HgCl2	Hg(II) to Hg(0)	First monocot for Hg phytoremediation	2003	20
Eastern cottonwood (P. deltoides)	merA	Cytoplasm	25 μM HgCl2	Hg(II) to Hg(0)	merA transformed into a forest tree	2003	18
Peanut (A. hypogaea)	merA	Cytoplasm	100 μM HgCl2	Hg(II) to Hg(0)	merA transformed into peanut	2003	26
A. thaliana	merA	Cytoplasm of root cells	80 μM HgCl2	Hg(II) to Hg(0)	Root-specific expression of merA	2005	22*
Salt marsh cordgrass (S. alterniflora)	merA	Cytoplasm	500 μM HgCl2	Hg(II) to Hg(0)	First wetland grass for Hg phytoremediation	2006	19*
Chlorella	merA	Cytoplasm	40 μM HgCl2	Hg(II) to Hg(0)	First transgenic alga for Hg bioremediation	2006	7*
A. thaliana	merC	Cell membrane	10 μM HgCl2	Hypersensitivity to Hg(II)	First use of Hg membrane transporter	2006	48**
Tobacco	ppk	Cytoplasm	10 μM HgCl2	Hg chelation	First use of a Hg-scavenging agent	2006	53**
Eastern cottonwood	merA/B	Cytoplasm	10 μM PMA	PMA to Hg(II) to Hg(0)	merA and merB transformed into a forest tree	2007	23
Tobacco	merA/B	Plastid	300 μM PMA/HgCl2	PMA to Hg(II) to Hg(0)	100-fold increase in Hg translocation and accumulation	2007	44**
A. thaliana	merP	Cell membrane	10 μM HgCl2	Possibly Hg(II) chelation	Localization to the cell membrane	2009	49**

PMA: phenylmercuric acetate; Organic-Hg: PMA and methyl mercury; MerA/B: expression of merA and merB genes; References in bold: relevant;

^*special interest;

^**outstanding interest

Heaton et al, [20] developed merA transgenic rice (Oryza sativa) resistant to 250 μM HgCl₂; the wild type was resistant up to 150 μM HgCl₂. This was the first report of a transgenic monocot for phytoremediation of Hg. Also, salt marsh cordgrass (Spartina alterniflora) modified for Hg phytoremediation was resistant up to 500 μM HgCl₂ [19*]. It should be pointed out that both these monocots showed natural tolerance to Hg and further studies are needed to understand the mechanism of such resistance. Recently, Huang et al, [7*] demonstrated that transgenic Chlorella expressing the merA gene withstood up to 40 μM HgCl_2. Microalgae have certain advantages over land plants for phytoremediation, including larger surface area in contact with the contaminant in water, and a natural capacity to bind Hg [28].

Unfortunately, mercuric ion reductase did not protect against the more toxic and environmentally relevant organic-Hg. Both the merA and merB genes are needed to protect cells from organic-Hg. When a codon optimized merB was used to transform the Arabidopsis thaliana nuclear genome, it provided resistance up to 1 μM phenylmercuric acetate (PMA) [17]; although this is a very low level of resistance, this concentration was lethal to untransformed plants. The organic-Hg concentration needed to inhibit growth in untransformed plants was 100-fold less than the level required for HgCl₂, confirming higher toxicity. Transgenic Eastern cottonwood trees expressing merB failed to confer resistance to PMA [18].

In 2000, Bizily et al, [16**] reported the first transgenic plant expressing both the merA and merB genes; Arabidopsis thaliana transgenic lines showed resistance up to 5 μM PMA and 10 μM CH₃Hg, a 5-fold increase in resistance when compared to transgenic plants expressing only the merB genes. Transgenic Eastern cottonwood trees expressing both genes were tolerant up to 10 μM of PMA [23]. Bizily et al, [15] showed that transgenic plants in which organomercurial lyase was localized to the endoplasmic reticulum (ER) and to the cell wall (CW) conferred as much resistance (up to 5 μM PMA) as transgenic plants expressing both genes in the cytoplasm. Organomercurial lyase localized to the ER showed 41-times as much activity as organomercurial lyase in the cytoplasm, suggesting that organic-Hg may be localized to membrane bound organelles (Table 1).

MerAB Engineered via the Chloroplast Genome

In order to provide higher levels of resistance (> 5–10 μM PMA), the plastids may need protection from toxic mercury. There are different types of plastids present in plant organs and tissues. Chloroplasts present in green tissues are metabolically very active and are involved in photosynthesis, sulfur/nitrogen metabolism and synthesis of starch, amino acids, fatty acids, pigments, vitamins, etc. Chloroplasts are the primary source of world’s food and oxygen, sustaining life on this planet. Other types of plastids include colorful chromoplasts present in flowers, fruits and vegetables. Amyloplasts are present in tubers like potato or in endosperm tissues in seeds, primarily performing the function of storage. Non-green proplastids are present in roots and developing tissues. In spite of variation in their appearance and functions, all plastids contain identical genome structure and sequence.

The idea of protecting essential metabolic reactions occurring within plastids, has been our working hypothesis for expressing merA and merB genes within plant chloroplasts [29**]. Chloroplast has been shown to be the main target for Hg poisoning [12, 30, 31]. Transgene containment by the maternal inheritance of the plastid genome [32] or cytoplasmic male sterility [33], is another major advantage of the chloroplast transformation system. Chloroplast genetic engineering is especially advantageous for the development of transgenic plants when multiple genes are required for effective phytoremediation [34]. By using chloroplast transformation, multigenetic pathways can be developed in a single transformation event without the need for backcrosses of independent lines or re-transformation. Because plastids have retained the genetic machinery from bacterial ancestor, their genomes can transcribe and translate operons [34]. Chloroplast genetic engineering also circumvents common drawbacks of nuclear transformation, including gene silencing and position effect. Site-specific transgene integration by homologous recombination [35] eliminates the position effect. High expression levels, up to 47% of total leaf protein in healthy plants [36**] or accumulation of transcripts 150 fold higher than nuclear transgenic plants [37] did not result in post-transcriptional or post-translational gene silencing. Such high expression levels are achieved because there are up to 10,000 copies of transgenes in each transformed plant cell. Also, the addition of transcriptional and translational enhancer elements is needed to optimize expression of transgenes. However, there is no need for codon optimization of bacterial genes for their expression in plastids [29**].

In spite of these advantages, there are a few limitations of the chloroplast transformation technology. Unlike transformation of the nuclear genome of different crop species using the same transformation vectors, species specific vectors are necessary for chloroplast transformation. Therefore, it is important to have the chloroplast genome sequence, especially the intergenic spacer regions and endogenous regulatory sequences (promoter, 5′UTR, 3′UTR) to create species specific chloroplast vectors. Until 2004, complete genome sequences were available for only six crop species. However, plastid genome sequences are becoming available at a rapid pace, facilitating the application of chloroplast genetic engineering to more species than ever before [38–42]. The second challenge is achieving homoplasmy (integration of transgenes into each chloroplast genome and elimination of untransformed chloroplast genome). This has been easily accomplished in plant species that are regenerated via organogenesis, after several rounds of selection. Chloroplast genomes of more than ten crop species have been transformed via organogenesis, including cauliflower, lettuce, cabbage, petunia, poplar, tobacco, potato, tomato, etc [43]. However, achieving homoplasmy in crops regenerated via somatic embryogenesis is challenging although carrot, cotton, and soybean chloroplast genomes have been transformed already, with desired agronomic traits [43]. The versatility of the chloroplast system has led to its application in developing transgenic plants with enhanced agronomic traits [43]. Therefore, chloroplast genetic engineering is an ideal approach for mercury phytoremediation [29**].

In these studies, the native bacterial merA and merB genes were integrated into the tobacco chloroplast genome [29**,44**]. The transgenic plants were resistant to very high concentrations of PMA, up to 400 μM. This is the only report to date in which transgenic plants not only survived such high concentrations of PMA but also showed better growth than control untreated plants [29**]. Transgenic plants grew exceedingly well in soil contaminated with 300 μM PMA or HgCl_2, accumulating Hg in the root to levels surpassing the concentration in soil, up to 2000 μg/g [44**]. There are several reasons for successful phytoremediation via the chloroplast genome. Mercury, especially in the organic form, is targeted to the chloroplast. Mercuric ion reductase functions better in chloroplasts because of the abundance of NADPH. Protection of the root in the transgenic plant was possible because root proplastids are active and express the mer genes under the control of a constitutive promoter. Kumar et al, [45] showed that root proplastids expressed up to 70% of foreign protein as leaf chloroplasts. Interestingly, transgenic plants were able to accumulate 100-fold and 4-fold more Hg in the shoot in the presence of PMA or HgCl₂ than untransformed plants. This shows that organic-Hg is transported more efficiently than inorganic forms. Finally, the transgenic plants were shown to volatilize Hg⁰ efficiently, independent of the form of Hg in the soil, confirming that both mercuric ion reductase and organomercurial lyase were active in transgenic chloroplasts [44**].

Other Mechanisms for Phytoremediation of Mercury

The mer operon has three known membrane transporter genes involved in the process of translocating Hg²⁺ into the cell: merC, merP, and merT [46,47]. Recently, Sasaki et al, [48**] showed that expression of the merC gene in Arabidopsis thaliana enhanced hypersensitivity to otherwise sub-lethal concentrations of HgCl₂. Therefore, plants can be genetically modified with bacterial metal transporters to enhance heavy metal uptake. In a subsequent report, Arabidopsis thaliana transformed with the merP gene showed resistance to 10 μM HgCl₂ instead of becoming hyper-sensitive [49**]. Because merP requires merT for Hg²⁺ translocation, it is possible that merP was trapping Hg at the cell membrane, thereby protecting the cytoplasm. To take advantage of these transport mechanisms, transgenic plants should be modified with the merA gene as well; this should provide resistance against higher levels of transported Hg. Alternatively, the merC can be coupled to a chelator gene like polyphosphate kinase (ppk) or metallothionein (mt), to develop transgenic plants that could accumulate Hg. Recently, it has been shown that by using chloroplast genetic engineering, membrane proteins can be targeted to the chloroplast inner envelope membrane and accumulated at very high levels [50**]. This opens the possibility of transforming the chloroplast genome with mer transporters to enhance Hg accumulation or phytoremediation.

One public concern regarding the use of the merAB system is the release of Hg⁰ into the atmosphere. Therefore, an alternative approach would be the sequestration of ionic mercury inside the cell by binding it to a chelating molecule. The bacterial ppk gene coding for polyphosphate kinase synthesized polyphosphates, which are negatively charged; the long phosphate polymers reduced cytotoxicity of heavy metals by chelation [51]. Evidence for this was first provided when nuclear transgenic tobacco plants expressing the ppk gene showed enhanced tolerance and accumulation of Hg²⁺ when grown in 10 μM HgCl₂ [52,53**]. Although the levels of resistance were low, the possibility of using metal-scavenging molecules for phytoremediation was demonstrated.

Future Prospects for Mercury Phytoremediation

It is clear that plants can be genetically modified to enhance their tolerance, uptake, translocation, accumulation and volatilization capabilities for Hg phytoremediation. Several general conclusions can be drawn from the discussed reports. Transgenic plants grow exceedingly well in soil contaminated with PMA or HgCl_2, accumulating Hg in the root to levels surpassing the concentration in soil, up to 2000 μg/g or volatilizing mercury supplied in organic or inorganic form. However, none of these plants were tested in the field to demonstrate real potential of this approach. Availability of metal transporters, translocators, and chelators could further enhance phytoremediation capabilities. Our recent ability to express membrane proteins opens the possibility of transforming the chloroplast genome with mer transporters to enhance Hg accumulation or phytoremediation. Alternatively, the merC can be coupled to a chelator gene like polyphosphate kinase (ppk) or metallothionein (mt), to develop transgenic plants that could accumulate Hg.

Plants tend to accumulate most mercury in roots, and translocation to shoot tends to be a major limitation, dependent upon concentration and cellular integrity [20, 27, 44**]. It seems that for translocation to occur, saturation of the roots to levels only withstood by transgenic plants is required. A recent report of a hundred fold increase in Hg translocation to the shoot [44**] is quite encouraging. However, in order to utilize the large biomass above ground, the translocation problem should be addressed further. Other interesting observations include promotion of growth in mer transgenic lines compared to untreated controls, and increased chlorophyll content when grown in the presence of mercury [25,29**,54].

Conclusions

Toxicity of methyl mercury is magnified due to its high accumulation in tissues; over 90% of methyl mercury is absorbed into the blood stream from the gastrointestinal tract when compared to 0.1% of elemental mercury [10]. Methyl mercury accumulated in fish is therefore 990-fold more neuro-toxic than elemental mercury in thermometers. US EPA cautions pregnant women against frequent consumption of fish. Therefore, it is very important to develop technologies to clean up our environment. The potential of transgenic plants for Hg phytoremediation has been demonstrated, but important questions need to be answered to explore the full potential of this technology. A better understanding of Hg-forms transported and accumulated, and their saturation and steady-state levels in different tissues would help develop phytoremediation systems that are cost-effective. New mercury phytoremediation technologies would rely on different gene combinations to enhance uptake, translocation, chelation or detoxification and release of Hg⁰ into the atmosphere (Figure 1). Ultimately, the scientific community should make an effort to address the most important questions that deter the application of phytoremediation and produce plants with the best capabilities, including plants with large biomass, rapid growth rate, adaptation to wider climatic conditions, and expression of multiple genes in different cellular compartments. The use of edible plant species should be avoided in phytoremediation applications, while plant species that out-cross with wild relatives or cultivated plants may be used only when containment measures are applied including harvesting transgenic plants prior to the plant reproductive age, using a male sterility or maternal inheritance system, or planting away from wild relatives to minimize outcross.

Figure 1. Schematic representation of a genetically modified plant cell for optimal mercury phytoremediation.

Mercury phytoremediation technologies should utilize multiple genes to enhance uptake, translocation and detoxification in different cellular compartments, including plastids and endoplasmic reticulum. Both chelation and Hg⁰ volatilization are promising approaches for phytoremediation. Solid dark arrows indicate rapid movement across membranes or via specialized Hg transporters (merC) while dashed arrows indicate slow movement mediated by cellular ion transporters.