Subcellular targeting of methylmercury lyase enhances its specific activity for organic mercury detoxification in plants

Abstract

Methylmercury is an environmental pollutant that biomagnifies in the aquatic food chain with severe consequences for humans and other animals. In an effort to remove this toxin in situ, we have been engineering plants that express the bacterial mercury resistance enzymes organomercurial lyase MerB and mercuric ion reductase MerA. In vivo kinetics experiments suggest that the diffusion of hydrophobic organic mercury to MerB limits the rate of the coupled reaction with MerA (Bizily et al., 2000). To optimize reaction kinetics for organic mercury compounds, the merB gene was engineered to target MerB for accumulation in the endoplasmic reticulum and for secretion to the cell wall. Plants expressing the targeted MerB proteins and cytoplasmic MerA are highly resistant to organic mercury and degrade organic mercury at 10 to 70 times higher specific activity than plants with the cytoplasmically distributed wild-type MerB enzyme. MerB protein in endoplasmic reticulum-targeted plants appears to accumulate in large vesicular structures that can be visualized in immunolabeled plant cells. These results suggest that the toxic effects of organic mercury are focused in microenvironments of the secretory pathway, that these hydrophobic compartments provide more favorable reaction conditions for MerB activity, and that moderate increases in targeted MerB expression will lead to significant gains in detoxification. In summary, to maximize phytoremediation efficiency of hydrophobic pollutants in plants, it may be beneficial to target enzymes to specific subcellular environments.

Figure 1

Modification of the bacterial merB gene for CW- and ER-specific expression in plants. A, WT 642-bp bacterial coding sequence of merB (Bizily et al., 1999) was altered by two cycles of nested PCR to make gene constructs targeting the protein product to the CW (CW-merB) and ER (ER-merB). Arrows show the position of the mutagenic primers used. B, Bacterial disc sensitivity assays with PMA were used to compare the functionality of MerB, CW-MerB, and ER-MerB expressed in E. coli as compared with the a bacterial strain containing only the empty vector plasmid [(−) merB] or the WT bacterial sequence (merB; see “Materials and Methods”). C, Western-blot analysis of MerB, CW- MerB, and ER-MerB expression showed that three different MerB antibodies all react similarly to the modified MerB proteins. Blots of the bacterial extracts were probed with rabbit polyclonal antibody (pAb-MerB) and mouse monoclonal antibodies mAb10E2 and mAb2H8. CW-MerB and ER-MerB ran at slightly higher M_rs than MerB as would be predicted from their increased sequence lengths. CW-MerB and ER-MerB were quantified (see “Materials and Methods”) and normalized relative to MerB levels in strains expressing WT merB. D. MerB, CW-MerB, and ER-MerB protein levels were determined on western blots performed on parallel samples of the crude protein extracts examined in the p-chloromercuribenzoic acid (PCMB) spectrophotometric MerB enzyme assays (Table I). The filter was reacted with pAb-MerB and 2-min exposure is shown. MerB, CW-MerB, and ER-MerB bands were quantified using a film exposed for 20 s (see Table I). The E. coli protein polynucleotide phosphorylase (PNPase), probed with a polyclonal antibody, is shown to confirm the equal loading and transfer of samples to the membrane. Band intensities were normalized relative to WT MerB protein levels from three repetitions of this experiment (see Table I).

Figure 2

Resistance of Arabidopsis merB, CW-merB, and ER-merB lines to organic mercury. Arabidopsis seeds were germinated on 0.8% (w/v) agarose plates containing standard one-half-strength Murashige and Skoog plant growth medium (Bizily et al., 2000). The various plant lines discussed in the text are distributed on the plates (B–D) as shown in A. The plates were dosed with 0 μm PMA (B), 1 μm PMA (C), or 5 μm PMA (D). Plants were grown for 16 d at 22°C with 16 h of light per day. The slight growth of the merA negative control line on 1 μm PMA results from the heavy seed density used in this experiment. Plant lines “a” and “b” are unrelated to these experiments. E and F, Root growth of WT, AB-1, CW-2, and ER-3 lines on one-half-strength Murashige and Skoog medium without (E) and with (F) 2 μm PMA. Seeds were germinated on plates and plants were grown vertically for 10 d.

Figure 3

Organic mercury to Hg(0) conversion by selected Arabidopsis merB, CW-merB, and ER-merB lines. PMA to Hg(0) conversion (Reactions 1 and 2) rates were measured for each transgenic line. The enzyme activity [nanograms of Hg(0) per minute per gram plant tissue [fresh weight]) for each line is reported as the average of three separate assays with 10 seedlings each and the se among these three assays.

Figure 4

Western-blot analysis of MerB, CW-MerB, and ER-MerB expression levels in transgenic Arabidopsis lines. For each line, 30 2-week-old seedlings were removed from mercury-free growth medium and ground in liquid nitrogen. Crude cell protein extracts were resolved on 12.5% (w/v) PAGE (Bizily et al., 2000). Western blots were probed with the pAb-MerB and an anti-phosphoenolpyruvate (PEP)-carboxylase polyclonal antisera. Crude protein loading was reduced 10-fold for lines AB-1 and AB-5. CW-MerB and ER-MerB bands were quantified and normalized relative to the MerB level in the line AB-5.

Figure 5

Immunofluorescence localization of MerB and ER-MerB in fixed Arabidopsis leaf cells. A, MerB WT line. B, MerA line with no MerB (negative control). C and D, ER-MerB line. The confocal images shown are reconstructed from stacks of 0.5-micron photographs. Under equivalent laser parameters, MerB (A) and ER-MerB (C and D) showed different patterns of localization. The dimension of each cell is approximately 25 × 40 μm.