Selective, light-driven enzymatic dehalogenations of organic compounds

Bhavin Siritanaratkula, Shams T. A. Islama, Torsten Schubertb, Cindy Kunzeb, Tobias Gorisb, Gabriele Diekertb and Fraser A. Armstrong*a
aInorganic Chemistry Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QR, Oxfordshire, UK
bInstitut für Mikrobiologie, Friedrich-Schiller-Universität Jena, Lehrstuhl für Angewandte und Ökologische Mikrobiologie, Philosophenweg 12, 07743 Jena, Germany

Received 4th August 2016 , Accepted 24th August 2016

First published on 7th September 2016

Tetrachloroethene reductive dehalogenase (PceA), a corrinoid-containing enzyme from Sulfurospirillum multivorans, is highly active for the sequential reduction of the organohalide tetrachloroethene (PCE) to trichloroethene (TCE), then regiospecifically to cis-1,2-dichloroethene (cDCE). We demonstrate direct electron transfer from graphite and semiconductor electrodes to PceA adsorbed onto the electrode surface. Colloidal TiO2 nanoparticles modified with PceA efficiently carry out the sequence of dehalogenation reactions under UV light irradiation.


Enzymes can be coupled to semiconductors that provide electrons or holes under light irradiation, thus directing catalysis of highly selective redox reactions. So far, most of these reactions have been related to solar fuels (artificial photosynthesis). Systems based upon non-toxic TiO2 have been constructed for H2 production and CO2 reduction, using hydrogenases1 and carbon monoxide dehydrogenases (CODH),2,3 respectively. As part of a program to extend research to alternative organic targets, a flavocytochrome c3 fumarate reductase was recently used to demonstrate C[double bond, length as m-dash]C hydrogenation using electrons from dye-sensitized TiO2 nanoparticles.4

In a completely different context, semiconductors have been studied for their application in light-driven decomposition of chlorinated organic compounds that are common toxic contaminants in water. For example, trichloroethene (TCE) has been used as a model target for studying photocatalytic decomposition of organohalides on TiO2.5,6 This non-specific process is enhanced by the presence of O2, as the mechanism probably proceeds by generation of oxygen radicals on the bare TiO2 surface. These radicals react with TCE through a complex series of steps to yield a mixture of products such as phosgene, HCl, CO and CO2.

Enzymes offer possibilities for highly controlled dehalogenation reactions, some of which could have analytical or synthetic use. Sulfurospirillum multivorans (previously known as Dehalospirillum multivorans) is a bacterium that can use tetrachloroethene (commonly known as perchloroethylene, or PCE) or trichloroethene (TCE) as electron acceptors in anaerobic respiration.7,8 Tetrachloroethene is thereby selectively reduced via trichloroethene (TCE) to cis-1,2-dichloroethene (cDCE), as shown in eqn (1) and 2.

image file: c6ra19777a-u1.tif(1)
image file: c6ra19777a-u2.tif(2)

The terminal reductase in this process, the PCE reductive dehalogenase (PceA),9 contains a corrinoid (norpseudo-B12)10 and iron sulfur clusters as cofactors; its structure was recently determined.11 Two [4Fe–4S] clusters are located close to the co-containing norpseudo-B12, allowing electron transport from the enzyme surface to the buried active site.

By analogy with ongoing strategies in artificial photosynthesis, the highly specific reductive dechlorination reactions catalysed by reductive dehalogenases should be possible using light-activated semiconductors. Unlike artificial photosynthesis, however, chemical energy is not stored: sequential conversions of PCE to TCE and TCE to cDCE are both thermodynamically favourable, based on estimated reduction potentials (pH 7) of +0.58 V and +0.54 V respectively.12 As well as greatly accelerating rates, the enzyme ensures both substrate specificity and regiospecificity (the latter for eqn (2)).

We have now explored the extension of enzyme-semiconductor strategies to achieve selective reductive dehalogenation of compounds such as PCE and TCE, using UV light and colloidal anatase TiO2 (bandgap 3.2 eV). The activity of PceA has previously been studied by chemical assays9 and cyclic voltammetry in the presence of electron mediators, but no evidence for direct interfacial electron transfer with a conducting surface has been reported.13 Therefore, we first investigated the direct adsorption of PceA onto different electrodes, in order to establish that direct interfacial electron transfer to PceA is feasible.

Results and discussion

All electrochemical measurements were performed under anaerobic conditions in a glove box (Belle Technologies, O2 < 3 ppm). Fig. 1 shows cyclic voltammograms obtained with PceA adsorbed on a stationary pyrolytic graphite edge (PGE) electrode. Briefly, the surface of the PGE electrode was abraded with sandpaper, then 1 μL of PceA solution (21 μM) was applied. Upon injection of an aliquot of TCE, dissolved in ethanol, yielding 0.8 mM final concentration in the aqueous cell solution, a sharp reduction wave appears with an onset potential of −0.45 V vs. SHE at pH 7. At an electrode such as PGE that shows metallic conductivity properties, the onset potential for an adsorbed enzyme catalyzing an irreversible reaction is related to the potential of the redox centre at which electrons enter or leave the enzyme:14 in this case one of the two [4Fe–4S] clusters is the likely entry site. The decrease in reduction current that is observed on subsequent cycles is mainly due to evaporation of the volatile substrate, and to a lesser extent due to film loss. The isolated norpseudo-B12 cofactor was also tested with a PGE electrode, but did not show any activity towards catalytic TCE reduction.
image file: c6ra19777a-f1.tif
Fig. 1 Cyclic voltammograms of PceA on a pyrolytic graphite ‘edge’ (PGE) electrode, before and after injection of the substrate, trichloroethene (TCE). Reaction conditions: Tris buffer 0.1 M, pH 7.0, 25 °C, scan rate 20 mV s−1, electrode is stationary.

To establish that PceA adsorbs on TiO2 nanoparticles (P25: anatase/rutile ratio approx. 3/1), the UV-vis absorbance of a solution of PceA was recorded before and after stirring with TiO2 particles (1 mg mL−1) for 20 min (Fig. 2). At the enzyme concentration for the photochemical reaction described in the following section (21 nM), the UV-vis absorbance of the enzyme was too low to be observed clearly, therefore a more concentrated solution was used (0.1 μM). Even at this higher concentration, the enzyme was almost totally adsorbed from solution onto the TiO2 particles after stirring for 20 min. It is therefore reasonable to assume that at the lower enzyme concentration used for the photochemical experiments, the adsorption of PceA onto TiO2 is at least 90% complete, although it is unlikely that all enzyme molecules are attached in an optimal way for interfacial electron transfer.

image file: c6ra19777a-f2.tif
Fig. 2 UV-vis absorbance of PceA (0.1 μM) in Tris buffer pH 7.0, before and after stirring with TiO2 particles (1 mg mL−1).

Fig. 3 shows the activity of PceA adsorbed on a TiO2 electrode. The electrode was constructed by depositing TiO2 nanoparticles on an conducting indium tin oxide (ITO) surface, as described previously.15 Compared to metallic-like PGE, n-type TiO2 attains significant conductivity only when a more negative potential is applied (the flat-band potential is a guide15) and background cycles each show a prominent trumpet-like shape that shifts to more negative potential with increasing pH. Adsorption of enzyme does not alter the voltammetric response compared to bare TiO2. As shown in Fig. 3A, injections of TCE (0.8 mM) result in a catalytic net reduction current that commences once the electrode potential has entered the conductive region of TiO2. The background-subtracted cyclic voltammograms shown in Fig. 3B confirm that PceA is active in the pH range 6–8, and that pH 7 is a suitable condition for experiments with TiO2. In similar experiments, using PCE (0.8 mM) as the substrate instead of TCE (Fig. 3C and D), reduction currents were also observed. Control experiments with no enzyme confirmed that neither PCE or TCE were reduced on either bare PGE or TiO2 electrodes; therefore the observed reduction current is due to enzyme catalysis. These measurements showed that PceA can receive electrons from TiO2 and use the electrons in the reductive dehalogenation of TCE and PCE.

image file: c6ra19777a-f3.tif
Fig. 3 Cyclic voltammograms of PceA adsorbed on a TiO2 electrode. Panel (A) shows the current before (black line) and after (red line) injection of TCE (final concentration 0.8 mM). Panel (B) shows the blank-subtracted voltammograms at pH 6–8. Panels (C) and (D) are the same experiments, but with injection of 0.8 mM PCE instead of TCE. Reaction conditions: Tris buffer, 25 °C, 20 mV s−1, stationary electrode, no gas flow.

Light-driven conversion of TCE or PCE to cDCE under anaerobic conditions was demonstrated in a colloidal system. Irradiation with UV light generates electron–hole pairs in TiO2 particles. The electrons are transferred to PceA to drive the reduction of PCE or TCE, and the holes are quenched by electrons from the amine buffer, Tris.1 Particular care was taken to minimize losses of chlorinated ethenes by using a gas-tight vessel and avoiding rubber septa that absorb the volatile compounds. In a typical experiment, a solution of PceA (0.17 nmol) in 8 mL of Tris buffer (Tris 100 mM, pH 7.0), was mixed with TiO2 particles (P25, 8 mg) and the suspension was placed in a vial suitable for near-UV transmission and closed tightly with a Teflon seal. A solution of the substrate, either tetrachloroethene (PCE) or trichloroethene (TCE) in ethanol, was injected into the suspension before sealing the vial and commencing irradiation with a UV lamp (365 nm, 8 W, 1.5 cm distance). Small aliquots (0.5 mL) of the solution were sampled periodically by syringe extraction through the Teflon seal, and the product was detected and quantified by GC-MS headspace analysis. The temperature of the suspension increased by 5–6 °C (typical values being 24 °C to 30 °C) over 4 h of irradiation (it was not practical to thermostat the vial). Errors arising from the sample transfers for GC-MS analysis and calibration were minimized as far as practically feasible.

Fig. 4A shows the consumption of TCE and concurrent production of cDCE under UV irradiation, reproducibly reaching 85% conversion after 2 h. The subsequent decrease in activity varied from one experiment to another but is likely due to slow enzyme deactivation caused by oxidative damage from the valence-band holes. Control experiments with TiO2 but no enzyme showed a decrease to ca. 70% of the starting TCE concentration after 4 h: the latter is ascribed to decomposition on bare TiO2 and, to a lesser extent, losses during sampling. It is known that chlorinated organic compounds undergo slow photo-decomposition on TiO2 under anaerobic conditions, giving CO2 and HCl as final products.16 In our own study (see control reaction included in Fig. 4A) we also noted that TCE undergoes decomposition5 on bare TiO2, albeit at a much slower rate than the enzymatic pathway. Photocatalysis of TCE on bare TiO2 does not produce cDCE,5 and PceA is required for the specific reduction of TCE to cDCE to occur.

image file: c6ra19777a-f4.tif
Fig. 4 Enzymatic reductive dechlorination by UV irradiation of TiO2–PceA. (A) Time courses of concentrations of trichloroethene (TCE, solid black line) and cis-1,2-dichloroethene (cDCE, red line) in sealed, aqueous TiO2–PceA suspensions, under UV irradiation, commencing from an initial TCE concentration of 0.55 mM. The control reaction with no enzyme is shown in dashed black line; no cDCE was detected in the control. (B) A similar set up, but starting with tetrachloroethene (PCE, blue line) in solution at 0.26 mM. Normalized concentrations of TCE and cDCE are shown as black and red lines, respectively. In the control reaction without enzyme (not shown), neither TCE nor cDCE could be detected. Conditions given in text.

For the experiment starting with PCE in the solution (Fig. 4B), production of both TCE and cDCE was observed after irradiation, with the TCE concentration reaching a maximum value after approximately 0.75 h, before decreasing due to conversion to cDCE. When only cDCE was present at the start of the reaction, no difference between reaction solutions with and without the enzyme was observed (see ESI, Fig. S1), showing that some of the slight decrease in cDCE concentration (after more than 2 h referring to Fig. 4A) is due to slow decomposition on bare TiO2. Reductive dechlorination by PceA thus terminates at cDCE, which is not reduced further. Cyclic voltammograms of PceA with cDCE injections also showed no reduction current (see ESI, Fig. S2), in agreement with previous reports that cDCE is not a substrate for this enzyme.9

Together, the results demonstrate rapid and sequential photo-driven conversion of PCE to TCE and finally cDCE by PceA.

From the maximum product concentration obtained and the molecular mass of PceA based on crystallography (52.3 kDa calculated for a monomer),11 we estimated the turnover number (TON) at 18[thin space (1/6-em)]000 per enzyme active site for cDCE production after 4 h, and an average turnover frequency (TOF) of 2.4 s−1 (per enzyme active site). Both TON and TOF are lower limits for their true values, because not all enzyme molecules are likely to be adsorbed onto TiO2 in an electroactive configuration, and the product cDCE is slowly decomposed on bare TiO2. The TOF from the photochemical experiments is an order of magnitude lower than the TOF value (ca. 90 s−1) measured by a chemical assay using reduced methyl viologen.11

In an attempt to see if visible light could be utilized, we modified the TiO2 nanoparticles with a Ru-based dye [(Ru(bpy)2(4,4-(PO3H2)2bpy))Br2 (bpy = 2,2-bipyridine)] as used in previous systems with CODH or flavocytochrome c3 fumarate reductase.2,4 Upon visible light irradiation (λ > 420 nm), the dye–TiO2–PceA system was able to convert TCE to cDCE, although at a much lower rate (see ESI, Fig. S3). It may be possible to replace the dye–TiO2 system with a native visible-light active semiconductor, and this is currently under investigation.

This report provides proof of principle that reductive dehalogenases can be harnessed for solar-driven transformations of halogenated organic compounds. Several reductive dehalogenases have been characterized, each one displaying a characteristic spectrum of substrates and products,17,18 and future research could incorporate these various enzymes into similar semiconductor-based light systems for selective halogen/H substitutions.


We have established the scope for selective light-driven reductive dehalogenation of organohalides using enzyme-modified semiconductor nanoparticles. Specifically, our discovery of a hybrid enzyme/materials pathway for the direct conversion of PCE to TCE and cDCE provides an interesting and conceptually valuable addition to the library of solar-to-chemical processes currently being studied.



Tris buffer solutions were prepared by dissolving Tris (Tris(hydroxymethyl)aminomethane, Fisher, 99.8%) and (NH4)2SO4 (Fisher, 99.8%) in purified water (Millipore, 18 MΩ cm) and titrating to the desired pH using HCl or NaOH.

Tetrachloroethene (PCE, Sigma Aldrich, 99%), trichloroethene (TCE, Sigma Aldrich, 99.5%), cis-1,2-dichloroethene (cDCE, Sigma Aldrich, 97%), and ethanol (Sigma Aldrich, 99.8%) were used without further purification.

Enzyme purification

The dehalogenase PceA was purified from a Sulfurospirillum multivorans mutant strain producing PceA with a C-terminal Strep-Tag. The strain was cultivated in 12 L pyruvate/fumarate-containing medium8 in the presence of 0.72 mM FeSO4 and kanamycin. The cells were harvested under oxic conditions before disruption via French Press under anaerobic conditions. The PceAstrep product was purified under anoxic conditions via gravity flow using Strep-Tactin (IBA, Göttingen, Germany) as described previously.9,11 A total of 2.3 mg of PceA-Strep were obtained from 1.9 g cell protein. The enzyme was stored at −80 °C.

Protein film electrochemistry

All electrochemical measurements were conducted in a glove box under an N2 atmosphere (Belle Technologies, O2 < 3 ppm). A single-compartment cell with a water jacket was used with a 3-electrode setup: PGE or TiO2 (as working electrodes), Pt wire (as counter electrode), and Ag/AgCl (as reference electrode). An Autolab PGSTAT30 potentiostat controlled by Nova software (EcoChemie) was used to record voltammograms. The surface of the PGE electrode was abraded with sandpaper (P400 Tufbak Durite) and rinsed with deionized water before applying enzyme solution (typically 1 μL) to the electrode surface. After waiting for approximately 30 s, the electrode was placed in the cell and measurements were initiated. The compounds PCE, TCE, and cDCE were dissolved in EtOH before injecting into the cell.

Electrode preparation

Pyrolytic graphite ‘edge’ (PGE) electrodes were fabricated in-house; a small piece of pyrolytic graphite was embedded in epoxy resin within a Teflon casing containing a metal contact rod. The graphite was oriented so that the basal planes were perpendicular to the surface (approximately 0.03 cm2).

The TiO2 electrodes were fabricated with TiO2 particles (P25, Degussa) on ITO slides (SPI Supplies, 1.1 mm thick, 8–12 Ω) using the doctor's blade method. The TiO2 coated ITO slides were calcined in air at 450 °C for 30 min. A copper wire was attached with silver paint, then the exposed surfaces were covered with epoxy.

UV-vis spectroscopy

Absorbance spectra were recorded with a Perkin-Elmer Lambda 19 UV-vis spectrometer using quartz cuvettes. The spectrum of the PceA solution (5.95 μg mL−1 in Tris buffer pH 7) was measured before and after stirring (following centrifugation) with TiO2 particles (1 mg mL−1) for 20 min.

Photochemical reactions

A sample of TiO2 particles (P25, Degussa, 7 mg) was sonicated in Tris buffer (pH 7, 7 mL) for 15 min, then 7 μL of PceA solution (1.19 mg mL−1) was added to the suspension which was stirred for a further 20 min. The substrate solution, either PCE or TCE in EtOH, was added and the reaction vessel was sealed with a Teflon cap. Note that TCE is known to dissolve rubber, therefore rubber stoppers should be avoided for prolonged reactions. A UV lamp (UVL-28 EL series, 365 nm, 8 W, held at 1.5 cm distance) was used to illuminate the vessel under continuous stirring. Periodically, a 0.5 mL sample of the liquid phase was taken for product detection by GC-MS headspace analysis (Agilent 7890B, HP5 column, He carrier gas). The GC sampling vial was incubated at 60 °C for 5 min before a sample of the headspace was taken. The GC oven was held at 40 °C for 5 min from injection, then ramped up to 290 °C at a rate of 50 °C min−1. Calibration was established with standard solutions of PCE, TCE, and cDCE. The cell was not thermostatted: during irradiation, the temperature of the solution increased from 24 °C to 30 °C over the course of 4 h.


We thank James Wickens for assistance with GC-MS measurements. Financial support for CK, TS, GD and TG in the framework of the German Research Foundation (DFG) Research Unit FOR 1530 is gratefully acknowledged. FAA is a Royal Society-Wolfson Research Merit Award holder.


  1. E. Reisner, D. J. Powell, C. Cavazza, J. C. Fontecilla-Camps and F. A. Armstrong, J. Am. Chem. Soc., 2009, 131, 18457–18466 CrossRef CAS PubMed.
  2. T. W. Woolerton, S. Sheard, E. Reisner, E. Pierce, S. W. Ragsdale and F. A. Armstrong, J. Am. Chem. Soc., 2010, 132, 2132–2133 CrossRef CAS PubMed.
  3. A. Bachmeier and F. Armstrong, Curr. Opin. Chem. Biol., 2015, 25, 141–151 CrossRef CAS PubMed.
  4. A. Bachmeier, B. J. Murphy and F. A. Armstrong, J. Am. Chem. Soc., 2014, 136, 12876–12879 CrossRef CAS PubMed.
  5. J. Fan and J. T. Yates, J. Am. Chem. Soc., 1996, 118, 4686–4692 CrossRef CAS.
  6. M. D. Driessen, A. L. Goodman, T. M. Miller, G. A. Zaharias and V. H. Grassian, J. Phys. Chem. B, 1998, 102, 549–556 CrossRef CAS.
  7. A. Neumann, H. Scholz-Muramatsu and G. Diekert, Arch. Microbiol., 1994, 162, 295–301 CrossRef CAS PubMed.
  8. H. Scholz-Muramatsu, A. Neumann, M. Meßmer, E. Moore and G. Diekert, Arch. Microbiol., 1995, 163, 48–56 CrossRef CAS.
  9. A. Neumann, G. Wohlfarth and G. Diekert, J. Biol. Chem., 1996, 271, 16515–16519 CrossRef CAS PubMed.
  10. B. Kräutler, W. Fieber, S. Ostermann, M. Fasching, K.-H. Ongania, K. Gruber, C. Kratky, C. Mikl, A. Siebert and G. Diekert, Helv. Chim. Acta, 2003, 86, 3698–3716 CrossRef.
  11. M. Bommer, C. Kunze, J. Fesseler, T. Schubert, G. Diekert and H. Dobbek, Science, 2014, 346, 455–458 CrossRef CAS PubMed.
  12. T. M. Vogel, C. S. Criddle and P. L. McCarty, Environ. Sci. Technol., 1987, 21, 722–736 CrossRef CAS PubMed.
  13. G. Diekert, D. Gugova, B. Limoges, M. Robert and J.-M. Savéant, J. Am. Chem. Soc., 2005, 127, 13583–13588 CrossRef CAS PubMed.
  14. S. V. Hexter, T. F. Esterle and F. A. Armstrong, Phys. Chem. Chem. Phys., 2014, 16, 11822 RSC.
  15. A. Bachmeier, V. C. C. Wang, T. W. Woolerton, S. Bell, J. C. Fontecilla-Camps, M. Can, S. W. Ragsdale, Y. S. Chaudhary and F. A. Armstrong, J. Am. Chem. Soc., 2013, 135, 15026–15032 CrossRef CAS PubMed.
  16. P. Calza, C. Minero and E. Pelizzetti, Environ. Sci. Technol., 1997, 31, 2198–2203 CrossRef CAS.
  17. B.-E. Jugder, H. Ertan, S. Bohl, M. Lee, C. P. Marquis and M. Manefield, Front. Microbiol., 2016, 7 Search PubMed.
  18. T. Schubert and G. Diekert, in Organohalide Respiring Bacteria, ed. L. Adrian and F. E. Loffler, Springer-Verlag Berlin Heidelberg, in press Search PubMed.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19777a

This journal is © The Royal Society of Chemistry 2016