Nature's hydrides: rapid reduction of halocarbons by folate model compounds

Abstract

Halocarbons R–X are reduced to hydrocarbons R–H by folate model compounds under biomimetic conditions. The reactions correspond to a halide–hydride exchange with the methylenetetrahydrofolate (MTHF) models acting as hydride donors. The MTHF models are also functional equivalents of dehalohydrogenases but, unlike these enzymes, do not require a metal cofactor. The reactions suggest that halocarbons have the potential to act as endocrinological disruptors of biochemical pathways involving MTHF. As a case in point, we observe the rapid reaction of the MTHF models with the inhalation anaesthetic halothane. The ready synthetic accessibility of the MTHF models as well as their dehalogenation activity in the presence of air and moisture allow for the remediation of toxic, halogenated hydrocarbons.

---

Introduction

The structural similarity between methylenetetrahydrofolate (MTHF) 2H^1–5 and Thauer's hydrogenase 3H⁶ has led us to investigate imidazolidines 1H as model compounds for the reactivity of these fascinating biomolecules (Fig. 1).

Fig. 1 Structural relationship between imidazolidines 1H, methylenetetrahydrofolate (MTHF) 2H, and Thauer's hydrogenase 3H. Different structural formulae resulting from different N-protonation positions can be found in the literature for 2H. The structure presented in this figure is the one most commonly used.¹

A type of reactivity shared by 2H and 3H is the transfer of negatively charged “hydridic” hydrogen leading to the very stable cationic species [2]⁺ and [3]⁺.

In addition to its function as a hydride donor, MTHF is involved in several other biochemical pathways including folate-dependent DNA methylation, purine, pyrimidine, and amino acid syntheses, as well as the methylation of homocysteine to methionine. MTHF is also the precursor to 5-methyltetrahydrofolate, an important intermediate in its own right.^1–5

The recent discovery that the hydride donor system 2H/[2]⁺ acts as a photobiological cofactor in the light-activated repair enzyme DNA photolyase⁵ demonstrates that MTHF offers significant surprises even after more than sixty years of extensive research. For comprehensive reviews on Thauer's hydrogenase 3H see ref. 6a–d.

Results

Reduction of halogenated hydrocarbons

During our initial studies,⁷ we were surprised to find that imidazolidines 1H (R: methyl, tert-butyl)⁸ are unstable in chloroform and resemble many organometallic compounds and hydrido complexes in this respect. A closer examination revealed that the reason for this instability was the reduction of chloroform to dichloromethane by 1H acting as a hydride donor. By-product of the reduction is the salt [1]⁺Cl⁻ (Scheme 1). We now report that this type of reduction is remarkably general and allows for the dehalogenation of a wide variety of halogenated hydrocarbons including those of toxicological, environmental, and pharmacological significance.

Scheme 1 Reduction of carbon–halogen bonds R′–X by imidazolidine 1H (R: tert-butyl).

To further explore the scope of the reductive dehalogenation, the reactivity of 1H towards a series of chloromethanes and bromomethanes was investigated. With the exceptions of dichloromethane, chloromethane, and bromomethane, all chlorine- and bromine-substituted methanes are reduced by 1H. The reductions occur at room temperature, but the time required for a complete conversion ranges from weeks for chloroform to only minutes for bromoform and carbon tetrabromide. Reactions of 1H with allyl bromide and ethyl 2-bromoacetate resulted in the formation of complex product mixtures. Details about the reaction as well as the compatibility of functional groups are the subject of on-going studies and will be reported in a separate publication.

The nature of the employed solvent (ethanol, hexanes), an excess of 1H, or an excess of the halogenated methane do not significantly influence the reaction rates. Remarkably, the dehalogenation reactions are not inhibited by air or moisture.

An excess of the halogenated hydrocarbon and prolonged reaction times lead to further oxidation of [1]⁺X⁻ to [4]⁺X⁻ (Scheme 1).

The unsaturated imidazoline analogues of 1H were previously found to be incompatible with biomimetic conditions due to their high sensitivity towards moisture.^7c To investigate if the lack of reduction observed for some of the halogenated substrates is due to thermodynamic reasons or is the result of kinetic factors the reaction energies ΔG° for the halide–hydride exchange were determined by quantum chemical calculations (Table 1).⁹ A computational solvation model for water was added to simulate biological conditions (see Methods).¹⁰

Table 1 Computational free energies ΔG° (CBS-QB3 level,¹¹ H₂O-SMD solvent model,¹⁰ kcal mol⁻¹) for the reductions of selected halomethanes (A → B) by 1H (R: tert-butyl)

A	B	X:Cl	X:Br
CX4	CHX3	−48.3	−37.9
CHX3	CH2X2	−44.3	−39.1
CH2X2	CH3X	−41.0	−36.5
CH3X	CH4	−39.8	−29.3
PhX	PhH	−35.9	−31.3

---

The calculated energies show that all reactions are thermodynamically favourable. The lack of reactivity observed for bromomethane, chloromethane, and dichloromethane is accordingly only the result of low reaction rates.

The observed reaction rates show no correspondence to the calculated reaction energies ΔG° but closely match the order of decreasing carbon–halogen bond energies (CBS-QB3 level, kcal mol⁻¹): CH₃Br (77.1), CHCl₃ (77.0), CH₂Br₂ (71.7), CCl₄ (71.9), CHBr₃ (65.3), CBr₄ (62.9).

To test for the ability of 1H to reduce polybrominated commercial flame retardants,¹² 1,3,5-tribromobenzene was investigated (Scheme 2). Dehalogenation to 1,3-dibromobenzene occurred within minutes, even at room temperature, although the total reduction to benzene requires prolonged heating at 150 °C.

Scheme 2 Reaction of imidazolidines 1H with 1,3,5-tribromobenzene.

As carbon–halogen bonds are found in many medicinal drugs, the dehalogenation reactions observed for the MTHF model 1H may well occur in vivo with MTHF itself, with obvious implications for drug metabolism as well as toxicology.

As a test case, the reaction of the inhalation anaesthetic halothane, CF₃CH(Br)Cl, with 1H was examined (Scheme 3). Dehalogenation was observed (NMR) at room temperature within minutes with 2-chloro-1,1,1-trifluoroethane, CF₃CH₂Cl,¹³ being formed as the sole reduction product.

Scheme 3 Reaction of imidazolidines 1H with 2-bromo-2-chloro-1,1,1-trifluoroethane (halothane).

The high reactivity and selectivity observed suggests that 1H may become a new substitute for the tin hydrides¹⁸ and hydrosilanes¹⁹ previously used for selective debrominations.

Discussion

While in vivo studies will inevitably be required to further study the biochemical implications of these findings, the reactivity of 1H does suggest that MTHF-dependent biochemical pathways are likely to be disrupted by halogenated hydrocarbons. It is of interest to note in this context that MTHF is the precursor to 5-methyltetrahydrofolate – a compound under current investigation for the treatment of depression and schizophrenia.⁴

The reported reductive dehalogenation reactions are highly unusual for a simple, neutral, and stable organic molecule like 1H, but have precedence in the fascinating reactivity of dehalohydrogenases.¹⁵

Dehalohydrogenases have been isolated from a variety of microorganisms^14–16 and have attracted great interest for their ability to reduce a number of highly toxic, halogenated hydrocarbons like the notorious 2,3,5,6-tetrachlorodibenzodioxin.¹⁶ A notable difference between our system 1H and dehalohydrogenases reported to date is that the latter require metal cofactors like iron-sulphur proteins or cobalamines.¹⁷ The fact that dehalohydrogenase-type activity is possible with metal-free folate models 1H suggests that metal-free dehalohydrogenases can exist as well.

Methods

Quantum chemical calculations

All calculations were carried out with Gaussian 09, Revision D.01.⁹ The energies were obtained from full optimizations with the CBS-QB3 method¹¹ and the SMD solvent model¹⁰ for water. Minima were verified by the absence of virtual frequencies.

Starting materials

The imidazolines 1H (R: tert-butyl, methyl, para-methyphenyl) were prepared from the respective 1,2-diaminoethanes as described elsewhere.⁷ Halothane and 1,3,5-tribromobenzene were purchased from Sigma-Aldrich Inc.

Reduction of halocarbons

To the best of our knowledge the only previously reported reduction of a halogenated hydrocarbon by a MTHF analogue is that of CCl₄.⁸ Reductions were carried out with a slight excess of the imidazolidines 1H (1.1 equivalent per reducible halogen). In nonpolar solvents like hexanes or diethyl ether, the onset of the reduction is indicated by the appearance of turbidity with subsequent precipitation of the salts [1]⁺X⁻. Control experiments in polar solvents or under inert gas (99.995% argon) with degassed materials did not show noticeably different reaction rates or products. For the less reactive aryl halides and chloromethanes, heating was required to drive the reaction to completion. The progress of the reductions was monitored by NMR and GC-MS. After completion of the reaction, the salts were isolated by filtration under inert gas, washed with diethyl ether, and dried in vacuo. Yield 85–95%. Traces of [4]⁺X⁻ could be removed from [1]⁺X⁻ by recrystallization from 95% ethanol.

Reduction of halothane

To retain the highly volatile product 2-chloro-1,1,1-trifluoroethane (bp +6 °C)^13a the reduction of halothane was monitored in a flame-sealed NMR tube.

Spectroscopic data

2-Bromo-2-chloro-1,1,1-trifluoroethane (halothane). ¹H-NMR (600 MHz, C₆D₆): δ 4.58 [q, ³J(¹H,¹⁹F) = 5.4 Hz]. ¹³C-NMR (600 MHz, C₆D₆): δ 50.2 [q, ²J(¹³C,¹⁹F) = 40.9 Hz, CHClBr], 121.6 [q, ¹J(¹³C,¹⁹F) = 278.0 Hz, CF₃]. ¹⁹F-NMR (600 MHz, C₆D₆vs. CFCl₃): δ −76.1 [d, ³J(¹H,¹⁹F) = 5.3 Hz].

2-Chloro-1,1,1-trifluoroethane. ¹H-NMR (600 MHz, C₆D₆): δ 2.78 [q, ³J(¹H,¹⁹F) = 8.5 Hz]. ¹³C-NMR (600 MHz, C₆D₆): δ 40.1 [q, ²J(¹³C,¹⁹F) = 37.7 Hz, CH₂], 123.1 [q, ¹J(¹³C,¹⁹F) = 275.7 Hz, CF₃]. ¹⁹F-NMR (600 MHz, C₆D₆vs. CFCl₃): δ −71.6 [t, ³J(¹H,¹⁹F) = 8.5 Hz].

1,3,5-Tribromobenzene. ¹H-NMR (400 MHz, CDCl₃): δ 7.61 [s]. ¹³C-NMR (400 MHz, CDCl₃): δ 123.4 [C-2,4,6], 133.0 [C-1,3,5]. GC-MS: t_r = 10.28 min m/z (rel. int%): 314(100)[M⁺], 235(30), 156(15), 74(32).

1,3-Dibromobenzene. ¹H-NMR (400 MHz, CDCl₃): δ 7.67 [m], 7.43 [m], 7.11 [m]. ¹³C-NMR (400 MHz, CDCl₃): δ 123.1 [C-1,3], 130.3 [C-4,6], 131.2 [C-5], 134.2 [C-2]. GC-MS: t_r = 8.57 min m/z (rel. int%): 236(100)[M⁺], 155(41), 75(32).

1,3-Di-tert-butylimidazolidinium chloride. Colourless crystals, mp 202–203 °C (from chloroform). ¹H-NMR (400 MHz, CDCl₃): δ 1.55 [18H, s, C(CH₃)₃], 4.05 [4H, s, CH₂CH₂], 8.85 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, CDCl₃): δ 28.2 [C(C̲H₃)₃], 45.3 [CH₂CH₂], 57.2 [C(C̲H₃)₃], 154.1 [N₂CH⁺].

1,3-Di-tert-butylimidazolidinium bromide. Colourless crystals, mp 230–231 °C (dec.) (from chloroform). ¹H-NMR (400 MHz, CDCl₃): δ 1.56 [18H, s, C(CH₃)₃], 4.10 [4H, s, CH₂CH₂], 8.46 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, CDCl₃): δ 28.3 [C(C̲H₃)₃], 45.5 [CH₂CH₂], 57.2 [C̲(CH₃)₃], 153.1 [N₂CH⁺].

1,3-Di-tert-butylimidazolium chloride. Colourless crystals, mp 209–210 °C (from chloroform). ¹H-NMR (400 MHz, CDCl₃): δ 1.80 [18H, s, C(CH₃)₃], 7.54 [4H, s, HC=CH], 10.54 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, CDCl₃): δ 30.3 [s, C(C̲H₃)₃], 60.8 [C̲(CH₃)₃], 119.7 [HC=CH], 134.5 [N₂CH⁺].

1,3-Di-tert-butylimidazolium bromide. Colourless crystals, mp 219–220 °C (from chloroform). ¹H-NMR (400 MHz, CDCl₃): δ 1.85 [18H, s, C(CH₃)₃], 7.39 [4H, s, HC=CH], 10.37 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, CDCl₃): δ 30.4 [C(C̲H₃)₃], 61.2 [C̲(CH₃)₃], 118.9 [HC=CH], 135.2 [N₂CH⁺].

1,3-Dimethylimidazolidinium bromide. Colourless crystals, mp 189–191 °C (from chloroform). ¹H-NMR (400 MHz, CDCl₃): δ 3.35 [18H, s, NCH₃], 4.01 [4H, s, NCH₂], 9.54 [1H, s, N₂CH⁺]. ¹H-NMR (400 MHz, DMSO-D₆): δ 3.08 [18H, s, NCH₃], 3.85 [4H, s, NCH₂], 8.41 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, CDCl₃): δ 35.2 [NCH₃], 50.9 [CH₂CH₂], 159.4 [N₂CH⁺]. ¹³C-NMR (400 MHz, DMSO-D₆): δ 34.1 [NCH₃], 50.2 [CH₂CH₂], 158.3 [N₂CH⁺].

1,3-Di-para-methylphenylimidazolidine. Colourless crystals, mp 157–158 °C. ¹H-NMR (400 MHz, CDCl₃): δ 2.28 [6H, s, C–CH₃], 3.59 [4H, s, CH₂CH₂], 4.60 [2H, s, N₂CH₂], 6.58 [4H, d, ³J = 8.5 Hz, meta-CH], 7.10 [4H, d, ³J = 8.5 Hz, ortho-CH]. ¹³C-NMR (400 MHz, CDCl₃): δ 20.4, [C–C̲H₃], 46.8 [CH₂CH₂], 66.4 [N₂CH₂], 112.5, 126.7, 129.8, 144.4. GC-MS: t_r = 17.9 min m/z (rel. int%): 252(72)[M⁺], 251(100), 133(39), 118(16), 105(81), 91(30), 65(10).

1,3-Di-para-methylphenylimidazolium bromide. Colourless crystals, mp 292–293 °C (from methanol). ¹H-NMR (400 MHz, DMSO-D₆): δ 2.35 [6H, s, C–CH₃], 4.56 [4H, s, CH₂CH₂], 7.36 [4H, d, ³J = 8.4 Hz, meta-CH], 7.53 [4H, d, ³J = 8.4 Hz, ortho-CH], 9.89 [1H, s, N₂CH⁺]. ¹³C-NMR (400 MHz, DMSO-D₆): δ 20.3, [C–C̲H₃], 48.2 [CH₂CH₂], 118.1, 129.9, 133.6, 136.4, 150.9 [N₂CH⁺].

Conclusions

Our observation that dehalohydrogenase activity is possible with a simple, metal-free MTHF model 1H suggests that metal-free dehalohydrogenases may likewise exist.

The ready synthetic accessibility of the model compounds 1H as well as their dehalogenation activity in the presence of air and moisture suggest their use for the remediation of toxic, halogenated hydrocarbons. The reactions suggest that halocarbons are likely to act as endocrinological disruptors of folate dependent pathways.

Acknowledgements

The authors are grateful for editorial comments by professors Gordon Lange and Mario Monteiro at different stages of the manuscript. N. S. M. and M. K. D. wrote the paper and M. K. D. carried out the quantum chemical calculations. All authors contributed about equally to the experiments. We thank one of the referees for enquiring about the reactivity of allyl bromide and ethyl 2-bromoacetate. The compatibility of other functional groups with the reported reduction is under investigation.

Notes and references

1. R. G. Kallen and W. P. Jencks, J. Biol. Chem., 1966, 241, 5851–5863.
2. (a) L. Jaenicke, Angew. Chem., 1961, 73, 449–480; (b) L. Jaenicke, Annu. Rev. Biochem., 1964, 33, 287–312; (c) J. T. Fox and P. J. Stover, Folic Acid and Folates, ed. G. Litwack, Elsevier Academic Press Inc., San Diego, 1st edn, 2008, vol. 79, ch. 1, pp. 1–44; (d) E. P. Quinlivan, A. D. Hanson and J. F. Gregory, Anal. Biochem., 2006, 348, 163–184.
3. (a) E. R. Werner, N. Blau and B. Thoöny, Biochem. J., 2011, 438, 397–414; (b) M. Födinger, W. H. Hörl and G. Sunder-Plassmann, J. Nephrol., 2000, 13, 20–33; (c) K. Robien and C. M. Ulrich, Am. J. Epidemiol., 2003, 157, 571–582.
4. (a) A. L. Miller, Altern. Med. Rev., 2008, 13, 216–226; (b) M. Fava and D. Mischoulon, J. Clin. Psychiatry, 2009, 70(5), 12–17; (c) S. M. Stahl, CNS Spectr., 2007, 12, 739–744.
5. (a) A. Sancar, Chem. Rev., 2003, 103, 2203–2237; (b) R. F. Pauszek III, G. Kodali and R. J. Stanley, J. Phys. Chem. A, 2014, 118, 8320–8328.
6. (a) S. Shima and R. K. Thauer, Chem. Rec., 2007, 7, 37–46; (b) B. H. Geierstanger, T. Prasch, C. Griesinger, G. Hartmann, G. Buurman and R. K. Thauer, Angew. Chem., Int. Ed. Engl., 1998, 37, 3300–3303; (c) S. Shima, E. J. Lyon, M. Sordel-Klippert, M. Kauß, J. Kahnt, R. K. Thauer, K. Steinbach, X. Xie, L. Verdier and C. Griesinger, Angew. Chem., Int. Ed. Engl., 2004, 43, 2547–2551; (d) E. J. Lyon, S. Shima, G. Buurman, S. Chowdhuri, A. Batschauer, K. Steinbach and R. K. Thauer, Eur. J. Biochem., 2004, 271, 195–204.
7. (a) M. K. Denk, S. Gupta, J. Brownie, S. Tajammul and A. J. Lough, Chem.–Eur. J., 2001, 7, 4477–4486; (b) M. K. Denk, A. Hezarkhani and F. Zheng, Eur. J. Inorg. Chem., 2007, 3527–3534; (c) M. K. Denk, J. M. Rodezno, S. Gupta and A. J. Lough, J. Organomet. Chem., 2001, 617–618, 242–253.
8. E. Rabe and H. W. Wanzlick, Liebigs Ann. Chem., 1973, 40–44.
9. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, GAUSSIAN 09 (Revision D.01), Gaussian Inc., Wallingford, CT, 2013.
10. A. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 6378–6396.
11. J. A. Montgomery, M. J. Frisch, J. W. Ochterski and G. A. Petersson, J. Chem. Phys., 2000, 112, 6532–6542.
12. (a) C. A. de Wit, Chemosphere, 2002, 46, 583–624; (b) The Stockholm convention (2001) and its annexes list polychlorinated and brominated hydrocarbons that have been phased out due to their persistence in the environment, their bioaccumulation, their toxicity and capability to act as endocrinological disruptors (http://chm.pops.int/). Excluding the list of newly proposed compounds, the list now (2015) comprises 23 chlorinated and brominated hydrocarbons.
13. (a) J. Chapman, R. Hill, J. Muir, C. W. Suckling and D. J. Viney, J. Pharm. Pharmacol., 1967, 19, 231–239; (b) F. J. Weigert and J. K. Karel, J. Fluorine Chem., 1987, 37, 125–149.
14. J. Dolfing, A. J. van den Wijngaard and D. B. Janssen, Biodegradation, 1993, 4, 261–282.
15. (a) P. Adriaens and D. Grbić-Galić, Chemosphere, 1994, 29, 2253–2259; (b) A. Neumann, G. Wohlfarth and G. Diekert, J. Biol. Chem., 1996, 271, 16515–16519; (c) X. Maymó-Gatell, Y. T. Chien, J. M. Gossett and S. H. Zinder, Science, 1997, 276, 1568–1571; (d) W. Schumacher, C. Holliger, A. J. Zehnder and W. R. Hagen, FEBS Lett., 1997, 409, 421–425; (e) G. Wohlfarth and G. Diekert, Curr. Opin. Biotechnol., 1997, 8, 290–295; (f) C. Holliger, G. Wohlfarth and G. Diekert, FEMS Microbiol. Rev., 1998, 22, 383–398; (g) A. Mägli, M. Messmer and T. Leisinger, Appl. Environ. Microbiol., 1998, 64, 646–650; (h) L. Adrian, U. Szewzyk, J. Wecke and H. Görisch, Nature, 2000, 408, 580–583.
16. (a) M. Bunge, L. Adrian, A. Kraus, M. Opel, W. G. Lorenz, J. R. Andreesen, H. Görisch and U. Lechner, Nature, 2003, 421, 357–360; (b) H. Smidt and W. M. de Vos, Annu. Rev. Microbiol., 2004, 58, 43–73.
17. (a) M. Bommer, C. Kunze, J. Fesseler, T. Schubert, G. Diekert and H. Dobbek, Science, 2014, 346, 455–458; (b) K. A. Payne, C. P. Quezada, K. Fisher, M. S. Dunstan, F. A. Collins, H. Sjuts, C. Levy, S. Hay, S. E. Rigby and D. Leys, Nature, 2015, 517, 513–516.
18. (a) R. J. Strunk, P. M. DiGiacomo, K. Aso and H. G. Kuivila, J. Am. Chem. Soc., 1970, 92, 2849–2856; (b) A. Sanders and W. P. Giering, J. Org. Chem., 1973, 38, 3055; (c) T. Ando, F. Namigata, H. Yamanaka and W. Funasaka, J. Am. Chem. Soc., 1967, 89, 5719–5721.
19. (a) C. Chatgilialoglu, D. Griller and M. Lesage, J. Org. Chem., 1988, 53, 3641–3642; (b) M. Ballestri, C. Chatgilialoglu, K. B. Clark, D. Griller, B. Geise and B. Kopping, J. Org. Chem., 1991, 56, 678–683; (c) C. Chatgilialoglu, Chem.–Eur. J., 2008, 14, 2310–2320.

---