O OH HO HO HO S Ph N OSO3 – K+ Myrosinase PhCH2N C S 1 2 NO2 O P (EtO)2 S 3 NO2 O P (EtO)2 S + [S] Liver Microsomes or CF3CO3H 410 J. CHEM. RESEARCH (S), 1997 J. Chem. Research (S), 1997, 410–411† Oxidation of Isothiocyanates to Isocyanates using Dimethyldioxirane; Relevance to Biological Activity of Isothiocyanates† Nicola E. Davidson and Nigel P. Botting* School of Chemistry, University of St. Andrews, St. Andrews, Fife KY16 9ST, UK The reaction of organic isothiocyanates with dimethyldioxirane in acetone produces isocyanates in good yields, trapped out as the ureas by reaction with isopropylamine.Organic isothiocyanates are widely distributed in plants, including cruciferous vegetables1 such as Brussels sprouts, cauliflower and broccoli. They in fact exist as glucosinolates2 1 and are only released by the action of the enzyme myrosinase, when the plant is damaged or cooked. The occurrence of isothiocyanates in the diet means that there is considerable interest in their biological activities. Some, including benzyl isothiocyanate 2 and phenethyl isothiocyanate, two of the most commonly occurring compounds, have anti-carcinogenic properties.Thus administration of benzyl isothiocyanate prior to treatment with the carcinogen N-nitrosodiethylamine inhibited tumour formation in A/J mice,3 while phenethyl isothiocyanate inhibits induction of lung tumourigenesis by NNK [4-(methylnitrosamino)- 1-(3-pyridyl)butan-1-one] in the forestomach and lung of mice.4 Two potential mechanisms for this anti-carcinogenic action have been proposed.Firstly, the isothiocyanates have been shown to induce phase 2 detoxification enzymes.5 Increased levels of these enzymes result in more rapid removal of the carcinogenic species. Secondly, the isothiocyanates have also been found to inhibit metabolic activation of the carcinogen.6 For example, NNK requires a-hydroxylation by cytochrome P-450 enzymes in order to produce the active carcinogen. Phenethyl isothiocyanate specifically inactivates the P-450 enzymes responsible for this hydroxylation thus preventing activation.The relative importance of these two effects is still open to debate. The chemical mechanism of the inhibition of the cytochrome P-450 enzymes by isothiocyanates is not understood. Isothiocyanates are reactive compounds, readily attacked by nucleophiles. Indeed, they are often used to label proteins by reaction with free amino groups, e.g.using fluorescein isothiocyanate. However, cytochrome P-450 enzymes catalyse oxidation reactions and so it is likely that when the isothiocyanates initially interact with the enzyme they undergo an oxidation reaction, producing a more reactive species which is then responsible for the inactivation. This may also explain the specificity observed, in that there is some activation of the isothiocyanate required in order to produce the inhibitor giving an example of suicide (mechanism-based) inhibition.The insecticides parathion 3 and malathion,7 containing a P�S group, undergo oxidative desulfuration by mammalian liver microsomes to give a P�O group and elemental sulfur (Scheme 2). Model studies using trifluoroperacetic acid resulted in the same transformation. Furthermore, analogous conversions of thiocarbonyl groups to carbonyl groups have also been observed.8 If a similar pathway operates for the oxidation of isothiocyanates, a much more reactive isocyanate would be produced at the enzyme active site.This could then acylate an appropriate amino acid side chain causing inactivation of the cytochrome P-450. There is some evidence for the conversion of 2-naphthyl isothiocyanate to 2-naphthyl isocyanate by rat liver microsomes, but only in very low yields (s0.6% conversion).9,10 The chemical oxidation of isothiocyanates has thus been examined as a model for these biological systems. Prior to this work there were only two previous reports of the conversion of isothiocyanates to isocyanates.In 1890 Kuhn and Lieber reported that heating an isothiocyanate with mercuric oxide at 170 °C gave ca. 20% of the corresponding isocyanate.11 More recently,12 the conversion was achieved using palladium(II) chloride in refluxing 1,4-dioxane. Good to excellent yields of isocyanates were obtained using a range of alkyl and aryl derivatives. The other product in this case was thionyl chloride.However to date no nonmetal- catalysed oxidative conversion of isothiocyanates to isocyanates has been reported. The electrophilic nature of isothiocyanates means that the oxidising agent must be carefully chosen to reduce the possibility of competing nucleophilic attack on the central carbon of the heterocumulene system. Dimethyldioxirane (DMD) thus proved to be the most appropriate reagent, as it is a very reactive non-nucleophilic oxidising agent with acetone as its only by-product.13 When benzyl isothiocyanate was reacted with a solution of DMD14 in dry acetone at room temperature analysis by GCMS showed complete consumption of the isothiocyanate after 15 min and only one major product.The product had an identical retention time and mass spectrum to authentic benzyl isocyanate. The identity of the compound was confirmed by addition of isopropylamine to the reaction solution, which resulted in a decrease in the peak due to the isocyanate and the appearance of a new peak, shown to be due to the 1-isopropyl-3-benzylurea by its mass spectrum.The reaction presumably proceeds via an oxathiirane type intermediate 4 (Scheme 3), formed via either initial transfer of oxygen to sulfur and cyclisation, or direct insertion of oxygen into the carbon–sulfur double bond. Similar mechanisms are proposed for the oxidation of other thiocarbonyl compounds. A range of isothiocyanates were then reacted with DMD and the isocyanates isolated as their urea derivatives 5 following trapping with isopropylamine. The optimised yields of the 1-isopropylureas were good to excellent for a range of isothiocyanates (Table 1), including both alkyl and aryl deriva- *To receive any correspondence.†This is a Short Paper as defined in the Instructions for Authors, Section 5.0 [see J. Chem. Research (S), 1997, Issue 1]; there is therefore no corresponding material in J. Chem. Research (M). Scheme 1 Scheme 2R N C S DMD Acetone R N C S+ O– R N C O S R N C O + [S] 4 Trapping with isopropylamine N N O R H H 5 J.CHEM. RESEARCH (S), 1997 411 tives. The best yields were obtained using a five-fold excess of DMD. It could be argued that the amine and isothiocyanate initially react to form the thiourea which can then be oxidised to the urea by DMD. However, examination by both GCMS and TLC, against the authentic thiourea, gave no evidence for its formation during the reaction. It was also observed that reaction of the isothiocyanates with isopropylamine was much slower than the observed oxidation of the isothiocyanate to the isocyanate.It thus seemed most likely that the reaction followed the pathway as given in Scheme 3. However, attempts were made to isolate the isocyanates directly from the reaction solution. Unfortunately, these were hampered by both the small scale of the reactions, due to the low concentrations of DMD available, and by competing hydrolysis, illustrated by the recovery of 1,3-diphenylurea in 94% yield from the attempted isolation of phenyl isocyanate.The latter arises from the difficulty observed in obtaining sufficiently dry solutions of DMD in acetone. The presence of the free isocyanate in the reaction solution was finally demonstrated by means of FTIR studies. In acetone solution benzyl isothiocyanate gave peaks at 2169 and 2092 cmµ1. The addition of DMD gave rise to new peaks at 2276 and 2341 cmµ1. The former peak was identical to that produced by authentic benzyl isocyanate.The peak at 2341 cmµ1 was found to be due to dissolved carbon dioxide, presumably arising from hydrolysis of the isocyanate. Indeed the hydrolysis of authentic isocyanate could be monitored in both acetone and the DMD solution. The fate of the sulfur produced in the reaction has not been elucidated. However in some of the GCMS analyses there is evidence for a species similar tmethyl disulfide or dimethyl sulfone.The observation of such oxidised sulfur species accounts for the requirement for excess DMD in order to obtain good yields. A number of other oxidising agents were also examined. m-Chloroperbenzoic acid (MCPBA) and trifluoroperacetic acid will carry out the transformation, although in both cases competing reactions lower the yields. With the MCPBA reactions, competing nucleophilic attack by m-chlorobenzoic acid, a contaminant and by-product, produces the amide following rearrangement and loss of CO2.With trifluoroperacetic acid the aqueous conditions employed for reaction cause hydrolysis of the isocyanate. It was found that using equimolar amounts of both these oxidising agents the yields of benzyl isocyanate from the corresponding isothiocyanate were 25 and 5% respectively, compared to a yield of 58% with equimolar DMD. The ozonolysis of benzyl isothiocyanate was also briefly examined, in dichloromethane at 0–5 °C. In this case most of the isothiocyanate remained unchanged, with only small amounts (ca. 3%) of the isocyanate and some benzaldehyde (ca. 10%). It was not clear whether the latter came from direct reaction of the isothiocyanate or from the isocyanate. In summary these studies have shown that organic isothiocyanates can be efficiently converted to isocyanates via oxidation with DMD. These observations lend credence to the proposal that a similar reaction may be catalysed by cytochrome P-450 enzymes, during their reaction with, and inactivation by, isothiocyanates.The non-nucleophilic nature of DMD makes it a good reagent for this transformation, although it has also been observed to much lesser extents with other oxidising agents. Experimental Dimethyldioxirane (DMD) was synthesised according to the method of Mello et al.14 as a solution in acetone. The concentration was determined by NMR analysis of the oxidation of methyl phenyl sulfide in CDCl3.14 The GCMS analyses were carried out using a Hewlett-Packard 5890A gas chromatograph, with SGE BP1 column and a linear temperature gradient from 30 to 300 °C, attached to a Finnigan MAT Incos mass spectrometer. 1H and 13C NMR spectra were obtained using a Varian 2000 FT spectrometer (1H, 300 MHz; 13C, 75.42 MHz) and a Varian Gemini FT spectrometer (1H, 200 MHz; 13C, 50.31 MHz). FTIR spectra were recorded on a Perkin Elmer 1710 spectrophotometer. General Procedure.·In a typical reaction phenyl isothiocyanate (150 mg, 1.11 mmol) was added to a solution of DMD (440 mg, 6.0 mmol) in dry acetone (180 ml) and stirred under nitrogen at room temperature for 15 min.Analysis by GCMS indicated the reaction was complete and isopropylamine (1.12 g, 18.9 mmol) was added at 0 °C. This mixture was then stirred under nitrogen for 1.5 h prior to filtration and concentration under reduced pressure. The crude product was purified by column chromatography on silica, eluting with ethyl acetate–light petroleum (bp 40–60 °C) (30:70).This afforded the product as a white solid (200 mg, 89%), mp 160 °C (lit.,15 156 °C); vmax (nujol)/cmµ1 3350 (NH), 1650 (CO), 700, 750 (Ph); dH (200 MHz; CDCl3) 1.23 [6 H, d, J 7 Hz, CH(CH3)2], 3.97 [1 H, m, CH(CH3)2], 5.53 (1 H, d, J 7 Hz, NHCH), 6.97–7.32 (5 H, m, ArH), 7.54 (1 H, br s, ArNH); dC (50.31 MHz; CDCl3) 23.68 [CH(CH3)2], 42.47 [CH(CH3)2], 120.59, 123.42, 129.54 (aromatics), 139.63 (quat. aromatic), 156.38 (CO); m/z (EI) 178 (M+, 13%), 119 (2, PhNCO+), 93 (100, PhNH2 +), 77 (4, Ph+).N. P. B. is currently a RSE/SOED Research Fellow. N. E. D. wishes to thank St. Leonard’s College and the University of St. Andrews for funding. Received, 20th June 1997; Accepted, 22nd July 1997 Paper E/7/04356B References 1 M.-T. Huang, T. Osawa, C.-T. Ho and R. T. Rosen, Food Phytochemicals for Cancer Prevention (Fruits and Vegetables), ACS Symposium Series 546, American Chemical Society, Washington, 1994. 2 Y. S. Zhang and P.Talalay, Cancer Res., 1994, 54, 1976. 3 G. R. Fenwick, R. K. Heaney and W. J. Mullin, CRC Rev. Food Sci. Nutr., 1983, 18, 123. 4 L. W. Wattenberg, J. Natl. Cancer Inst., 1977, 58, 395. 5 L. W. Wattenberg, Cancer Res., 1992, 52, 2085. 6 C. S. Yang, T. J. Smith and H.-Y. Hong, Cancer Res., 1994, 54, 1982. 7 K. A. Ptashne and R. A. Neal, Biochemistry, 1972, 11, 3224. 8 B. Testa and P. Jenner, Drug Metabolism Rev., 1981, 12, 1. 9 M.-S. Lee, Chem. Res. Toxicol., 1992, 5, 791. 10 M.-S. Lee, Environ. Health Perspect., 1994, 102, 115. 11 B. Kuhn and M. Lieber, Chem. Ber., 1890, 23, 1536. 12 S. M. Parashewas and A. A. Danopoulos, Synthesis, 1983, 8, 638. 13 W. Adam, R. Curci and J. O. Edwards, Acc. Chem. Res., 1989, 22, 205. 14 R. Mello, M. Fiorentino, C. Fusco and R. Curci, J. Am. Chem. Soc., 1989, 111, 6749. 15 J. W. Boehmer, Recl. Trav. Chim. Pays-Bas, 1936, 55, 379. Scheme 3 Table 1 Optimised yields of 1-isopropylureas obtained from the oxidation of organic isothiocyanates with DMD and trapping with isopropylamine. Reaction carried out in acetone at room temperature, with a 5-fold excess of DMD, using 1 mmol of isothiocyanate (see General Procedure) Isothiocyanate Isolated yield of (R·N�C�S) ureaa 5 (%) Benzyl Phenethyl Phenyl Butyl 84 67 89 71 aAll the products gave satisfactory spectral d