Enantioselective Sulfoxidation of Thioanisole by Cascading a Choline Oxidase and a Peroxygenase in the Presence of Natural Deep Eutectic Solvents

A bienzymatic cascade for selective sulfoxidation is presented. The evolved recombinant peroxygenase from Agrocybe aegeri-tra catalyses the enantioselective sulfoxidation of thioanisole whereas the choline oxidase from Arthrobacter nicotianae provides the H 2 O 2 necessary via reductive activation of ambient oxygen. The reactions are performed in choline chloride-based deep eutectic solvents serving as co-solvent and stoichiometric reductant at the same time. Very promising product concentrations (up to 15 mM enantiopure sulfoxide) and catalyst performances (turnover numbers of 150,000 and 2100 for the peroxygenase and oxidase, respectively) have been achieved. Chiral sulfoxides are widely found in drugs, chiral intermediates as well as chiral ligands in catalysis. the HNO 3 -promoted sulfoxidation aromatic sulfides, synthetic methodologies to the portfolio transition catalysis,

A bienzymatic cascade for selective sulfoxidation is presented. The evolved recombinant peroxygenase from Agrocybe aegeritra catalyses the enantioselective sulfoxidation of thioanisole whereas the choline oxidase from Arthrobacter nicotianae provides the H 2 O 2 necessary via reductive activation of ambient oxygen. The reactions are performed in choline chloride-based deep eutectic solvents serving as co-solvent and stoichiometric reductant at the same time. Very promising product concentrations (up to 15 mM enantiopure sulfoxide) and catalyst performances (turnover numbers of 150,000 and 2100 for the peroxygenase and oxidase, respectively) have been achieved.
Chiral sulfoxides are widely found in drugs, chiral intermediates as well as chiral ligands in catalysis. [1] Since the first report by Märcker in 1865 on the HNO 3 -promoted sulfoxidation aromatic sulfides, [2] many further synthetic methodologies have been added to the portfolio including transition metal catalysis, peracids and hydrogen peroxide or hypohalites. [3] Nevertheless, the issue of overoxidation to the (mostly undesired) sulfone remains to be solved satisfactorily. More importantly, enantioselective methods, highly desired especially for products in the pharmaceutical and agrochemical sectors are still scarce. [3c,4] To address the latter challenge, biocatalysis is of particular interest due to the often inherent enantioselectivity of enzymes. [5] Therefore, it is not very astonishing that chiral sulfoxidation (of benzyl methyl sulfide, thioanisole) is a common model reaction for monooxygenases and peroxidases. [6] One issue that needs to be addressed in biocatalytic sulfoxidation reactions is the poor solubility of most thiols in aqueous media, preferred in biocatalysis. Thioanisole, for example, is soluble in to approximately 4 mM. As a consequence, the majority of biocatalytic reports on thioanisole sulfoxidation use such low reagent concentrations, which is not attractive neither from an economic nor an environmental point-of-view. [7] In recent years, (natural) deep eutectic solvents ((NA)DES) have enjoyed an increasing popularity in biocatalysis. [8] Particularly, their often stabilizing effect on enzymes but also their solubilisation of less hydrophilic reagents are of interest. Especially choline chloride (ChCl) is the basis of many (NA)DES. Recently, we have demonstrated that ChCl can be used not only as solvent component but also as sacrificial electron donor for the reductive activation of molecular oxygen by the enzyme choline oxidase from Arthrobacter nicotianae (AnChOx) to promote perhydrolase- [9] and peroxygenase-catalysed oxyfunctionalisation reactions. [10] Inspired by the recent contribution by Brenna and co-workers on the peroxygenase-catalysed sulfoxidation of thiols [6] we became interested in expanding this reaction concept to ChCl-based DES exploiting the dual functionality of these DES as co-solvent and as co-substrate ( Figure 1 and Figure 2).
In a first set of experiments we used AnChOx in various DES to generate H 2 O 2 , which then spontaneously oxidised thioanisole ( Figure 1).
Especially urea-and propanediol-containing DES gave the highest conversions. Interestingly, these were not necessarily those solvents in which AnChOx exhibits its highest activity. [10] In carbohydrate-based DES, the overall sulfoxidation activity was the lowest. Possibly, carboxylate groups activate the in situ formed H 2 O 2 for the spontaneous sulfoxidation reaction. [11] Overall, this procedure appeared to be a viable method for the preparation of racemic sulfoxides. However, especially if biological activity of the sulfoxide products is desired, enantioselective sulfoxidation is preferred. Therefore, we also evaluated peroxygenase-catalysed, enantioselective oxygenation of thioanisole ( Figure 2). [6] Pleasingly, in the bienzymatic system, enantiomerically pure (R)-sulfoxide was formed in all experiments (ee > 99 %, Ta-  ble S4). Nevertheless, conversions in this system never exceeded approx. 50-60 %.
As the best results were obtained with Urea-containing DES, we further characterised the influence of several relevant reaction parameters of the efficiency of the bienzymatic sulfoxidation reaction.
First, we evaluated the influence of the water content in the DES reaction mixture (Figure 3). In accordance to our previous findings with bienzymatic hydroxylation and epoxidation reactions, [10] no sulfoxidation activity was found in neat DES and at least 25 % (v/v) of water (buffer) were necessary. This is, however, in stark contrast to previous findings on the perhydrolase-catalysed reactions in DES. [9,12] In the latter case, significant enzyme activity was found also in neat DES. For all further experiments, a 1 : 1 volumetric ratio of buffer and DES was used.
Increasing the concentration of the H 2 O 2 -generation catalyst (AnChOx) influenced the overall product formation in a saturation-type manner (Figure 4).
Interestingly, in buffer, the highest product formation was observed already at AnChOx concentrations as low as 3 μM whereas in the presence of 50 % (v/v) of ChCl-Urea the overall productivity steadily increased over the entire AnChOx concentration range investigated. This may be attributed to the reduced activity of AnChOx in the presence of DES. [10] In any case, it can be asserted that the H 2 O 2 generation rate (under these conditions) was overall rate limiting (vide infra).

Communications
Previous studies using hydrolases in DES had established a stabilising effect of DES on the biocatalyst. [12] We therefore also compared the influence of the reaction temperature on the bienzymatic sulfoxidation reaction in buffer and DES-media ( Figure 5). Very much to our surprise, no stabilising effect of the DES was observed.
Investigating the pH dependency of both reaction systems ( Figure 6) we found no sharp optimum; the highest reaction rates were observed in alkaline media.
Finally, we also varied the rAaeUPO concentration to investigate its effect on the initial rate and the final product concentration (Figure 7). Interestingly, varying the peroxygenase concentration between 50 nM and 200 nM had no significant effect in both reaction media, confirming our previous assumption that the rAaeUPO-catalysed sulfoxidation is not the overall rate-limiting step of the bienzymatic cascade. Interestingly, while the rate of the overall reaction in the DESsystem was approximately half of the buffer system (in accordance with the lower AnChOx activity under these conditions Figure 4) the final product concentration in all experiments was essentially identical (13.9 mM � 1.5 mM corresponding to 55 % conversion, Table S5-6).
Overall, in the current contribution we have demonstrated that sulfoxidation reactions are feasible in ChCl-based DES. By using AnChOx as in situ H 2 O 2 generation catalyst, we could use the DES not only as a (co-)solvent but also as a reagent for the reaction. Recently, there is an increasing interest in this dual use of DES. [9][10]13] Using AnChOx alone, full conversion of thioanisole was achieved albeit producing racemic products. If combined with rAaeUPO, enantiomerically pure sulfoxide was obtained exclusively, albeit at the expense of significantly reduced yields. Today, we are lacking a plausible explanation for this observation. One possible reason may be inhibition of AnChOx by the   (R)-sulfoxide, but further experiments will be necessary to identify the molecular reason, alleviate it and bring the production system to preparative applicability. It is worth mentioning that in none of our experiments the corresponding overoxidation product (sulfone) was observed.
Another interesting outcome of this study is that in case of the proposed bienzymatic sulfoxidation reaction, no specific benefit of DES was found. The presence of ChCl-based DES did not improve the stability or the activity of the biocatalysts. This is in stark contrast to previous findings on hydrolases in DES [9,12,14] or even the AnChOx-rAaeUPO-system if used for hydroxylation and epoxidation reactions. [10] This discrepancy calls for further studies aiming at more systematic investigation and understanding of the effects of DES on biocatalysts!