A chemoselective and continuous synthesis of m-sulfamoylbenzamide analogues

A chemoselective and continuous synthesis of m-sulfamoylbenzamide analogues Arno Verlee1, Thomas Heugebaert1, Tom van der Meer2,3, Pavel I. Kerchev2,3, Frank Van Breusegem2,3 and Christian V. Stevens*1 Full Research Paper Address: 1Department of Sustainable Organic Chemistry and Technology, Faculty of Bioscience Engineering, Ghent University, Campus Coupure, Coupure Links 653, B-9000 Ghent, Belgium, 2Department of Plant Systems Biology, VIB, Ghent University, Technologiepark 927, B-9000 Ghent, Belgium and 3Department of Plant Biotechnology and Bioinformatics, Ghent University, Technologiepark 927, B-9052 Ghent, Belgium Email: Christian V. Stevens* - * Corresponding author Open Access Beilstein J. Org. Chem. 2017, 13, 303–312. doi:10.3762/bjoc.13.33 Received: 27 November 2016 Accepted: 03 February 2017 Published: 16 February 2017 This article is part of the Thematic Series "Automated chemical synthesis". Guest Editor: I. R. Baxendale © 2017 Verlee et al.; licensee Beilstein-Institut. License and terms: see end of document. Keywords: flow chemistry; medium-throughput synthesis; m-sulfamoylbenzamide analogues Abstract For the synthesis of m-sulfamoylbenzamide analogues, small molecules which are known for their bioactivity, a chemoselective procedure has been developed starting from m-(chlorosulfonyl)benzoyl chloride. Although a chemoselective process in batch was already reported, a continuous-flow process reveals an increased selectivity at higher temperatures and without catalysts. In total, 15 analogues were synthesized, using similar conditions, with yields ranging between 65 and 99%. This is the first automated and chemoselective synthesis of m-sulfamoylbenzamide analogues. Introduction Small molecules are commonly used for their ability to regulate or assist in different biological processes. Typically, drug development starts with the screening of large libraries of relatively similar compounds, where only milligrams of material are needed for primary testing. Upon identification of a primary hit, the synthetic protocol must then be quickly expanded to tens of grams for early in vivo toxicity studies and hundreds of grams for further toxicology studies and clinical trials [1]. These swiftly changing requirements appear throughout the clinical development of active pharmaceutical ingredients (APIs) and place specific and conflicting burdens on synthetic protocols. An early synthesis must be extremely fast and flexible, as current high-throughput compound screening takes less than one week for a set of 10,000 compounds [2], which is far beyond the current synthetic capabilities. Once a suitable hit is identified on the other hand, the synthetic prerequisites change completely, and a robust and scalable protocol is needed. Over the past few years, flow chemistry has emerged as a potential solution to these conflicting prerequisites [3-11]. Flow processing is suitable for automation, thus allowing the fast synthesis of com- 303 Beilstein J. Org. Chem. 2017, 13, 303–312. pound libraries, but as opposed to, e.g., combinatorial chemistry, the developed protocols are directly useful for scale-up. A class of small molecules where these principles can apply for are m-sulfamoylbenzamides. These compounds proved to be effective against Huntington’s and Parkinson’s disease [12-14]. They inhibit the Sirtuin 2 (SIRT2) deacetlyse protein (Figure 1, AK-1, AK-7) resulting in improved motor skills [12,13,15]. Furthermore, m-sulfamoylbenzamide analogues (Figure 1, C2-8) are able to suppress polyglutamine (polyQ) aggregation [14], which is a major cause of neurodegeneration in Huntington’s disease. Although there are numerous reports available on the study of these analogues, an automated, chemoselective alternative to the synthesis is not yet available. shown by Yang et al. [18]. By transferring this reaction to a multistep flow set-up, we envisioned an improved chemoselectivity. This phenomenon is not unusual for flow chemistry. Typical batch reactions are mixed by stirring; however, perfect homogeneity is not immediately obtained. Ideal mixing conditions can only be achieved with microreactors or micromixers [22]. The small diameters of these microreactors lead to almost ideal mixing conditions [23-26], resulting in an improved chemoselectivity. Furthermore, the use of an automated process leads to the possibility to produce libraries of compounds in a fast manner. In addition, an alternate biocompatible and water miscible solvent would result in a flexible and automated chemoselective synthesis, delivering stock solutions suitable for initial testing at the outlet of the reactor. Results and Discussion Development of a continuous-flow process Figure 1: m-Sulfamoylbenzamides as Sirtuin 2 inhibitors (SIRT2) or suppressor of polyglutamine aggregation (polyQ). The most common synthetic approach starts from m-(chlorosulfonyl)benzoic acid [15-17]. This synthetic approach is a twostep procedure and therefore needs two subsequent work-up steps, limiting the yield and resulting in a more time-consuming synthetic approach. Yang et al. [18] reported a one-pot synthetic strategy for m-sulfamoylbenzamide analogues starting from m-(chlorosulfonyl)benzoyl chloride. In this study the difference in reactivity between the sulfonyl and aroyl chloride is exploited resulting in a chemoselective synthesis for these analogues. The yields varied between 46% and quantitative yield, relatively short reaction times were required and dichloromethane was used as solvent. The coupling of carboxylic acids with amines in flow through a benzotriazole activation [19], or with immobilized reagents as for the synthesis of grossamide [20] is already known. However, we wanted to use m-(chlorosulfonyl)benzoyl chloride since this can be synthesized in one single step. Furthermore, acid chlorides show a high reactivity [21] making m-(chlorosulfonyl)benzoyl chloride an ideal starting material as was Although a continuous-flow process shows many advantages compared to batch reactions, there are some difficulties which should be overcome or be avoided. A general concern is the clogging of the channels. There are numerous reports about handling solids in flow. For example, the use of ultrasound [2732] can reduce the particle size of the precipitates, and preventing the clogging of the small channels. A second example is the Coflore agitating cell reactor [32]. This type of reactor uses transverse mixing motions which keeps the solids in suspension, and prevents clogging. The Coflore reactor was successfully used for the synthesis of N-iodomorpholinium hydroiodide salt [33]. However, it takes specialized machinery and time to develop a system which can pump slurries. Therefore, a reduction in the formation of solids is preferable. Furthermore, we wanted to avoid the use of dichloromethane as solvent and use a biocompatible and water miscible alternative. A series of initial batch reactions were performed to e (...truncated)