|Home | About | Journals | Submit | Contact Us | Français|
The synthesis of small organic molecules as probes for discovering new therapeutic agents has been an important aspect of chemical-biology. Herein we report a reagent-based, diversity-oriented synthetic (DOS) strategy to probe chemical and biological space via a “Click, Click, Cyclize” protocol. In this DOS approach, three sulfonamide linchpins underwent cyclization protocols with a variety of reagents to yield a collection of structurally diverse S-heterocycles. In silico analysis is utilized to evaluate the diversity of the compound collection against chemical space (PC analysis), shape space (PMI) and polar surface area (PSA) calculations.
The synthesis of small organic molecules as probes for discovering new therapeutic agents has been an important aspect of chemical-biology.1 Essential to this goal are two fundamental features i) the production and access to libraries of skeletally diverse small molecules and ii) biological evaluation and identification of new probes.2 Such small molecules have had a dramatic effect in recent years providing invaluable insight into biological targets and the development of therapeutic agents for curing disease.3 The generation of an open-data, high-throughput screening environment of diverse small-molecule libraries has provided both a number of new molecular probes as well as a novel insight into unmined chemical space.2 In contrast to natural product-based targeted libraries premised on improving the biological activity of the corresponding natural product, diversity-oriented-synthesis (DOS) derived libraries aim to discover new molecules that exhibit biological effects beyond those associated with the natural product. In this regard, DOS has emerged as a powerful strategy in the generation of structurally complex and skeletally diverse small molecules.4
Synthetic protocols combined with rational design of small molecules based on structural diversity, complexity and inherent physiochemical properties, has emerged as a rich area in chemical biology.5 The ability to generate a collection of small molecules that combine not only skeletal and peripheral complexity from a central building block, while remaining diverse in comparison to each other has been a challenging goal. Libraries synthesized utilizing a DOS approach have been generated through a number of approaches. Seminal papers by Evans in 1988 and Schreiber 2000 reported the generation of substructural motifs as ligands for diverse receptors.6 Recently, notable examples of DOS strategies have been reported by Spring, Park and Shair.3,4 One of the more notable strategies employing both reagent-based7 and functional group pairing attributes is the build/couple/pair (B/C/P) paradigm pioneered by Schreiber and coworkers.8
Recently, a number of reports of sultams, the cyclic analogs of sulfonamides, have emerged demonstrating a broad-spectrum bioactivity (Fig. 1), yet not “preordained bioactivity” as is the case with targeted, medicinally active natural products. In particular, reports include anti-HIV activity,9 antidepressant activity,10 inhibitors of RSV,11 selective tumour necrosis factor,12 and metalloproteinase.13 In addition to this potent biological profile, sultams and their sulfonamide precursors possess a number of advantageous chemical properties. This potency, when coupled with their unique chemical properties, elevates sultams as promising candidates for drug discovery.
Despite these attributes, general strategies towards the synthesis of sultam libraries are lacking in the literature.14 To address this challenge, we report a reagent-based DOS strategy termed “Click, Click, Cyclize” en route to structurally diverse sultams from common sulfonamide linchpins.15,16 In this strategy, skeletal diversity is incorporated into each small molecule via a chosen orthogonal reagent used to cyclize each linchpin. As in functional-group pairing approaches, this DOS strategy provides a pathway to a collection of diverse sultams.
Taking the aforementioned approach into hand, three unique sulfonamide linchpins 2, 9 and 15 were designed to yield a collection of sultams utilizing the aforementioned “Click, Click, Cyclize” protocol.15 In this regard, linchpin 2 was synthesized via a “Click” mono-sulfonylation of ethylenediamine with 2-bromobenzene sulfonamide 1, followed by a second “Click” sulfonylation with tosychloride (TsCl) to generate the desired linchpin 2 in high yield (Scheme 1).17 Utilizing a variety of reagents, five sultams and bis-sulfonamides (3–7) were readily synthesised. Initial cyclization of linchpin 2 was achieved via a microwave-assisted, Cu-catalyzed, intramolecular N-arylation yielding the corresponding sultam 3 in 70%.18 Alternatively, cyclization of linchpin 2 with either 1,2-dibromoethane or 1,3-dibromopropane provided the desired piperazine 4 and diazepine 6 in good yield. In contrast, cyclization of linchpin 2 with carbonyl diimidazole gave the corresponding imidazolidin-2-one 5 in 92% yield. Finally, allylation followed by RCM yielded 7 in 88% yield, via a “click-cyclize” 2-step protocol.
Building on these results, sulfonamide linchpin 9 was synthesized via sulfonylation of 2-bromobenzylamine 8 with 2-chloroethanesulfonyl chloride followed by an aza-Michael reaction with n-butylamine (Scheme 2).19 It was envisioned that cyclization with four commercially available reagents would yield four skeletally diverse sultams (10–13).
Sultam 10 was synthesised via cyclization of linchpin 9 utilizing methyl 3-bromo-2-(bromomethyl)propionate. Utilizing the same cyclization protocol as for the synthesis of 4 and 5, sultams 11 and 12 were synthesized via cyclization of linchpin 9 with 1,2-dibromoethane or 1,3-dibromopropane, respectfully. Finally, sultam 13 was synthesized utilizing a microwave-assisted, copper-catalyzed N-arylation protocol.
Utilizing a previously developed Cu-catalyzed, N-arylation protocol, sulfonamide linchpin 15 was readily synthesized on scale from sulfonylchloride 14 (Scheme 3).18 The first cyclization route, utilized a CDI cyclization protocol yielding sultam 16 in good yield. In an attempt to synthesize sultam 18, linchpin 15 was treated with 3-bromo-2-(bromomethyl)propionate in DMF at 60 °C. However, the desired product 18 was not isolated and instead sulfonamide 17 was isolated in 78% yield. Finally, linchpin 15 readily underwent cyclization with 1,2-dibromoethane to yield bicyclic sultam 19 in good yield.
To evaluate the diversity that is contributed by this collection of molecules and hence their associated chemical descriptors, in silico algorithms were utilized to evaluate the S-heterocycles reported. With each molecule possessing its own set of unique descriptors, every small molecule has a discrete point in chemical space. Therefore, the more chemical space probed by a collection of molecules, the greater the associated diversity. This metric of diversity in chemical space can be represented by a principle component (PC) analysis (Fig. 2). In order to gauge the chemical diversity of the sultams and sulfonamides (3–7, 10–13 and 16–19) reported herein, we plotted them in a chemical space plot corresponding to a set of five BCUT descriptors relative to 3770 FDA approved compounds as reported in ZINC database (Fig. 2).20 This plot demonstrated that this collection of molecules both covered a significant area of chemical space but also the compounds did not cluster together according to the corresponding linchpin they were derived from.
Building on this analysis, sultams and sulfonamides (3–7, 10–13 and 16–19), were plotted according to the normalized principal moment of inertia (PMI) formalism of Sauer and Schwartz, in order to gauge the shape-based distribution (Fig. 3).21 The PMI plot is a rapid and visual way to demonstrate diversity corresponding to the area of shape space covered by a collection of molecules. This is a significant property for a collection of molecules to possess, as broad biological activity has been correlated to shape space.21 Hence, screening collections possessing a high degree of molecular shape diversity increases the chances of a broad range of biological activity. Each molecule was aligned to principal inertial moment axes in SYBYL,22 and the normalized PMI values were computed via a program developed in-house (available upon request to the authors). With this plot in hand, Fig. 3 demonstrates a large coverage of shape space for sultams and sulfonamides (3–7, 10–13 and 16–19). Of note is the coverage by compounds 10–13 (red data points) derived from linchpin 8 further demonstrating the diversity achieved utilizing a “Click, Click, Cyclize” approach.
In addition to diversity in both chemical and shape space, polar surface area of a small molecule is a key feature in terms of diverse bioactive molecules involved in ligand-receptor binding. Rigid scaffolds bearing diverse polar surface areas interact differently with various key interactions such as hydrogen bonding, electrostatic and other non-covalent interactions. This is further exemplified by reports that demonstrate the diverse biological activity associated with small molecules with diverse polar surface areas resulting from different orientations of heteroatoms.4a In this regard, polar surface area distribution of sultams 3, 4, 5 and 7 were plotted (Fig. 4). Comparison among the four further demonstrates the degree of diversity achieved from linchpin 2 utilizing a “Click, Click, Cyclize” protocol. Surface electrostatic profiles were calculated by projecting the Gasteiger-Marsili charge distribution onto a Connolly surface generated via the MOLCAD tool in SYBYL.22
In summary, we have utilized a “Click, Click, Cyclize” strategy to synthesize a collection of skeletally diverse heterocycles in a DOS approach. Three distinct sub-sets of molecules were prepared via the cyclization of sulfonamide linchpins with a variety of reagents. In silico analysis using a variety of metrics demonstrates the degree of diversity from this collection in regards of chemical space, shape space and polar surface area.
This publication was made possible by the Pilot-Scale Libraries Program (P41 GM076302), the National Institutes of General Medical Sciences (KU Chemical Methodologies and Library Development Center of Excellence, P50 GM069663) and by Grant Number P20 RR015563 from the National Center for Research Resources, a component of the National Institutes of Health, and the State of Kansas. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the NCRR or NIH.
†Electronic supplementary information (ESI) available: General experimental details, experimental data for compounds 2–17 and 19 and 1H and 13C NMR spectra of compounds. See DOI: 10.1039/b927161a