|Home | About | Journals | Submit | Contact Us | Français|
An operationally simple, one-pot–two-step, cascade method has been developed to afford quinazolino[1,2,3]-triazolo[1,4]benzodiazepines. This unique, atom-economical transformation engages five reactive centers (amide, aniline, carbonyl, azide, and alkyne) and employs environmentally benign iodine as a catalyst. The method proceeds via sequential quinazolinone-forming condensation and intramolecular azide-alkyne 1,3-dipolar cycloaddition reactions. Substrate scope, multicomponent examples, and mechanistic insights are discussed.
Quinazolinones, benzodiazepines, and triazoles are considered privileged heterocycles and are found in many bioactive compounds.1–3 The quinazolinone core is a ubiquitous pharmacophore appearing in natural and synthetic products.1a Natural products (Figure 1) with the quinazolinone moiety include rutaecarpine (11a–c; a Chinese herbal medication isolated from the fruit of Evodia rutaecarpa with applications in the treatmeent of headache, cholera, and dysentery) and febrifugine (21a; a Chinese herb used in the treatment of malaria). Methaqualone (3, Figure 1) was once a commonly prescribed synthetic drug (on the market in the U.S. since the 1960’s) that exhibits hypnotic and sedative effects.1a,1d Pharmaceuticals derived from the benzodiazepine core, binding to GABAA receptors at the benzodiazepine site, 2a emerged in many marketed drugs during the 1970’s. Two of these are Alprazolam (4, Xanax™)2b and Diazepam (5, Figure 1, Valium™),2c which continue to be prescribed today for the treatment of anxiety and insomnia. Compounds such as 6 (plus analogs, Figure 1), derived from the further elaborated [1,2,3]triazolo[1,4]benzodiazepine skeleton, display activity as protease inhibitors.3
The pharmacological records of these privileged motifs prompted us to develop an atom-economical, one-pot–two-step, cascade method for the production of compounds incorporating the three aforementioned heterocycles. Many procedures have appeared in the literature for the synthesis of quinazolinones; these generally utilize o-aminobenzamides and aldehydes or ketones as starting materials and Bronsted/Lewis acids or Bronsted bases as catalysts to effect imine formation and subsequent ring closure (Scheme 1a).4a–i The quinazolinone synthesis methodology developed by Wang et al. is attractive because environmentally benign, inexpensive molecular iodine is used catalytically (5 mol%) at 50 °C to promote product formation.4d Furthermore, the thermally driven intramolecular azide-alkyne 1,3-dipolar cycloaddition is well-documented, usually employing elevated temperatures (80–120 °C is common) to afford the 1,5-fused-1,2,3-triazole (Scheme 1b).1c,3,5a–c
The method described herein (Scheme 2) draws on these ideas to explore the use of molecular iodine to promote a cascade sequence of reactions. Step one consisting of anilino-keto condensation to form Schiff base 10 and subsequent nucleophilic attack by the amide nitrogen onto the imine to form an aminal (quinazolinone 11). These two iodine promoted condensations pre-organize the alkyne and azide for an intramolecular 1,3-dipolar cycloaddition (step two) to form the complex pentacyclic system 9.6
These transformations were demonstrated by reacting o-amino-N-(prop-2-yn-1-yl)benzamide (7) with o-azidoacetophenone (8a) in MeOH using 10 mol% I2 at 50 °C (open to the atmosphere). This one-pot–two-step cascade reaction delivered pure quinazolino[1,2,3]triazolo[1,4]-benzodiazepine 9, which precipitated from the reaction mixture in 70% isolated yield (Scheme 3).
This method affords many advantageous features. For instance, exploiting inexpensive, non-toxic, air-stable iodine as catalyst4h in a one-pot–two-step cascade process characterizes this as an operationally simple procedure. Additionally, this is a highly atom-economical reaction, engaging five reactive centers (aniline, amide, carbonyl, azide, and alkyne) spanning two starting materials. An other attractive aspect is that each starting material can be accessed in one straightforward step from commercial reagents. o-Amino-N-(prop-2-yn-1-yl)benzamide (7) was prepared by the chemoselective nucleophilic opening of isatoic anhydride (12) with propargyl amine and subsequent decarboxylation of the in situ formed carbamic acid, providing the propargylated amide/free aniline in 88% yield in pure form (Scheme 4a; no column chromatography necessary).7 The second starting material, o-azidoacetophenone (8a: R2 = Me; R3 = H), was obtained in one pot by diazotization of o-aminoacetophenone (13a) and subsequent displacement with sodium azide in high yield (Scheme 4b; 8b–d were also prepared in this fashion; 8b: R2 = Me; R3 = [3,4]dioxole; 8c: R2 = Ph; R3 = H; 8d: R2 = Ph; R3 = p-Cl).8 Substituted isatoic anhydrides are an ideal starting point for this method but, unfortunately, are not generally available. Fortunately, substituted o-amino-N-(prop-2-yn-1-yl)benzamides (15a–c; 15a: R1 = o-OMe; 15b: R1 = m-OMe; 15c: R1 = p-Br) can also be prepared by 1,1′-carbonyldiimidazole (CDI) mediated coupling of substituted anthranilic acids (14a–c) with propargyl amine (Scheme 4c; note that aniline protection is unnecessary).9 o-Azidobenzaldehyde (17) can also serve as the electrophilic carbonyl center in this methodology. Direct displacement of the nitro group of o-nitrobenzaldehyde (16) with sodium azide affords 17 (Scheme 4d).8
The versatility/limitations of this method were evaluated by examining substrate scope (Figure 2, isolated yields listed) using the conditions outlined in Scheme 3. These results show that R-group substitution can have a significant effect on product yield. For instance, electron-donating substituents in both the R1 (o-OMe and m-OMe with respect to the aniline: 19, 20, 22, and 23) and R3 (dioxole: 21–23) positions are well tolerated by this system. One explanation for this -OMe tolerance at R1 is that the donating character of the methoxy group enhances aniline nucleophilicity and in turn facilitates imine formation. It is also interesting to note that when R1 is substituted with a halogen (Br), the yield drops substantially (18 = 30%; the inductively withdrawing character of the Br may have had the opposite effect of the -OMe group). Not surprisingly, when R2 is a phenyl group, the lowest yields are obtained (24 = 25%, 25 = 11%, and 26 = 11%) as a consequence of steric congestion. Furthermore, when an aldehyde (17; R2=R3=H) is employed in place of a ketone, the final quinazolino[1,2,3]triazolo[1,4]benzodiazepine is delivered in good yield (27a = 62%) together with the air- or I2-oxidized side product 27b (16%).
Formation of quinazolinones 27a and 27b was intriguing and prompted us to explore the possibility of converting 27a to 27b under oxidative conditions. Indeed, the facile conversion of 27a to 27b can be achieved with DDQ (Scheme 5) in good isolated yield (70%).
Multicomponent reactions are important because they deliver complex, diverse final products from relatively simple starting materials in one operational step.10 We were interested in investigating the possibility of developing a multicomponent version of this method (Scheme 6); if successful, a starting material synthetic step would be circumvented. Indeed, multicomponent preparations of quinazolinones have been reported, which typically utilize isatoic anhydride (12), ammonium acetate, and an aldehyde.11a–c,1d Two successful multicomponent experiments–delivering quinazolino[1,2,3]triazolo[1,4]benzodiazepines 9 and 27a – are delineated in Scheme 6. In these two examples, propargyl amine, isatoic anhydride (12), o-azidoacetophenone (8a) or o-azidobenzaldehyde (17), and 20 mol% I2 (at 0.2 M MeOH) were simultaneously combined in one pot. The in situ-formed o-amino-N-(prop-2-yn-1-yl)benzamide (7; Scheme 6) went on to deliver the targeted heterocycles, albeit in lower yields compared to the two-component versions.
In conclusion, an atom-economical, cascade, five-centered, one-pot–two-step method has been developed, exploiting the inexpensive, environmentally friendly catalyst I2. The substrate scope has been examined showing that this method is tolerant of various functional groups and aryl substitutions. We have also shown that multicomponent examples of this method are possible. All final compounds have been sent to the Molecular Libraries Small Molecule Repository (MLSMR) with the goal of identifying biologically active molecules.
The authors thank the National Science Foundation (CHE-0910870) and the National Institutes of Health (GM089153) for generous financial support. We also thank Ms. Kelli M. Farber (Department of Chemistry, University of California, One Shields Avenue, Davis, California, 95616) for assistance in the collection of HRMS data.