An Efficient, Stereocontrolled and Versatile Synthetic Route to Bicyclic Partially Saturated Privileged Scaffolds

Herein, we describe the development of a simple, high yielding and stereocontrolled strategy for the synthesis of a series of triazolopiperazines and other biologically relevant fused scaffolds from optically active amino acids. This route was applied to the sysnthesis of 22 sacffolds containing new, previously inaccessible vectors and used to access a novel analogue of ganaplacide. “Privileged” scaffolds, have an inherent ability to bind and modulate the activity of protein targets, making them highly favourable scaffolds for drug discovery and as such have been utilised in a variety of discovery programs. 1 Many synthetic drugs also share common scaffolds, which led to Black’s suggestion that “The most fruitful basis for the discovery of a new drug is to start with an old drug.” 2 The key to the construction of privileged scaffold libraries for drug discovery is the development of reactions of broad scope with stereocontrol and that allow access to multiple vectors in a scaffold. 3 present A simple, high yielding and scalable synthesis for diastereoselective access to privileged fused bicyclic heteroaromatic scaffolds.

Herein, we describe the development of a simple, high yielding and stereocontrolled strategy for the synthesis of a series of triazolopiperazines and other biologically relevant fused scaffolds from optically active amino acids. This route was applied to the sysnthesis of 22 sacffolds containing new, previously inaccessible vectors and used to access a novel analogue of ganaplacide.
"Privileged" scaffolds, have an inherent ability to bind and modulate the activity of protein targets, making them highly favourable scaffolds for drug discovery and as such have been utilised in a variety of discovery programs. 1 Many synthetic drugs also share common scaffolds, which led to Black's suggestion that "The most fruitful basis for the discovery of a new drug is to start with an old drug." 2 The key to the construction of privileged scaffold libraries for drug discovery is the development of reactions of broad scope with stereocontrol and that allow access to multiple vectors in a scaffold. 3 Heterocycles with increased levels of three-dimensionality present an underexplored and underrepresented class of privileged scaffolds. The incorporated heteroatoms enable potential binding interactions, and the limited structural flexibility of the cyclic scaffold reduces the entropic binding penalty resulting in an increased possibility to form productive binding interactions. 4 As a result there has been a recent surge in the development of methodologies to access threedimensional heterocyclic scaffolds. 5 Partially saturated bicyclic piperazine-based privileged scaffolds are present in a number of drugs for a variety of clinical indications including treatments of malaria, hyperglycaemia, sex-hormone related disorders and cancer (Figure 1a). 6 Despite the promising and varied applications for these privileged scaffolds, there exists no consistent synthetic strategy for accessing them. Published routes include annulating the 5membered heterocycle onto the 6-membered piperazine ring; selectively reducing the pyrazine ring of the fully aromatic triazolopyrazine and forming a 6-membered lactam on the 5membered heterocycle. 6,7 These varied routes lack the ability to generate scaffolds with enantioselective substitutions around the piperazine ring. We describe a highly modular, efficient and versatile approach to synthesising fused bicyclic heteroaromatic privileged scaffolds starting from optically pure amino acids ( Figure 1b). We anticipated that assembling the 5-membered heterocycle first would provide the scaffold around which the partially saturated piperazine ring could be constructed. Alkylation of the bridging nitrogen would set up the scaffold for a one-pot deprotection and reductive amination with cisdiastereoselective control driven by the pre-set amino acid chiral centre (see SI, 1.2.6. for model).
Initial studies were carried out towards the synthesis of 5,6,7,8-tetrahydro-[1,2,4]triazolo[1, 5-a]pyrazine scaffolds. To this end, a number of exemplar amino acids 1a-g were converted into the corresponding amino hydrazides 3a-g via the corresponding amino ester intermediates 2a-g. These transformations proceeded in near quantitative yields on multi- gram scales. In parallel, the required imidate building blocks 5af were formed from the corresponding nitriles 4a-f using the Markovnikov transformation 8 (Scheme 1). The combination of the amino hydrazide (3a-g) and imidate (5a-f) building blocks in the presence of acetic acid gave triazoles 6a-p (Table 1). 9 Subsequent alkylation of the triazoles with α-bromoketones gave the cyclisation precursors 7a-p, which were set up for cyclisation to form the 5,6-bicyclic scaffolds 8a-p. Treatment of intermediates 7a-p with catalytic palladium dihydroxide and ammonium formate enabled the deprotection of the pendant nitrogen, unveiling it for the reductive amination. A 3:1 ratio of methanol and water was used, the addition of water was though to increase the rate of imine reduction and thus supress the undesired ketone reduction. 10 The final scaffolds (8a-p) were generated in 51-91% and >20:1 d.r. Significantly, the methodology enabled the incorporation of functionalities that provide vectors for compound elaboration, including amino, alcohol and sulphur groups in the final triazole-based 5,6-bicyclic compounds (8a-p). In addition, common functionality used in drug discovery were included, such as trifluoromethyl and cyclopropyl groups. Due to the reductive conditions of the final step, alkene moieties were reduced to simple alkyl chains. To access a primary amine substituent, it was necessary to use a nitro group to mask this reactive moiety. Finally, for access to sulphur containing substituents it was necessary to increase the catalytic amount of palladium dihydroxide to excess such that the poisoning effect of the sulphur could be overcome. The desired cisgeometry between the two piperazine substituents of the 5,6bicyclic scaffolds 8a-p was conclusively demonstrated by X-ray crystallography of compound 8n (see SI). The X-ray structure also confirmed the conservation of the stereochemistry of the chiral centre derived from the starting amino acid. 11 Thus we have developed an efficient, high yielding synthetic strategy towards the synthesis of triazolopiperazines with stereocontrol and substitution at 3 of a possible 4 positions, leaving the amine unsubstituted for derivatisation. To expand the scope of this work and explore the potential to generate closely related privileged scaffolds with novel vectors in a stereocontrolled manner, other 5-membered heterocycles were used to form the fused ring system. To this end a synthetic strategy was sought which continued to utilise optically pure amino acids as an effective chiral source and our methodology to drive the stereo-control (Scheme 2).
The methodology was further applied to the synthesis of 5,7-bicycles (Scheme 3). The 5-membered heteroarene (6a) was alkylated with β-haloketones to furnish the cyclisation precursor 18 in an excellent 93% yield. The 5,7-heterocycle 19 was achieved in a 20% yield and >20:1 diastereoselectivity, which was a pleasing result given the slower rate of cyclisation for larger rings. The by-product of the reaction was identified as 20 resulting from reduction of the uncyclised ketone.
Synthesis of final scaffolds without an R 3 substituent was explored (see Table 1). Diverging from the previous synthesis after formation of heterocycle 6a, the alkylation was carried out with ethyl bromoacetate to furnish ester 21 (Scheme 4).

Compound
Step (    Reduction, followed by oxidation with IBX furnished the hemiaminal 23. Application of the reductive palladium dihydroxide conditions yielded the desired amine 24 in a good yield. The importance of this work is highlighted by the lead optimisation work carried out in the development of ganaplacide by Novartis, a drug currently in Phase II clinical trials for the treatment of malaria. The SAR studies previously conducted were limited to double substitutions of the same group at each position around the piperazine ring due to the methodology available at the time, preventing exploration of various substitution patterns, functional groups and stereoselective control of the substitution (See SI). 23 These limitations could be overcome by using our approach. To demonstrate its applicability, the synthesis of a novel analogue with a single substituent in the R 2 -and R 3 -positions was completed using our methodology (Scheme 5).
The key intermediate (32) of a ganaplacide analogue was achieved with substitution patterns and functional groups not accessible with the previous synthetic routes utilised. Intermediate 32 could be easily elaborated into the ganaplacide analogue 50 in 3 steps (see SI). Our modular synthetic route has the potential to access a wide range of novel analogues of ganaplacide than were previously inaccessible, such as those with single substitution patterns. In conclusion, we have developed a simple, highly scalable route to a series of fused piperazine-and diazepane-based scaffolds, which would enable the efficient construction of fragment-like, lead-like or drug-like screening libraries. We illustrated the utility of this work in the synthesis of a previously inaccessible analogue of ganaplacide.