Heterocyclic Merging of Stereochemically Diverse Chiral Piperazines and Morpholines with Indazoles

Abstract

We report a heterocyclic merging approach to construct novel indazolo-piperazines and indazolo-morpholines. Starting from chiral diamines and amino alcohols, novel regiochemically (1,3 and 1,4) and stereochemically diverse (relative and absolute) cohorts of indazolo-piperazines and indazolo-morpholines were obtained within 6–7 steps. The key transformations involved are a Smiles rearrangement to generate the indazole core structure and a late stage Michael addition to build the piperazine and morpholine heterocycles. We further explored additional vector diversity by incorporating substitutions on the indazole aromatic ring, generating a total of 20 unique enantiomerically pure heterocyclic scaffolds.

Graphical Abstract

Introduction

Heterocycles are ubiquitous in biologically active substances including chemical probes and approved therapeutics. Among them, piperazines and morpholines are regarded as privileged scaffolds owing to their expansive biological activities.^1–8 However, both heterocycles are typically only N-substituted and rarely feature substituents on the ring sp³ carbon atoms. The lacking substitution on their ring carbon atoms potentially attenuates their biological target recognition landscape. The addition of substituents on the carbon atoms might also lead to greater target selectivity. Thus, synthetic methodologies aimed at the efficient and stereoselective carbon formation of piperazine and morpholine compounds are of high utility.^9–14 We have previously reported methods leading to the systematic preparation of diverse enantiomerically pure piperazines and morpholines.^15–21 These studies led us to consider that a structurally unique collection of compounds could be achieved in the context of merging these stereodiverse ring systems with other biologically valuable heterocycles (Scheme 1). Novel ring systems resulting from the formal union of two simpler heterocyclic scaffolds would expand shape diversity and offer additional vectors for substitution. Furthermore, we hypothesize that scaffolds formed from the union of two heterocyclic components will result in biological activities which are unique from their individual heterocycles. Toward this new objective, we demonstrate the concept of heterocyclic merging by synthesizing stereochemically diverse chiral piperazines and morpholines fused to an aromatic indazole nucleus.

Scheme 1.

Schematic representation of heterocyclic merging

Results and Discussion

Our previous studies preparing 2, 5- and 2, 6-substituted piperazines hinted towards a method of synthesizing indazole-fused piperazines and morpholines with a previously unreported merged heterocyclic structure.^{17, 19} Specifically, in efforts to conduct a base promoted aza-Michael reaction to generate the desired piperazine skeleton, we observed the product resulting from a Smiles rearrangement (Scheme 2, B to D).^{22, 23} To form the desired piperazine products, we circumvented the Smiles rearrangement by developing non-basic conditions leading to efficient aza-Michael cyclization. However, it did not escape our notice that the unanticipated indazole N-oxide product still retained the prospect for piperazine ring formation, albeit fused to an indazole ring. Since indazoles are found in a range of biologically active compounds,^24–27 we thought this hybridization could be leveraged to explore novel heterocyclic chemical space. However, exploiting the piperazine formation hinged on identifying an N-protecting group that could effectively be removed to liberate the nucleophilic amino group that would undergo the requisite aza-Michael reaction.

Scheme 2.

Synthesis of 2,5- and 2, 6-disubstituted piperazines. Pg = Cbz or Boc.

To test our hypothesis, we synthesized compound 1a (Scheme 3) from tert-butyl (S)-(1-((2-nitrophenyl)sulfonamido)propan-2-yl)carbamate¹⁷ and performed the Smiles reaction by treating it with DBU in DMF to obtain the N-oxide product 2a. The Boc protected amine of 2a was then unmasked with TFA to yield the aza-Michael precursor 3. A simple basic workup to neutralize excess TFA was performed in anticipation that compound 3 might undergo a spontaneous aza-Michael reaction to produce the cyclized piperazine product 4, but no cyclized products were observed and we questioned whether the conformational rigidity within the indazole ring might be creating a barrier to cyclization. To probe this, we performed the Michael addition by treating 3 with base (K₂CO₃, Cs₂CO₃) at room temperature and then heated the reaction to 100 °C; however, this also resulted only in recovery of the starting material. We next turned to an alternative strategy based on the initial reduction of indazole N-oxide of 2a prior to attempting the intramolecular conjugate addition. Sodium dithionate in aqueous DMSO was used to reduce the N-oxide delivering 5a in good yield.²⁸ Next, 5a was subjected to the Boc de-protection to yield the free amine (not shown), the key aza-Michael cyclization was attempted by treating with base (Cs₂CO₃) at room temperature which gave a modicum of conversion to the piperazine product/s (5% by LCMS) after 16h. Elevating the reaction temperature to 100 °C and stirring for 16h resulted in complete consumption of the starting material and delivery of the desired merged indazolo-piperazine scaffolds as a ~1.5:1 (crude NMR) ratio of piperazines 6a:7a (74% crude yield, Scheme 3). The generation of both diastereomers nonselectively is advantageous for efficiently achieving stereochemical diversity since purification can provide similar quantities of the equally valued products. Unfortunately, in a departure from our previous studies on monocyclic piperazines, the diastereomeric mixture (6a/7a) was not separable on normal silica gel chromatography. We thus resorted to chiral HPLC (Chiralpak IC, heptane/EtOH) to separate the diastereomers (60% isolated yield). The relative stereochemistry of the isolated products was assigned using ¹H and 2D NMR (COSY, HMBC, HSQC, and NOESY; see SI for details).

Scheme 3.

Synthesis of enantiomerically pure indazole fused 1, 3-substituted piperazines. rt = 25–30 °C.

Having achieved the desired route to the merged heterocycles we proceeded to examine the reaction scope. In keeping with the goals of systematic chemical diversity, to achieve stereochemical diversity we started from the enantiomeric R- scaffold 1b, and to probe appendage diversity we used benzyl substituted 1c (Scheme 3). To our delight, the expected products 6b/7b and 6c/7c were produced in good yields and the diastereomers 6c/7c could be separated using silica gel chromatography. It is noteworthy that the larger benzyl substitution did not affect the cyclization reaction or the diastereomeric ratio. The merged final products (6a/6b, 7a/7b) were analyzed using chiral HPLC, confirming that no erosion of the enantiomeric purity took place during the synthesis. In addition, we demonstrated the synthesis of 6b/7b on gram scale by starting from tert-butyl (S)-(1-((2-nitrophenyl)sulfonamido)propan-2-yl)carbamate.

Generating systematic chemical diversity also benefits from varying regioisomeric relationships. Having established an optimal synthetic route of 1,3 substituted piperazines, we pursued the synthesis of 1, 4-substituted merged indazolo-piperazines. Starting from linear precursors 8a/8b (Scheme 4), we performed the Smiles rearrangement and N-oxide reduction to obtain 10a/10b. As described above, the N-Boc removal of intermediate 10a/10b afforded the free amine which was taken on to the base promoted aza-Michael reaction at elevated temperature (100 °C) to yield the 1, 4-substituted indazolo-piperazines in moderate yield (11a/12a and 11b/12b, Scheme 4). In this case, we observed a 1:1 ratio of diastereomers which were separated using chiral HPLC, and then characterized and stereochemically assigned based on ¹H, ¹³C and 2D NMR (COSY, HMBC, HSQC, and NOESY; see SI for details).

Scheme 4.

Synthesis of enantiomerically pure indazole fused 1, 4-substituted piperazines. rt = 25–30 °C.

We decided to extend the concept of heterocyclic merging by using a different saturated scaffold. Our laboratories have previously investigated substituted morpholine scaffolds in a systematically diverse context.^{15, 16} Therefore, we next sought to apply the synthetic methodology to merge the morpholine and indazole heterocycles. Rather than a chiral diamine substrate, the morpholines would require access to analogously functionalized 1,2 amino alcohols that would set the stage for a late stage oxa-Michael to establish the fused morpholine skeleton. Our focus initiated with the synthesis of morpholines having a 1,3-substitution pattern fused to the indazole nucleus. Starting from the commercially available (S)-1-amino-2-propanol 13a, an efficient three step synthesis was applied to form the Smiles rearrangement precursor 15a in good yield (Scheme 5). As expected, on subjection of 15a to DBU, the indazole N-oxide 16a was obtained which was reduced to the oxa-Michael substrate 17a. To facilitate the morpholine ring formation we attempted a one pot TBAF mediated silyl ether cleavage and concomitant oxa-Michael addition to the desired product.²⁹ Unfortunately only free alcohol 18a could be obtained even on heating up to 60 °C. Next, we tried the oxa-Michael addition on pure alcohol 18a by examining different bases (NaH, KO^tBu, DBU), acids (TfOH, TFA) and Lewis acid (MgOTf), and scanning different solvents and temperatures (Scheme 5, Table S1). In most cases, we only observed a partial conversion to the morpholine products. Among the conditions examined, we focused on the one giving the best conversion (~50%) which was DBU in acetonitrile at room temperature (heating did not provide any improvement). Switching the base to triazabicyclodecene (TBD) in acetonitrile produced >95% conversion after 48h at room temperature. To establish the reproducibility, we tested the reaction at 2 mMol scale and were pleased that it provided the merged indazolo-morpholines 19a/20a in a ~1:1 diastereomeric ratio. Separation of the diastereomeric products by column chromatography and assignment of their relative configurations was achieved by ¹H, ¹³C and 2D NMR data (COSY, HMBC, HSQC, and NOESY; see SI for details). To achieve the corresponding antipodes of 19a/20a, the same synthetic sequence was applied starting from (R)-1-amino-2-propanol (13b) to produce 19b/20b (Scheme 5).

Scheme 5.

Synthesis of enantiomerically pure indazole fused 1, 3- and 1, 4-substituted morpholines. rt = 25–30 °C.

Turning our attention to merging the regioisomeric 1,4-morpholines, we started with the enantiopure alaninols 13c and 13d. Following an identical synthetic sequence as described above, the indazolo-1,4-morpholines19c/20c and 19d/20d were delivered in ~1:1 diastereomeric ratios which were separated by silica gel chromatography and assigned on the basis of ¹H, ¹³C NMR and 2D NMR (COSY, HMBC, HSQC, and NOESY; see SI for details).

Once they are generated, stereochemically diverse scaffolds serve as high-value substrates for applying reliable synthetic transformations for appendage substitutions. We finally sought to demonstrate the versatility of heterocyclic merging by incorporating an aryl chloride which could be utilized as a synthetic handle for further substitution reactions. Starting from commercially available -Boc protected diamine 21, sulfonylation followed by alkylation with ethyl 4-bromocrotonate provided the Smiles rearrangement precursor 23. Applying the sequence described in Scheme 1. We arrived at the final diastereomeric products 25/26 (Scheme 6). The chloro-substitution did not affect the diastereomeric ratio or reaction rates. We anticipate that the aryl chloride will prove competent to undergo range of transition metal catalyzed substitution reactions.

Scheme 6.

Synthesis of enantiomerically pure 9-chloro substituted indazole fused 1, 3-substituted piperazines. rt = 25–30 °C.

Conclusion

In conclusion, we describe here heterocyclic merging resulting from the fusion of piperazines and morpholines with indazoles to form the corresponding indazolo-piperazines and indazolo-morpholines. Regiochemically (1,3 and 1,4) and stereochemically diverse (relative and absolute) cohorts of indazolo-piperazines and indazolo-morpholines were generated from readily obtained starting materials. In total, the studies described here resulted in 20 unique scaffolds. In principle, heterocyclic merging can be expanded toward the fusion of other heterocyclic types, providing an efficient synthetic route that allows for stereochemical and regiochemical variation in the assembly. Considering the diversity of known heterocyclic structures, merging provides seemingly unlimited possibilities for populating chemical space with novel drug-like entities. We hypothesize that heterocyclic merging will afford compounds with biological activities that are unique from the simpler scaffolds that comprise their merged structure. This has motivated our current biological evaluation of these compounds in our laboratories, the results of which will be described in future reports.

Experimental Section

General procedure for Smiles rearrangement:

Into a round bottom flask equipped with magnetic stir bar and septum, the starting material (SM) (1 equiv.) was dissolved in dry DMF and was backfilled with nitrogen gas. DBU (2 equiv.) was added to the SM in solvent at room temperature and stirred for 15 mins, where the reaction mixture turned black immediately. The TLC and LCMS showed complete consumption of starting material. The reaction mixture was then diluted with ethyl acetate and washed with cold water (3X). The organic layer was collected and dried over Na₂SO₄ and solvent was evaporated under reduced pressure to afford crude product. The crude product was purified by silica gel chromatography (ethyl acetate/hexanes) to afford pure product.

General procedure for reduction of N-oxide:

To a round bottom flask equipped with magnetic stir bar was added starting material and it was dissolved in DMSO (5 fold). Sodium dithionate (10 equiv.) dissolved in water (1 fold) was then added to the SM in DMSO and mixture stirred at rt for 10 minutes by which time TLC and LCMS showed complete conversion to expected product. The mixture was diluted with ethyl acetate and the washed with ice cold water (3X). The organic layer was then dried over Na₂SO₄ and solvent was evaporated under reduced pressure to afford crude product. The crude product was then purified by silica gel chromatography (ethyl acetate/hexanes) to afford pure product.

General procedure for cyclization via aza Michael addition to afford Indazole fused piperazines:

The starting material was taken in heat dried Teflon coated screw cap vial equipped with magnetic stir bar and dissolved in dry 1,4-dioxane. Dry Cs₂CO₃ (2 equiv.) was added to the solution and mixture heated overnight at 100 °C in preheated metal block. Reaction was quenched by water after TLC and LCMS showed complete consumption of starting material. Product was extracted with EtOAc (3X) and dried over Na₂SO₄ and solvent was evaporated under reduced pressure to afford pure product. The crude product was then passed through column chromatography to afford product as mixture of diastereomers which were then separated by chiral HPLC.

General procedure for cyclization via oxa Michael addition to afford Indazole fused morpholines:

Into a round bottom flask equipped with magnetic stir bar and septum, the starting material (1 equiv.) was dissolved in ACN to which TBD (2.0 equiv.) was added and stirred at room temperature for 48 h after which time TLC and LCMS showed complete conversion to expected mixture of products. The reaction mixture was then diluted with water and extracted with EtOAc (3X). The organic layer was then dried over Na₂SO₄ and solvent was evaporated under reduced pressure to afford crude product. Purification by silica gel chromatography (ethyl acetate/hexanes) provided the pure products.