Polar and e‑Radical Strain-Release Strategies for Accessing Functionalized 2‑Oxabicyclo[2.1.1]hexan-3-ones from Bicyclo[1.1.0]butanes

Abstract

2-Oxabicyclo­[2.1.1]­hexanes (OBHs) have emerged as valuable saturated aromatic replacements, whereas their lactone analogues, 2-oxabicyclo[2.1.1]­hexan-3-ones (OBHOs) remain underexplored, because of limited synthetic access. Herein, we report complementary polar and electrochemical strain-release strategies for the modular synthesis of 1,4-difunctionalized OBHOs from bicyclo[1.1.0]­butane carboxylic acids. The two activation modes provide complementary perspectives toward the chemical space expansion of this underexplored scaffold.

Oxabicyclo­[2.1.1]­hexanes (OBHs) have recently emerged as valuable three-dimensional bioisosteres in medicinal chemistry, offering a compelling alternative to flat, aromatic motifs commonly found in drug molecules. , Their rigid, saturated bicyclic framework introduces defined spatial orientation and increased sp³ character, features that are increasingly associated with improved physicochemical and pharmacokinetic properties, such as enhanced solubility, metabolic stability, and target selectivity. − As bioisosteres, OBHs are particularly attractive as replacing units for substituted phenyl rings or other planar linkers (Scheme A). By mimicking the distance and relative orientation of substituents while disrupting planarity, these scaffolds can modulate ligand–receptor interactions and reduce liabilities such as off-target binding or poor bioavailability. The growing interest in oxabicyclo[2.1.1]­hexanes is also driven by advances in synthetic methodologies that allow their efficient and modular construction. − These developments have enabled broader exploration of OBH-containing libraries, facilitating their incorporation into lead optimization campaigns. While oxabicyclo[2.1.1]­hexanes (OBHs) have been widely validated as bioisosteres in medicinal chemistry, the corresponding cyclic lactones, 2-oxabicyclo[2.1.1]­hexan-3-ones (OBHOs), remain largely underrepresented. − Synthetic methods for their preparation and functionalization are still scarce, leaving an underexplored chemical space with significant potential for drug discovery and intellectual property development. Seminal work by Hall reported the synthesis and polymerization of a limited number of OBHOs, while Tang more recently investigated the anionic ring-opening polymerization of selected derivatives (Scheme B). However, the few reported examples rely on multistep syntheses and are mainly limited to C4-monosubstituted OBHOs. In striking contrast, numerous strategies have been developed for the preparation and validation of OBHs functionalized at both C1 and C4 positions. , Building on our ongoing interest in the development of three-dimensional sp³-rich scaffolds, − we report herein two complementary approaches for the synthesis of C1- and C4-functionalized OBHOs. The first approach relies on a polar strategy based on the reactivity of bicyclo[1.1.0]­butane carboxylic acids (BCBAs) with strong electrophiles (Scheme C). The second, complementary approach involves an electrochemically promoted strain-release functionalization of the same precursors. Inspired by the pioneering studies of Razin and co-workers (Scheme B), we envisioned that treatment of substituted bicyclobutane carboxylic acids 1 (BCBAs) with electrophiles (E) 2 would trigger cleavage of the central C–C σ-bond to form a C–E bond at the α-carbon, followed by intramolecular attack of the carboxylate at the γ-carbon to furnish C1,C4-functionalized OBHOs 3.

1. (A) Importance of Oxabicyclo[2.1.1]­hexanes (OBHs) as Saturated Bioisosteres. (B) Seminal Synthetic Work for the Preparation of OBHOs. (C) Our Present Modular Approach to OBHOs.

After optimization (see the Supporting Information), BCBA 1a (1.0 equiv), 1,3-dibromo-5,5-dimethylhydantoin 2a (DBDMH, 0.7 equiv), and 2,2,6,6-tetramethylpiperidine (TMP, 20 mol %) in acetonitrile at 0 °C afforded OBHO 3aa in high yield within 5 min (Scheme A, entry 1). Variations in TMP loading, temperature, and solvent did not improve the yield (entries 2–5). With optimized conditions in hand, the scope was explored with respect to both BCBAs 1 and electrophiles 2 (Scheme B). DCDMH 2b and DIDMH 2c afforded the corresponding halogenated OBHOs 3ab and 3ac in 88% and 90% yield, respectively. Notably, the reaction was readily scaled up to a 10 mmol scale, affording 3ac in a slightly reduced yield of 81% (see Supporting Information). BCB acids bearing electron-withdrawing (1b) or electron-donating (1c) aryl substituents were also well tolerated, providing 3bb, 3ba, 3ca, 3bc and 3cc in 79–94% yield. Electron-deficient aldehydes could also be employed at room temperature over 16 h. Pentafluorobenzaldehyde (2d) and p-nitrobenzaldehyde (2e) afforded 3ad and 3ae in 78% and 72% yield, whereas less electrophilic 3-pyridinecarboxaldehyde (2f) and p-chlorobenzaldehyde (2g) gave 3af and 3ag in only 30% and 28% yield, indicating that highly electrophilic partners are required. Consistent with this trend, tropylium tetrafluoroborate (2h) proved to be an efficient carbon-centered electrophile. Replacing TMP with NaHCO₃ (1.0 equiv) and using MeCN/H₂O (8:2) at 40 °C enabled reaction of 2h with two BCB acids, furnishing OBHOs 3ah and 3bh in 90% and 84% yield, respectively. We next investigated arynes, generated under mild conditions from ortho-silylaryl triflates, as carbon electrophiles. , Benzyne, generated in situ from 2-(trimethylsilyl)­phenyl triflate and KF/18-crown-6, reacted smoothly with BCBA 1a to form phenyl-substituted OBHO 3ai, as shown by crude ¹H NMR analysis (Scheme C). However, silica gel purification consistently converted 3ai into cyclobutene 4a. We therefore explored 4a as an intermediate toward more densely functionalized OBHOs bearing an additional C5 substituent. Treatment of 4a with DCDMH 2b afforded C1,C4,C5-trifunctionalized OBHO 5ab in 82% yield as an approximately 1:1 mixture of endo/exo diastereoisomers (50:50 dr). DIDMH 2c furnished iodinated 5ac in 85% yield with improved endo/exo selectivity (75:25 dr). To further validate this strategy, cyclobutene 4b was subjected to bromination conditions, affording the corresponding lactone in 86% yield and 65:35 exo/endo ratio. Mechanistically, electrophilic bromination of the alkene was initially proposed to proceed through Markovnikov addition to generate a benzylic carbocation, followed by intramolecular lactonization. Relative configurations of the endo and exo diastereoisomers were assigned by combining NOESY experiments with in silico GIAO NMR simulation (see Supporting Information). , Comparison of calculated and experimental ¹H NMR spectra showed excellent agreement, particularly for coupling constants, enabling unambiguous assignment of endo-5ab and exo-5ab, further supported by NOESY correlations. The iodinated analogues endo-5ac and exo-5ac were assigned by analogy, confirming the endo isomer as the major iodination product. We next investigated whether radical species could initiate the same strain-release sequence and expand the range of accessible OBHOs. Rueping, Cavallo et al. recently reported the electrochemical activation of acids 1, yielding unsubstituted OBHOs through redox-neutral rearrangement and C(4)-bromination as a byproduct. We instead sought to exploit electrochemistry for the direct, diversified functionalization of this scaffold. − A series of BCBAs (1a,1d–g) were subjected to electrooxidative reaction conditions in the presence of different pro-electrophilic partners 6a–c and 6k–p (Scheme A). Generally, the use of 2 equiv of 6, with the application of a constant current of 10 mA (2.0 F/mol₁) in an undivided cell equipped with a Ni cathode and a graphite anode, using a binary THF/ACN (2:1) solvent mixture with LiClO₄ electrolyte at room temperature provided the best conditions to realize the desired transformation. This set of parameters was readily obtained after a reoptimization of the conditions recently disclosed by us for the preparation of C(4)-heterosubstituted 2-oxabicyclo[2.1.1]­hexanes. Interestingly, the use of a base was generally found not pivotal, as cathodic hydrogen evolution is deemed responsible for the scavenging of the acid carboxylic proton. First, we subjected BCBA 1a to bromination, chlorination and iodination conditions in the presence of the respective tetrabutylammonium halides (6a–6c) as a green and readily available electrophilic halogenating species (Scheme C). Products 3aa, 3ab, and 3ac were thus obtained in useful yields (62%–75%) under the application of the standard conditions. To get access to unprecedented C(4)-nitrogenated OBHOs, tetrabutylammonium nitrite 6l and azide 6k were employed, rendering the respective products 3ak and 3al in 34% and 32% yield, respectively. In these cases, the moderate yield observed can be ascribed by a competitive oxidation of 1a (6l) or the low stability of the anodically produced radicals (6k). The introduction of sulfur-based moieties was also effectively achieved using thiophenols 6m–6o as pro-electrophiles (products 3am–3ao, 41%–65% yield), requiring the use of 2,6-lutidine as a base. Whose addition was necessary to generate thiolate anions, more prone to anodic oxidation with respect to the parent thiols. Finally, a formal oxygenation of the C(4)-position of OBHOs 3 was undertaken with the use of TEMPO 6p, for which a small set of differently substituted BCBAs 1 was investigated, resulting in the formation of products 3ap, 3dp–3gp in 43%–62% yield.

2. Polar Electrophile-Induced Synthesis of OBHOs.

^a Reaction performed under optimized reaction conditions: 1 (0.2 mmol), 2 (0.14 mmol), TMP (0.04 mmol) in ACN (0.1 M with respect to 1), 0 °C, 5 min.

^b Reaction performed under the optimized reaction conditions, except at room temperature for 16 h instead of 0 °C for 5 min.

^c Reaction performed under modified reaction conditions: 1 (0.2 mmol), 2 (0.2 mmol), NaHCO₃ (0.2 mmol), ACN/H₂O (4:1, 0.1 M with respect to 1), 40 °C, 16 h.

3. Electrochemical Synthesis of OBHOs.

^a Oxidation potentials of 1a, 6a–6c, 6k, 6l, 6n potassium salt, 6p and THF. Oxidation potentials collected by CV responses measured in a 3-electrode cell, ACN solvent, [analyte] = 2.10 mM, TEABF₄ electrolyte (0.10 M). See the Supporting Information for details.

^b Reactions were carried under air in the Electrasyn 2.0 apparatus (undivided cell), 1 (0.1 mmol), 6 (0.2 mmol), LiClO₄ (0.3 mmol), THF/ACN (2:1, 0.033 M with respect to 1), C­(+)||Ni(−), CCE, I = 10 mA, 2.0 F/mol ₁ , room temperature (rt).

^c Reaction run at I = 5.0 mA.

^d Reaction run using 2,6-lutidine (2.0 equiv).

^e Reaction run using 1.0 F/mol₁ instead of 2.0 F/mol₁ and at −30 °C instead of room temperature.

^f Reaction run using 2,6-lutidine (1 equiv), 6b or 6l (1.3 equiv).

^g Reaction run using 6b or 6l (0.6 equiv), I = 25 mA, 2.0 F/mol ₆ .

Unexpectedly, when running the chlorination reaction at −30 °C with a smaller excess of 6b (1.3 equiv), the formation of the desired compound 3ab was observed to occur alongside products 7ab and 8ab (1.5:2.5:1 ratio), as a consequence of THF oxidation–capture–ring opening events terminated by the final addition of the chloride anion (vide infra). In order to take advantage of this side pathway and improve the reaction selectivity toward compounds 7 or 8, two protocols deviating from the previously optimized conditions were identified (see the Supporting Information). Importantly, the progressive reduction of the equivalents of TBACl 6b was coherently observed to favor the intervention of THF as electroactive (7ab and 8ab) and/or nucleophilic (8ab) species. For the generation of 7ab, the presence of 2,6-lutidine (1 equiv) and 1.3 equiv of 6b were identified as optimal, in terms of yield and selectivity (7ab:8ab:3ab = 4:0:1, 52% yield for isolated 7ab). While the formation of 8ab was maximized in the presence of 0.6 equiv of 6b and when a current intensity of 25 mA was applied (8ab:7ab:3ab = 3.5:1:0, 47% yield for isolated 8ab). Assuming that products 7 and 8 arise from the competitive oxidation of the solvent, despite its higher anodic oxidation potential (vide infra) compared to the pro-electrophilic anion, we reasoned that the nitrite anion could also serve as a suitable nucleophile under these reaction conditions. Gratifyingly, replacement of 6b with tetrabutylammonium nitrite 6l afforded the corresponding products 7al (31% yield) and 8al (30% yield), which were isolated and fully characterized. Notably, compounds 7 and 8 demonstrate the versatility of this methodology for the one-step synthesis of OBHOs bearing a length-tunable arm amenable to further functionalization. Mechanistically, we propose that the preferential oxidation of electroactive species 6 occurs, over the anodic reaction of 1a, as suggested by cyclovoltammetry analysis (Scheme B) and literature precedents. ,, In the case of 6a–6c and 6k–6o, an electrophilic radical is generated and subsequently trapped by the strained bond of 1a, leading to intermediate A (Scheme D). Subsequent oxidation to stabilized carbocation B allows for the final intramolecular ring closure toward the desired products 3. In the case of TEMPO 6p on the other hand, a cation is produced by anodic oxidation, leading to the same intermediate B. Finally, in the case of products 7 and 8, anodic oxidation of THF is assumed to take place preferentially with anions having high oxidative potential (i.e., Cl^–, 1.03 V vs Fc/Fc⁺ and NO₂ ^–, 1.15 V vs Fc/Fc⁺). Moreover, although the oxidation potential of THF is higher (1.5 V vs Fc/Fc⁺) compared to 6b, 6l, or 1a (1.11 V vs Fc/Fc⁺), its far superior concentration is deemed likely to facilitate anodic oxidation (Scheme E). The so-formed THF radical cation is proposed to be intercepted by 1a, leading to intermediate C and, after subsequent oxidation and ring closure, to intermediate D. Chloride or nitrite anion-promoted ring-opening of D would lead to products 7. Conversely, in analogy with THF cationic polymerization processes, products 8 require another molecule of THF to be added first (E) before being terminally quenched by chloride or nitrite anion. In conclusion, we have developed complementary polar and electrochemical strain-release strategies for the modular synthesis of C1-, C4-, and C5-functionalized 2-oxabicyclo[2.1.1]­hexan-3-ones from readily available bicyclo[1.1.0]­butane carboxylic acids. Together, these activation modes considerably broaden the range of compatible reaction partners, providing efficient access to this underexplored class of bridged lactones. We anticipate that this platform will facilitate the future development and diversification of OBHO-based molecular scaffolds for applications in synthetic and medicinal chemistry.

The data underlying this study are available in the published article and the Supporting Information.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.6c03497

• Experimental procedures, optimization tables, characterization data (¹H, ¹³C, and ¹⁹F NMR and HRMS) for new chemical compounds, and further references provided by the authors (PDF)

∇.

Authors S.U. and G.M. contributed equally to this work. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.