1-(2-Benzyl-1,5-dimethyl-6,7,8-trioxabicyclo[3.2.1]octan-2-yl)ethan-1-ol

Abstract

The title compound, 1-(2-Benzyl-1,5-dimethyl-6,7,8-trioxabicyclo[3.2.1]octan-2-yl)ethan-1-ol, was synthesized for the first time by the selective reduction in keto ozonide under the action of the strong reducing agent LiAlH₄. The product was characterized by NMR, IR, HRMS, and elemental analysis.

1. Introduction

Cyclic organic peroxides are attractive compounds for drug discovery. Their importance was illustrated by the 2015 Nobel Prize in Medicine awarded to Youyou Tu for the discovery and development of the natural antimalarial peroxide artemisinin. Recent studies have demonstrated that cyclic synthetic peroxides exhibit antimalarial [1,2,3,4,5,6,7], anticancer [8,9,10,11,12,13], antifungal [14,15], antiparasitic, [16,17,18,19,20] and antiviral [21,22,23,24,25] activities. Notably, synthetic peroxide arterolane (Figure 1), which contains the ozonide core, is used in medical practice for the treatment of malaria [26]. Moreover, arterolane shows in vitro activity against α-coronavirus NL63, and β-coronavirus SARS-CoV-2 [27].

In light of the previous information, the expansion of the structural diversity of peroxides is an important task for the development of biologically active compounds based on them. [28,29,30,31,32,33,34,35,36,37]. One of the solutions to this problem is the synthesis of peroxides with functional groups and their transformation with the preservation of the O–O bond. This approach can provide access to hybrid molecules. The synthesis of hybrid molecules is a modern trend in medicinal chemistry [38,39,40].

Herein we report the synthesis of bridged ozonides containing a hydroxyl functional group via selective reduction in keto ozonide by a strong reducing agent LiAlH₄. This is very unusual because the ozonide cycle has proven to be moderately stable under such harsh conditions. Typically, peroxides are not resistant to reducing agents and decompose to form alcohols [41,42,43].

2. Results and Discussions

Transformations of peroxides with preservation of the peroxide cycle are nontrivial processes. Under the action of acid or base, peroxides can undergo transformations with cleavage of the O-O bond, for example, the Baeyer–Villiger, Criegee, Udris–Sergeev/Hock or Kornblum DeLaMare reactions and related processes [44]. Usually, the expansion of the structural diversity of peroxides is based on the following sequence: synthesis of peroxide with an ester group → hydrolysis of the ester group with the formation of a carboxyl group → modification of the carboxyl group (Scheme 1) [45].

Another promising approach to expand the diversity of peroxides is the introduction of a hydroxyl group by reduction in the ester or keto group. However, here it is necessary to keep a balance in which the peroxide cycle will be preserved, and the ester or keto group will be converted to alcohol. In this study, we found the key to ozonides 3a + 3b containing the OH-functional group (Scheme 2).

The starting keto ozonides 2a + 2b were synthesized by phosphomolybdic acid-catalyzed peroxidation of β,δ’-triketone 1 [46]. The molar ratio of 2a:2b was 1.5:1. Then, the resulting mixture of keto ozonides 2a + 2b was treated with LiAlH₄ in THF at −22 °C to give the title compounds 3a + 3b which were formed as the mixture of diastereomers in a 60% isolated yield with the molar ratio of 3a:3b = 2:1. The stereochemical assignment was based upon NMR analysis with 2D correlation spectroscopic techniques (HSQC, HMBC and NOESY). Surprisingly, the ozonide cycle was found to be moderately resistant to LiAlH₄. This fact may open up additional possibilities in the modification of peroxides for the development of effective biologically active compounds. Structure of 3a + 3b was confirmed by NMR, IR spectroscopy, HRMS, and elemental analysis.

3. Materials and Methods

Caution! Although we encountered no difficulties in working with peroxides, precautions, such as the use of shields, fume hoods, and the avoidance of transition metal salts, heating, and shaking, should be taken whenever possible.

¹H and ¹³C NMR spectra were taken with a Bruker AM-300 machine (Bruker AXS Handheld Inc., Kennewick, WA, USA) (at frequencies of 300 and 75 MHz) in CDCl₃ solution with TMS as the standard. J values are given in Hz. The high-resolution MS spectrum was measured on a Bruker microTOF II instrument (Bruker Daltonik Gmbh, Bremen, Germany) using electrospray ionization (ESI). The elemental analysis was performed on a 2400 Elemental Analyzer (Perkin ElmerInc., Waltham, MA, USA). The TLC analysis was carried out on silica gel chromatography plates Macherey-Nagel Alugram UV254; sorbent: Silica 60, specific surface (BET) ~500 m²/g, mean pore size 60 Å, specific pore volume 0.75 mL/g, particle size 5–17 µm; binder: highly polymeric product, which is stable in almost all organic solvents and resistant towards aggressive visualiz ation reagents. Chromatography of peroxides was performed on silica gel (0.040–0.060 mm, 60 A, CAS 7631-86-9). The solvents and reagents were purchased from commercial sources. A solution of H₂O₂ in Et₂O (6.0 M) was prepared by the extraction with Et₂O (5 × 100 mL) from a 35% aqueous solution (100 mL) followed by drying over MgSO₄. Then, part of Et₂O was removed in the vacuum of a membrane vacuum pump at 20–25 °C and titrated iodometrically [15,47].

Synthesis of 1-(2-benzyl-1,5-dimethyl-6,7,8-trioxabicyclo[3.2.1]octan-2-yl)ethan-1-ol (3a + 3b) (Figure 2) (Supplementary Materials).

LiAlH₄ (0.055 g, 1.45 mmol) was added to the solution of ozonides 2a + 2b (0.200 g, 0.724 mmol) in (dry) THF (10 mL) with stirring under an argon atmosphere at −22 °C. The reaction mixture was stirred at −22 °C for 3h. After that, an aqueous solution of NaOH (5M, 5 mL) was added with stirring to reaction mixture at −22 °C. Then, the reaction mixture was warmed to room temperature, CHCl₃ (30 mL) and H₂O (15 mL) were added. The organic layer was separated, and the aqueous layer was extracted with CHCl₃ (3 × 15 mL). The combined organic layers were washed with water (5 mL), dried over MgSO₄, and then concentrated in vacuo. Ozonides 3a + 3b were isolated by column chromatography on silica gel (EA—hexane, 10:1, v/v). Ozonides 3a + 3b: 0.121 g, 0.43 mmol, 60% yield, colorless oil, R_f = 0.30 (TLC, PE:EA, 5:1).

3a: ¹H NMR (300.13 MHz, CDCl₃), δ: 1.34 (d, J = 6.5 Hz, 3H, (H¹⁰)), 1.56 (s, 3H, (H⁷)), 1.58 (s, 3H, (H¹)), 1.74–2.30 (m, 4H, (H⁴ ,H⁵)), 2.34 (br.s. 1H, OH), 2.96 (d, J = 13.6 Hz, 1H, (H⁸)), 3.10 (d, J = 13.6 Hz, 1H (H⁸)), 3.62–3.72 (m, 1H, (H⁹)), 7.28–7.32 (m, 5H, (H^{o−, m−. p−})). ¹³C NMR (75.48 MHz, CDCl₃), δ: 19.5 (C¹), 19.6(C¹⁰), 20.7(C⁷), 21.6(C⁴), 31.4(C⁵), 36.8(C⁸), 47.8(C³), 69.4(C⁹), 108.9(C⁶), 114.2(C²), 126.6(C^p−), 128.2(C^m−), 130.9(C^o−), 138.2(Cⁱ⁻).

3a:¹H NMR (300.13 MHz, CDCl₃), δ: 1.26 (d, J = 6.5 Hz, 3H (H¹⁰)), 1.28–1.38 (m, 1H, (H⁴)), 1.53 (s, 3H, (H⁷)), 1.70 (s, 3H, (H¹)), 1.74–2.30 (m, 3H, (H⁴,H⁵)), 2.33 (br.s. 1H, OH), 2.71 (d, J = 13.6 Hz, 1H, (H⁸)), 3.05 (d, J = 13.6 Hz, 1H, (H⁸)), 3.88–3.97 (m, 1H, (H⁹)), 7.28–7.32 (m, 5H, (H^{o−, m−. p−})).¹³C NMR (75.48 MHz, CDCl₃), δ: 19.7(C¹), 20.9(C⁷), 21.1(C¹⁰), 24.8(C⁴), 32.8(C⁵), 40.4(C⁸), 46.8(C³), 71.0(C⁹), 108.6(C⁶), 113.4(C²), 126.5(C^p−), 128.1(C^m−), 131.1(C^o−), 137.6(Cⁱ⁻).

HRMS (ESI-TOF): m/z [M + Na]⁺: calculated for [C₁₆H₂₂NaO₄]⁺: 301.1410; found: 301.1409. Anal. Calcd for C₁₆H₂₂O₄: C, 69.04; H, 7.97. Found: C, 69.08; H, 8.00. IR (KBr): 3561, 2945, 1722, 1603, 1496, 1453, 1382, 1226, 1141, 1082, 944, 913, 827, 732, 705, 662 cm⁻¹.

4. Conclusions

In this work, we presented a synthesis of the previously unknown compound 3a + 3b (1-(2-benzyl-1,5-dimethyl-6,7,8-trioxabicyclo[3.2.1]octan-2-yl)ethan-1-ol) in 60% isolated yield. The ozonide cycle is surprisingly moderately resistant to LiAlH₄.