A convenient two-step synthesis of Coenzyme Q 1

Abstract

A convenient method for the preparation of Coenzyme Q₁ from cheap and readily available 3,4,5-trimethoxytoluene is developed. Coenzyme Q₁ is synthesized in a moderate yield by a two-step procedure involving the key reaction of an allyl bromide with Coenzyme Q₀ through a redox chain reaction. The reaction is efficient and can be used for the synthesis of other Coenzyme Q compounds.

Introduction

Coenzyme Q₁₀ (CoQ₁₀; Figure 1) is an isoprenoid quinone compound ¹ and plays a pivotal role in the electron transport chain in respiratory processes. ² CoQ₁₀ is a natural antioxidant that scavenges free radicals. ³ It is widely used in the treatment of cardiovascular disease and mitochondrial disorders. ⁴ Coenzyme Q₁ (CoQ₁; Figure 1) is an important fragment of the Coenzyme Q (CoQ) series which are active in the electron transport and oxidative phosphorylation processes in mitochondria. CoQ₁ also acts as a key intermediate in the synthesis of higher CoQ analogues. ⁵

Figure 1.

Structures of CoQ₁₀ and CoQ₁.

There have been several methods published for the preparation of CoQ₁, most of which involve a Lewis acid-catalyzed reaction between an allylic alcohol and hydroquinone, followed by oxidation to the quinones. Hegedus and Waterman ² and Sato et al. ⁶ synthesized CoQ₁ by reaction of π-allylic nickel complexes with quinones in 26% yield (Scheme 1, eq 1). Yamago et al. ⁷ reported a radical-mediated synthesis of substituted quinones with organotellurium compounds. ⁸ However, these reactions were quite sensitive to the reaction conditions, the key reagents π-allylnickel bromide complex and the requisite organotellurium reagent were difficult to prepare. Tabushi et al. ⁹ reported a β-cyclodextrin-catalyzed allylation–oxidation of hydroquinone to form CoQ₁ in 11% yield (Scheme 1, eq 2). Recently, Chen ⁵ and Borioni et al. ¹⁰ started from 3,4,5-tetramethoxytoluene (TMT) to obtain CoQ₁ in multiple steps (Scheme 1, eq 3). Unfortunately, all these methods generally gave low yields and complex by-products. Therefore, a general and practical method for efficient CoQ₁ synthesis is highly desirable. Here, we report a two-step synthesis of CoQ₁ by starting from 3,4,5-trimethoxytoluene with a total yield of 60% (Scheme 1, eq 4).

Scheme 1.

Various methods for the syntheses of CoQ _n : (a) previous work and (b) this work.

Results and discussion

First, a single-step synthesis of CoQ₀ is shown in Table 1; this oxidation reaction of TMT is conducted in acetic acid at 50°C in less than 2 h and without using any metal catalyst. This environmentally friendly procedure is based on the oxidant as an oxygen atom donor, and the acidic solvent acetic acid plays an important role in the transformation. The traditional method employing 30% H₂O₂ as an oxidant gave a yield of 50% (Table 1, entry 1). The use of Na₂S₂O₈ and (NH₄)₂S₂O₈ improved the reaction yield (Table 1, entries 3 and 4). The best yield was obtained using K₂S₂O₈ as an oxidant to afford the desired product CoQ₀ in 85% yield (Table 1, entry 2). However, when cerium ammonium nitrate (CAN) was used as the oxidant, we did not observe any product CoQ₀ (Table 1, entry 5).

Table 1.

Single-step synthesis of CoQ₀.

Entry	Oxidant	Yield (%)
1	30% H2O2	50
2	K2S2O8	85
3	(NH4)2S2O8	70
4	Na2S2O8	60
5	CAN	0

CAN: cerium ammonium nitrate.

Reaction conditions: TMT (0.01 mol), oxidant (1.5 equiv.), 2 h, open air.

Bold values represents the best condition.

Inspired by Xu and Li’s ¹¹ work on the alkylation of p-quinones by a redox chain reaction, we tried to synthesize CoQ₁ by the allylation of CoQ₀ with 1-bromo-3-methyl-2-butene 1; the results are shown in Table 2. Diethyl 1,4-dihydro-2,6-dimethyl-3,5-pyridinedicarboxylate (Hantzsch ester) was selected as an initiator which can accelerate the alkylation according to the literature, ¹² and the reaction works with many Lewis acids as catalysts in dichloromethane as the solvent at room temperature (r.t.). Typical Lewis acid catalysts were screened in the reaction; however, AlCl₃, ZnCl₂, and FeCl₃ did not catalyze the reaction (Table 2, entries 1–3). The solvents used for this reaction was crucial, acetone, tetrahydrofuran (THF), CH₃CN, or toluene as the solvent led to low yields of CoQ₁ (Table 2, entries 5–8). On the basis of these screening studies, the optimum conditions employed BF₃·Et₂O as the catalyst and dichloromethane as the solvent.

Table 2.

Redox chain reaction for the synthesis of CoQ₁.

Entry	Lewis acid	Solvents	Yield (%)
1	AlCl3	CH2Cl2	N.R.
2	ZnCl2	CH2Cl2	N.R.
3	FeCl3	CH2Cl2	N.R.
4	BF3·Et2O	CH2Cl2	70
5	BF3·Et2O	Acetone	8
6	BF3·Et2O	THF	32
7	BF3·Et2O	CH3CN	20
8	BF3·Et2O	Toluene	10

CoQ₁: coenzyme Q₁; N.R.: no reaction; THF: tetrahydrofuran; r.t.: room temperature.

Reaction conditions: CoQ₀ (0.01 mol), compound 1 (0.01 mol), Hantzsch ester (1 mmol), Lewis acid (0.01 mol), r.t., N₂ atmosphere.

Bold values represents the best condition.

Conclusion

In summary, we have developed a convenient synthetic protocol for the preparation of CoQ₁ from cheap and readily available 3,4,5-trimethoxytoluene (TMT) in two steps. The overall yield of CoQ₁ is 60%. The intermediate CoQ₀ was also obtained in 85% yield in the first step. The second redox chain reaction between allyl bromide and CoQ₀ provided a one-step procedure for the direct introduction of allyl groups on quinones in good yield. The reaction is efficient, clean, and the workup is easy. This method may find use for the synthesis of other CoQ compounds.

Experimental section

All reactions were monitored by thin-layer chromatography (TLC; SiO₂, petroleum ether (PE; b.p. 50°C–70°C)/ethyl acetate (EtOAc), 5:1). Melting points were measured on BUCHI Melting Point M-565. Nuclear magnetic resonance (NMR) and mass spectra were recorded on a Bruker Avance III-HD 400 NMR and a TripleTOF mass spectrometer, respectively. Potassium persulfate, ammonium persulfate, Hantzsch ester, and BF₃·Et₂O were purchased from Adamas, P. R. of China, and used without further purification.

General method for the preparation of 2,3-dimethoxy-5-methyl-1,4-benzoquinone (CoQ₀)

3,4,5-Trimethoxytoluene (1.82 g, 10 mmol) was dissolved in a mixture of acetic acid (10 mL) and catalytic H₂SO₄ (0.01 mL), then a solution of the oxidant (15 mmol) was added dropwise over 10 min. The mixture was stirred and heated at 50°C for 1 h. After cooling, the reaction mixture was extracted with CH₂Cl₂ (3 × 10 mL). The combined organic phases were washed with H₂O and NaHCO₃, then dried over anhydrous Na₂SO₄ and evaporated under reduced pressure. The residue was purified by silica gel column chromatography (PE/EtOAc, 5:1) to give CoQ₀.

CoQ₀, red-colored needles, m.p. 55°C–58°C (Lit. ¹³ 57°C–59°C).

¹H NMR (400 MHz, CDCl₃): δ = 6.44 (q, J = 1.7 Hz, 1H), 4.02 (s, 3H, OCH₃), 4.00 (s, 3H, OCH₃), 2.04 (d, J = 1.6 Hz, 3H, CH₃).

¹³C NMR (101 MHz, CDCl₃): δ = 184.4 (C =O), 184.2 (C =O), 145.0, 144.8, 144.0, 131.2, 61.2 (OCH₃), 61.1 (OCH₃), 15.4 (CH₃).

MS (ESI): m/z = 205 [M + Na]⁺.

The spectroscopic data are in accord with the literature. ¹³

General method for the preparation of 2,3-dimethoxy-5-methyl-6-(3-methyl-2-butenyl)-1,4-benzoquinone (CoQ₁)

1-Bromo-3-methyl-2-butene (1; 1.49 g, 0.01 mol), Hantzsch ester (0.25 g, 1 mmol), and CoQ₀ (1.82 g, 0.01 mol) were dissolved in dichloromethane (10 mL) under a nitrogen atmosphere. After stirring for 30 min, a solution of BF₃·Et₂O (1.2 mL, 0.01 mol) in dichloromethane (1 mL) was added, and the solution was stirred at r.t. for 2 h. The mixture was extracted with CH₂Cl₂ (3 × 10 mL). The combined organic phases were washed with H₂O and brine, then dried over anhydrous Na₂SO₄ and evaporated under reduced pressure. The residue was purified by silica gel column chromatography (PE/EtOAc, 8:1) to give CoQ₁.

¹H NMR (400 MHz, CDCl3): δ = 4.32 (t, 1H, J = 7.0 Hz, C=CH), 3.96 (s, 3H, CH3O), 3.94 (s, 3H, CH3O), 3.12 (d, 2H, J = 7.0 Hz, CH2), 2.14 (s, 3H, CH3), 1.75 (s, 3H, CH3), 1.65 (s, 3H, CH3).

¹³C NMR (101 MHz, CDCl₃): δ = 180.0 (C =O), 175.6(C =O), 144.1, 143.0, 142.7, 132.6, 126.5, 123.4, 60.5 (OCH₃), 60.3 (OCH₃), 30.4, 29.1, 25.4, 15.7(CH₃).

MS (ESI): m/z = 251 [M + H]⁺.

The spectroscopic data are in accord with the literature. ⁵