Synthesis of xanthohumol and xanthohumol-d3 from naringenin

A six-step synthesis of xanthohumol (1a) and its d₃-derivative (1b) from easily accessible naringenin is reported. The prenyl side chain was introduced by Mitsunobu reaction followed by the europium-catalyzed Claisen rearrangement and base-mediated opening of chromanone gave access to an α,β-conjugated ketone system. Compound 1b was used as an internal standard in stable isotope dilution assays of 1a in two Polish beers.

A six-step synthesis of xanthohumol and its d₃-derivative from easily accessible naringenin is reported.

Xanthohumol (1a, Scheme 1A) is a naturally occurring prenylated chalcone produced by lupulin glands in female inflorescences of hop plants.¹ In recent years, 1a has attracted significant attention due to its vast range of biological activities including cancer-preventive, antioxidative, anti-inflammatory, and antiviral.^2–8 Such properties combined with low toxicity to the human body make 1a a prospective therapeutic agent, diet supplement, or ingredient of cosmetics.⁹ Although 1a was isolated from natural sources in 1913,¹⁰ the first synthesis of this compound was reported as late as in 2007.¹¹ Several other syntheses have been reported since then, but only minor improvements have been achieved.^12–14Open in a separate windowScheme 1The background of the study.Bioactive compounds labeled with stable isotopes (deuterium, carbon-13) are widely applied in metabolomic studies for tracking metabolic pathways and as internal standards in stable isotope dilution assays.^15,16 Deuterated compounds are also considered as attractive drug candidates due to the influence of the kinetic isotope effect on pharmacokinetics.^17–19 Although approaches to ¹³C-enriched xanthohumol^20,21 and hydrogen/deuterium exchange in 1a²² were reported, no scalable and cost-effective synthesis of the deuterium-labeled derivative of 1a (i.e.1b) has been disclosed to date.Two main challenges have to be faced in the synthesis of 1a: (i) construction of a pentasubstituted aromatic ring containing a prenyl side chain and (ii) selection of suitable protecting groups for phenols. In the case of (i), phloroglucinol is used as a precursor and an acyl-substituent is introduced by Friedel–Crafts acylation with a subsequent Claisen–Schmidt condensation. The prenyl side-chain is introduced by Mitsunobu alkylation, followed by Claisen rearrangement. In the case of (ii), acid-sensitive alkoxymethyl protecting groups, removable under conditions in which 1a does not cyclize to isoxanthohumol (2a), are used most often (Scheme 1A).In this study, we have developed a synthetic approach for the formation of 1a and its deuterated analog 1b. We envisioned that both 1a and 1b can be directly obtained by the base-promoted chromanone ring-opening of 2a or 2b, which in turn can be obtained from easily accessible naringenin (3) (ca. 1 $/1 g) via two-step prenylation and Williamson etherification of the phenolic OH (Scheme 1B). The use of 3 as the starting material is beneficial as only one prenyl substituent has to be introduced.The synthetic route leading to 1a and 1b is depicted in Scheme 2. Our synthesis commenced from naringenin (3), which was selectively converted to diester 4 (Ac₂O, pyridine). O-Alkylation of 4 under Mitsunobu conditions (3-methyl-2-butene-1-ol, Ph₃P, DIAD), followed by the catalytic Claisen rearrangement of 5 (Eu(fod)₃, 1,2-dichloroethane, 80 °C) afforded prenyl-derivative 6. Notably, performing the latter reaction in 1,2-dichloroethane above its boiling point was superior in comparison to earlier reports.^23–26 Alkylation of 6 (CH₃I, Ag₂O or CD₃I, Ag₂O) afforded 7a/7b in good yields. An alternative approach to 7b involving the alkylation of phenolic OH under Mitsunobu conditions (CD₃OD, Ph₃P, DIAD) required a large excess of reagents and afforded the product in moderate yield. Basic hydrolysis (KOH, MeOH) of esters afforded isoxanthohumols 2a/2b. Although the chromane ring was stable during the hydrolysis, it could be opened under more harsh conditions (DBU, DMF, 70 °C),^27,28 leading to 1a/1b in good yields after a mild acidic workup.Open in a separate windowScheme 2The synthetic route to 1a and 1b.With 1a and 1b in hand, we investigated their MS-fragmentation patterns in electrospray ionization in positive and negative ion modes. The MRM transitions were found by an automatic procedure and they are listed in † for details). In positive ion mode, the most intensive product ions in fragmentation of 1a were ions with m/z values of 178.9, 299.0, 113.0, and 150.9. Corresponding ions with m/z +3 values can be found in the fragmentation of 1b. On the other hand, in the negative ion mode, the same product ions are observed both in the case of 1a and 1b, indicating that the CH₃/CD₃ groups were lost during fragmentations.The MRM transitions of 1a and 1b

Ionization mode	1a		1b
	Precursor ion	Product ion	Precursor ion	Product ion
ESI(+)	355.0	178.9	358.0	182.0
		299.0		302.0
		113.0		115.9
		150.9		107.9
		93.0		154.0
ESI(−)	353.0	119.1	356.0	119.1
		233.0		236.0
		295.1		295.2
		218.2		175.0
		175.0		218.1
		189.2		168.2

Open in a separate windowOne of the criteria for an effective internal standard is the coelution of the labeled and non-labeled compounds during the HPLC analysis. This is particularly important in case of deuterium-labeled compounds as with the increase in the number of deuterons in the molecule, retention times may be extended. The retention times of 1a and 1b under different HPLC conditions are listed in