Synthesis of 7,2′-Dihydroxy-4′,5′-Dimethoxyisoflavanone, a Phytoestrogen with Derma Papilla Cell Proliferative Activity

This paper reports a concise and scalable method for the synthesis of the phytoestrogen 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone 1 via an optimized synthetic route. Compound 1 was readily obtained in 11 steps and 11% overall yield on a gram scale from commercially available 3,4-dimethoxyphenol. The key features of the synthesis include the construction of the deoxybenzoin unit through a sequence of Claisen rearrangement, oxidative cleavage, and aryllithium addition and the efficient synthesis of the isoflavanone architecture from highly functionalized 2-hydroxyketone.


Introduction
Phytoestrogens are naturally occurring dietary compounds. They are found in a wide variety of foods, including fruits, vegetables, and grains [1][2][3][4]. These plant-derived compounds and their metabolites are structurally and functionally similar to those of mammalian estrogens, such as estradiol, and thereby exhibit weak estrogenic and anti-estrogenic effects by binding to estrogen receptors (ERs) [3,5,6]. There is increasing evidence that edible phytoestrogens have numerous health benefits, including anticancer, antioxidant, anti-inflammatory, hepatoprotective, antibacterial, and antiviral activities, that are closely related to the prevention and treatment of various types of cancers, cardiovascular diseases, osteoporosis, neurological diseases, diabetes/obesity, immune system dysfunction, menopause symptoms, and skin aging conditions, such as alopecia [1- 3,5,7,8]. Hence, phytoestrogens are promising non-steroidal estrogenic compounds and potential alternatives to estrogen replacement therapy (ERT) for human healthcare.
Phytoestrogens are classified into several subgroups, such as flavonoids, isoflavonoids, and lignans, based on their structural motifs and biosynthetic pathways. Furthermore, isoflavonoids vary among subclasses, such as isoflavones, isoflavanones, isoflavans, pterocarpanes, and coumestans [1, 2,9,10]. Owing to their structural rarity, as well as their unique and diverse range of biological functions, the chemistry related to the isolation, structural elucidation, and bioactivities of isoflavanones, along with their therapeutic applications, has been extensively investigated [10][11][12].
In 2003, 7,2 -dihydroxy-4 ,5 -dimethoxyisoflavanone (1) was first isolated from the heartwood of Dalbergia louvelii R. Viguier (Fabaceae), which was used as a folk medicine to treat bilharzia and malaria in Madagascar [13]. More recently, 1 and its structurally related phytochemicals possessing diverse substitution patterns and oxidation states were isolated from the bark of Dalbergia oliveri Prain, a traditional Thai medicine used for the treatment of to treat bilharzia and malaria in Madagascar [13]. More recently, 1 and its structurally related phytochemicals possessing diverse substitution patterns and oxidation states were isolated from the bark of Dalbergia oliveri Prain, a traditional Thai medicine used for the treatment of chronic ulcers in Southeast Asia (Figure 1) [14]. Isoflavanone 1 exhibits potent hair growth effects on immortalized dermal papilla cells (iDPCs). In a cell proliferation assay, 1 induced significant cell proliferative activity (54.1% at 10 μM, EC50 = 8.83 μM), which was more potent than that of Minoxidil (20.6% at 10 μM), a widely used medication for the prevention and treatment of hair loss [14]. Moreover, isoflavanone 1 induced the anagen phase of the hair cycle in a mouse model via subcutaneous (SC) injection [15]. In 2018, Kim et al. reported a sophisticated approach for the synthesis of 1. However, this synthetic route does not yield a large amount of isoflavanone 1, although the synthetic steps are relatively short [15]. Practically, the preparation of naturally occurring compounds in large quantities is a highly formidable task because the large-scale collection and isolation of natural products from natural sources are restricted. Therefore, we have been exploring an efficient and scalable synthetic strategy for the large-scale synthesis of 1 to identify the mechanism of its hair growth effects through in vivo animal model studies, as well as its wide range of health benefits. Herein, we report the synthesis of the natural isoflavanone 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1).

Results and Discussion
Our approach for the synthesis of phytoestrogen 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1), as shown in Figure 2, includes the efficient construction of an isoflavanone framework and a gram-scale synthetic pathway. Isoflavanone 1 was obtained from 2-hydroxyketone 2 via the annulation of the deoxybenzoin skeleton and the sequential deprotection of the two masked phenols in the final stage. The deoxybenzoin unit of 2 can be constructed through a sequence of aryllithium additions to arylacetaldehyde 3 and the subsequent oxidation of the resulting alcohol. It was expected that 3 could be easily prepared from commercially available 3,4-dimethoxyphenol 4 via Claisen rearrangement and the subsequent oxidative cleavage of the terminal alkene to introduce a crucial acetaldehyde side chain. In 2018, Kim et al. reported a sophisticated approach for the synthesis of 1. However, this synthetic route does not yield a large amount of isoflavanone 1, although the synthetic steps are relatively short [15]. Practically, the preparation of naturally occurring compounds in large quantities is a highly formidable task because the large-scale collection and isolation of natural products from natural sources are restricted. Therefore, we have been exploring an efficient and scalable synthetic strategy for the large-scale synthesis of 1 to identify the mechanism of its hair growth effects through in vivo animal model studies, as well as its wide range of health benefits. Herein, we report the synthesis of the natural isoflavanone 7,2 -dihydroxy-4 ,5 -dimethoxyisoflavanone (1).

Results and Discussion
Our approach for the synthesis of phytoestrogen 7,2 -dihydroxy-4 ,5dimethoxyisoflavanone (1), as shown in Figure 2, includes the efficient construction of an isoflavanone framework and a gram-scale synthetic pathway. Isoflavanone 1 was obtained from 2-hydroxyketone 2 via the annulation of the deoxybenzoin skeleton and the sequential deprotection of the two masked phenols in the final stage. The deoxybenzoin unit of 2 can be constructed through a sequence of aryllithium additions to arylacetaldehyde 3 and the subsequent oxidation of the resulting alcohol. It was expected that 3 could be easily prepared from commercially available 3,4-dimethoxyphenol 4 via Claisen rearrangement and the subsequent oxidative cleavage of the terminal alkene to introduce a crucial acetaldehyde side chain.
isolated from the bark of Dalbergia oliveri Prain, a traditional Thai medicine used for the treatment of chronic ulcers in Southeast Asia (Figure 1) [14]. Isoflavanone 1 exhibits potent hair growth effects on immortalized dermal papilla cells (iDPCs). In a cell proliferation assay, 1 induced significant cell proliferative activity (54.1% at 10 μM, EC50 = 8.83 μM), which was more potent than that of Minoxidil (20.6% at 10 μM), a widely used medication for the prevention and treatment of hair loss [14]. Moreover, isoflavanone 1 induced the anagen phase of the hair cycle in a mouse model via subcutaneous (SC) injection [15]. In 2018, Kim et al. reported a sophisticated approach for the synthesis of 1. However, this synthetic route does not yield a large amount of isoflavanone 1, although the synthetic steps are relatively short [15]. Practically, the preparation of naturally occurring compounds in large quantities is a highly formidable task because the large-scale collection and isolation of natural products from natural sources are restricted. Therefore, we have been exploring an efficient and scalable synthetic strategy for the large-scale synthesis of 1 to identify the mechanism of its hair growth effects through in vivo animal model studies, as well as its wide range of health benefits. Herein, we report the synthesis of the natural isoflavanone 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1).

Results and Discussion
Our approach for the synthesis of phytoestrogen 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1), as shown in Figure 2, includes the efficient construction of an isoflavanone framework and a gram-scale synthetic pathway. Isoflavanone 1 was obtained from 2-hydroxyketone 2 via the annulation of the deoxybenzoin skeleton and the sequential deprotection of the two masked phenols in the final stage. The deoxybenzoin unit of 2 can be constructed through a sequence of aryllithium additions to arylacetaldehyde 3 and the subsequent oxidation of the resulting alcohol. It was expected that 3 could be easily prepared from commercially available 3,4-dimethoxyphenol 4 via Claisen rearrangement and the subsequent oxidative cleavage of the terminal alkene to introduce a crucial acetaldehyde side chain. The synthesis of 1 begins with the preparation of the key intermediate, deoxybenzoin 2, as shown in Scheme 1. The allylation of commercially available 3,4-dimethoxyphenol 4 and subsequent Claisen rearrangement in N,N-diethylaniline [16] solely produced o-allylsubstituted phenol 6, which was readily converted to benzyl ether 7. The dihydroxylation of the terminal alkene of 7 and subsequent oxidative cleavage by NaIO 4 readily afforded arylacetaldehyde 3. Next, the lithiation of aryl bromide 8, which was prepared from commercially available 4-bromoresorcinol, and a spontaneous nucleophilic addition to the acetaldehyde of 3 afforded benzyl alcohol 9, which was converted to methoxymethyl (MOM)-protected deoxybenzoin 10 by pyridinium dichromate (PDC) oxidation. The selective MOM deprotection of the phenol adjacent to the ketone in deoxybenzoin 10 finally afforded 2-hydroxyketone 2, which is a key intermediate for the construction of the isoflavanone framework of 1. 60 3 of 9 substituted phenol 6, which was readily converted to benzyl ether 7. The dihydroxylation of the terminal alkene of 7 and subsequent oxidative cleavage by NaIO4 readily afforded arylacetaldehyde 3. Next, the lithiation of aryl bromide 8, which was prepared from commercially available 4-bromoresorcinol, and a spontaneous nucleophilic addition to the acetaldehyde of 3 afforded benzyl alcohol 9, which was converted to methoxymethyl (MOM)-protected deoxybenzoin 10 by pyridinium dichromate (PDC) oxidation. The selective MOM deprotection of the phenol adjacent to the ketone in deoxybenzoin 10 finally afforded 2-hydroxyketone 2, which is a key intermediate for the construction of the isoflavanone framework of 1.

Scheme 1. Preparation of key intermediate deoxybenzoin 2.
Using precursor 2, the isoflavanone skeleton of 1 was constructed, as shown in Scheme 2. Previously, Gouda et al. reported an efficient and scalable approach to construct the isoflavanone framework [17]. Therefore, Gouda's protocol was employed to obtain the fully functionalized isoflavanone 11 with paraformaldehyde and Et2NH in refluxing MeOH. As expected, isoflavanone 11 was obtained with a yield of 88% on a multi-gram scale. Finally, sequential deprotection reactions were performed for the two phenols in 11 with Bn and MOM protecting groups. However, initial attempts to remove the protecting groups of phenols were unsuccessful. In the presence of Pd/C or Pd(OH)2 (Pearlman's catalyst), the hydrogenolysis of the benzyl group in 11 unexpectedly yielded phenol 12 with a very low yield (<5%). Furthermore, under acidic conditions for the MOM deprotection of 12, phenol 12 was highly unstable and degradable, despite its structural simplicity. We assumed that the intrinsic structural instability of 12 was likely due to the presence of a free phenol moiety adjacent to the ketone in the 4-chromanone skeleton, leading to unexpected and inseparable messy mixtures, especially under acidic conditions. Therefore, the deprotection sequence for two masked phenols was changed, wherein the MOM ether was deprotected first, and the benzyl ether group was then cleaved under neutral conditions in the final stage via a hydrogenolysis reaction. Significantly, the careful deprotection of the MOM ether under acidic conditions afforded the desired phenol 13 without any degradation. Finally, the subsequent hydrogenolysis of the benzyl-protecting group successfully furnished the 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1) on a gram scale. The spectral data for synthetic 1 were identical to the reported data for the natural product in all aspects [13]. Using precursor 2, the isoflavanone skeleton of 1 was constructed, as shown in Scheme 2. Previously, Gouda et al. reported an efficient and scalable approach to construct the isoflavanone framework [17]. Therefore, Gouda's protocol was employed to obtain the fully functionalized isoflavanone 11 with paraformaldehyde and Et 2 NH in refluxing MeOH. As expected, isoflavanone 11 was obtained with a yield of 88% on a multi-gram scale. Finally, sequential deprotection reactions were performed for the two phenols in 11 with Bn and MOM protecting groups. However, initial attempts to remove the protecting groups of phenols were unsuccessful. In the presence of Pd/C or Pd(OH) 2 (Pearlman's catalyst), the hydrogenolysis of the benzyl group in 11 unexpectedly yielded phenol 12 with a very low yield (<5%). Furthermore, under acidic conditions for the MOM deprotection of 12, phenol 12 was highly unstable and degradable, despite its structural simplicity. We assumed that the intrinsic structural instability of 12 was likely due to the presence of a free phenol moiety adjacent to the ketone in the 4-chromanone skeleton, leading to unexpected and inseparable messy mixtures, especially under acidic conditions. Therefore, the deprotection sequence for two masked phenols was changed, wherein the MOM ether was deprotected first, and the benzyl ether group was then cleaved under neutral conditions in the final stage via a hydrogenolysis reaction. Significantly, the careful deprotection of the MOM ether under acidic conditions afforded the desired phenol 13 without any degradation. Finally, the subsequent hydrogenolysis of the benzyl-protecting group successfully furnished the 7,2 -dihydroxy-4 ,5 -dimethoxyisoflavanone (1) on a gram scale. The spectral data for synthetic 1 were identical to the reported data for the natural product in all aspects [13].

Conclusions
A concise and scalable synthesis of phytoestrogen 7,2′-dihydroxy-4′,5′-dimethoxyisoflavanone (1) was performed successfully. The key features of the synthesis include a deoxybenzoin intermediate obtained via a sequence of Claisen rearrangement and oxidative cleavage for the installation of the acetaldehyde side chain and the subsequent nucleophilic addition of functionalized aryllithium. Moreover, the scalable synthesis of the isoflavanone framework from the functionalized 2-hydroxyketone enabled the completion of isoflavanone 1 synthesis. Further in vivo studies on the hair growth effects and the elucidation of the therapeutic potential of phytoestrogen 1 as a promising treatment for hair loss diseases, such as alopecia, are currently in progress.

General Information
Unless noted otherwise, all starting materials and reagents were obtained from commercial suppliers and were used without further purification. All solvents used for the routine isolation of products and chromatography were reagent grade and glass-distilled. Reaction flasks were dried at 100 °C. Air-and moisture-sensitive reactions were performed under an argon atmosphere. Flash column chromatography was performed using silica gel 60 (230-400 mesh, Merck, Darmstadt, Germany) with the indicated solvents. Thin-layer chromatography was performed using 0.25 mm silica gel plates (Merck). Highresolution mass data were recorded by JMS-700 (JEOL, Tokyo, Japan), and methanol solvent was used to measure the MS-ESI spectra. Infrared (IR) spectra were measured on a 1600 FTIR spectrometer (Perkin-Elmer, Waltham, MA, USA). 1 H and 13 C NMR spectra were recorded on JEOL-500 (JEOL, Tokyo, Japan) as solutions in deuteriochloroform (CDCl3) and hexadeuterodimethyl sulfoxide (DMSO-d6). The melting point (m.p.) was measured using Electrothermal IA9100. Chemical shifts are expressed in parts per million (ppm, δ) downfield from tetramethylsilane and are referenced to the deuterated solvents (CHCl3 or HCD2SOCD3 for 1 H NMR and CDCl3 or DMSO-d6 for 13 C NMR). 1 H NMR data are reported in the order of chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; dd, doublet of doublets; dt, doublet of triplet; ddd, doublet of doublet of doublets; ddt, doublet of doublet of triplets; bs, broad singlet; m, multiplet and/or multiple resonance), number of protons, and coupling constant in hertz (Hz).

Conclusions
A concise and scalable synthesis of phytoestrogen 7,2 -dihydroxy-4 ,5dimethoxyisoflavanone (1) was performed successfully. The key features of the synthesis include a deoxybenzoin intermediate obtained via a sequence of Claisen rearrangement and oxidative cleavage for the installation of the acetaldehyde side chain and the subsequent nucleophilic addition of functionalized aryllithium. Moreover, the scalable synthesis of the isoflavanone framework from the functionalized 2-hydroxyketone enabled the completion of isoflavanone 1 synthesis. Further in vivo studies on the hair growth effects and the elucidation of the therapeutic potential of phytoestrogen 1 as a promising treatment for hair loss diseases, such as alopecia, are currently in progress.

General Information
Unless noted otherwise, all starting materials and reagents were obtained from commercial suppliers and were used without further purification. All solvents used for the routine isolation of products and chromatography were reagent grade and glass-distilled. Reaction flasks were dried at 100 • C. Air-and moisture-sensitive reactions were performed under an argon atmosphere. Flash column chromatography was performed using silica gel 60 (230-400 mesh, Merck, Darmstadt, Germany) with the indicated solvents. Thin-layer chromatography was performed using 0.25 mm silica gel plates (Merck). High-resolution mass data were recorded by JMS-700 (JEOL, Tokyo, Japan), and methanol solvent was used to measure the MS-ESI spectra. Infrared (IR) spectra were measured on a 1600 FTIR spectrometer (Perkin-Elmer, Waltham, MA, USA). 1 H and 13 C NMR spectra were recorded on JEOL-500 (JEOL, Tokyo, Japan) as solutions in deuteriochloroform (CDCl 3 ) and hexadeuterodimethyl sulfoxide (DMSO-d 6 ). The melting point (m.p.) was measured using Electrothermal IA9100. Chemical shifts are expressed in parts per million (ppm, δ) downfield from tetramethylsilane and are referenced to the deuterated solvents (CHCl 3 or HCD 2 SOCD 3 for 1 H NMR and CDCl 3 or DMSO-d 6 for 13 C NMR). 1 H NMR data are reported in the order of chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; dd, doublet of doublets; dt, doublet of triplet; ddd, doublet of doublet of doublets; ddt, doublet of doublet of triplets; bs, broad singlet; m, multiplet and/or multiple resonance), number of protons, and coupling constant in hertz (Hz).