Asymmetric synthesis of flavanols via Cu-catalyzed kinetic resolution of chromenes and their anti-inflammatory activity.

Flavanols are privileged heterocyclic compounds in medicinal chemistry. It is notable to develop an efficient and straightforward protocol for accessing chiral flavanols with precise control of the stereocenters. Here, a highly efficient kinetic resolution of chromenes was reported via Cu-catalyzed asymmetric hydroboration. This previously unidentified approach features a one-step synthesis of chiral flavan-3-ols containing two vicinal stereogenic centers via a highly efficient kinetic resolution (s factor up to 1060, >99% ee for most products). In addition, the anti-inflammation effects of these diversified flavan-3-ols were studied by the in vitro experiments and RNA sequencing analysis. These flavan-3-ols showed inhibitory effects on the secretion of pro-inflammation cytokines including interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α (TNF-α), as well as inhibiting the inflammation responses through down-regulating the gene transcriptions closely related to PI3K-Akt signaling pathway and TNF signaling pathway. The results suggested that these newly synthesized flavan-3-ols have the potential to be lead compounds for anti-inflammatory drugs.

INTRODUCTION

Flavonoids are prevalent structural motifs commonly found in numerous biologically active natural compounds and pharmaceuticals (–). There are many subgroups of the general structure of a 15-carbon skeleton. Among them, flavan-3-ols (2-phenylchroman-3-ols) and their derivatives, which feature two vicinal chiral centers at the C2 and C3, represent an important and special subclass of flavonoids. A number of natural flavan-3-ols including tupichinol A, (+)-catechin, (+)-gallocatechin, (−)-epicatechin-3-gallate, and theaflavin have been isolated from natural plants (Fig. 1A) (–). These natural products carry various interesting biological activities such as antioxidant, anticancer, antimicrobial, and antiviral (–). Given the interesting and extensive bioactivity of flavan-3-ols, as well as the difficulties and tediousness in isolating them in pure form from natural plants, having an effective asymmetric synthesis strategy to obtain flavan-3-ols is highly desirable. In particular, asymmetric catalysis allows easy manipulation of the stereo-configuration of a library of flavan-3-ols for biological investigations. However, only a very limited number of efficient synthetic methods for constructing chiral flavan-3-ols have been reported (–). The common approaches for the preparation of this important building block always involve multistep synthesis using Sharpless asymmetric dihydroxylation or asymmetric epoxidation (Sharpless epoxidation and Shi epoxidation) to construct the chiral center and a subsequent cyclization to generate the C ring of the flavan-3-ol framework (Fig. 1B) (). Thus, it is highly demanded to develop a new strategy for the construction of flavan-3-ol skeletons with excellent enantioselectivity.

Fig. 1.

Structures of chiral flavonoid, selected examples of flavan-3-ols, and asymmetric synthesis of flavan-3-ols.

(A) Structures of chiral flavonoid and selected examples of flavan-3-ols. (B) Multistep construction of chiral flavan-3-ols. (C) Kinetic resolution of 2H-chromenes by Cu-catalyzed asymmetric hydroboration.

Chiral organoboron compounds are versatile intermediates in organic synthesis, chemical biology, and material sciences because the C─B bond can be easily transformed into C─O, C─C, C─N, and C─halogen bonds in a stereospecific fashion (–). Transition metal–catalyzed asymmetric hydroboration of alkenes has been proven as one of the most direct and powerful methods for the preparation of chiral alkylboronic acid derivatives (, ). Rh (, ), Ir (), Co (–), and Cu (–) complexes have been successfully applied in asymmetric hydroboration of alkenes. Among the reported methods, Cu-catalyzed asymmetric hydroboration has gained increasing attention because of catalyst inexpensiveness, mild reaction conditions, and high levels of selectivity. To date, a series of readily available alkenes have been studied in Cu-catalyzed asymmetric hydroboration to give chiral organoboron compounds in high yields and enantioselectivities.

Kinetic resolution represents a simple and efficient way to afford both the chiral products and the enantio-enriched starting materials (–). So far, the scopes of reaction, catalysts, and substrates suitable for kinetic resolution are still limited, and achieving a high-resolution efficiency is a long-standing challenge. In 2013, Metz et al. () reported a kinetic resolution of racemic flavanones via Rh-catalyzed asymmetric transfer hydrogenation. Later, Cu-catalyzed asymmetric hydroboration has been applied to racemic 2-substituted 1,2-dihydroquinolines (). Recently, our group developed a novel kinetic resolution and dynamic kinetic resolution of 2H-chromenes and dihydroquinolines by rhodium-catalyzed asymmetric hydroarylation (, ). On the basis of our continuous efforts in constructing chiral bioactive flavonoids (–), here, we developed a kinetic resolution of racemic 2-substitued 2H-chromenes by Cu-catalyzed asymmetric hydroboration for one-pot synthesis of chiral flavan-3-ols with high kinetic resolution factors (Fig. 1C). In addition, more biological investigations of these diversified flavan-3-ols were further studied by the in vitro experiments and RNA sequencing (RNA-seq) analysis. In particular, these newly synthesized heterocycles suppressed the expression and secretion of pro-inflammation cytokines including interleukin-1β (IL-1β), IL-6, and tumor necrosis factor–α (TNF-α). Further data revealed that these compounds inhibit inflammation responses through down-regulating the gene transcriptions closely related to IL-17 signaling pathway, PI3K-Akt signaling pathway, and TNF signaling pathway, which suggested that these newly synthesized flavan-3-ols are potent lead compounds for treating inflammation diseases.

RESULTS

Reaction condition optimization

The asymmetric hydroboration/kinetic resolution of rac-flavene (2-aryl-chromene) 1a with B2pin2 (2) in MeOH was evaluated in the presence of Cu complex, followed by oxidative workup with NaBO3·H2O. Initially, various chiral bisphosphine ligands with different backbones using CuCl as the catalyst precursor were investigated. To our delight, the flavan-3-ol 3a with vicinal chiral centers was obtained essentially as a single diastereomer (dr > 19:1) in 50% isolated yield with 73% ee, and the recovered 1a was also obtained in 32% yield with 99% ee (s = 32), catalyzed by the in situ complex formed from 5.0 mole percent (mol %) CuCl and 5.5 mol % (R, R)-Ph-BPE L1 in tetrahydrofuran (THF) at room temperature (Table 1, entry 1). (R, R)-Me-Duphos L3 and (R, Sp)-Josiphos L4 could promote the full conversion of rac-flavene 1a, but the ee values of product 3a were inferior (Table 1, entries 3 and 4). When (R)-Binap L5 was used, the trace amount of the product was detected (Table 1, entry 5). Other diphosphine ligands, such as (R)-DM-Segphos L6, (R)-Difluorphos L7, and (R)-Xylyl-P-Phos L8, afforded product 3a with high enantioselectivities (74 to 86%), albeit with low enantioselectivities for the recovered substrate 1a (20 to 68%). The temperature factor was also investigated by using (R, R)-Ph-BPE L1 as the ligand. The kinetic resolution was highly selective when the reaction temperature was lowered to −35°C. Both the flavanol product 3a and the substrate flavene 1a were obtained in excellent yields (45 and 46%) and extremely high enantioselectivities (99%), with a selective factor of 1060 (Table 1, entry 11).

Table 1.

Optimization of reaction conditions.

Reaction conditions: CuCl (5.0 mol %), ligand (5.5 mol %), 1a (0.2 mmol), 2 (0.24 mmol), NaOtBu (10.0 mol %), MeOH (0.2 mmol), THF (0.6 ml), NaBO3·H2O (0.5 mmol), and H2O (0.6 ml). rt, room temperature.

Entry	Ligand	Temperature (°C)	(R)-1a* yield (%)	(R)-1a† ee (%)	3a*,‡ yield (%)	3a† ee (%)	Conv.§ (%)	s‖
1	L1	rt	32	99	50	73	58	32
2	L2	rt	22	88	37	46	66	7
3	L3	rt	trace	–	88	11	–	–
4	L4	rt	trace	–	58	7	–	–
5	L5	rt	–	–	Trace	–	–	–
6	L6	rt	58	38	31	79	33	12
7	L7	rt	68	20	16	74	21	8
8	L8	rt	45	62	36	−86	42	25
9	L1	0	37	>99	62	76	57	37
10	L1	−30	43	>99	48	93	52	145
11	L1	−35	46	99	45	>99	50	1060
12	L1	−40	63	50	28	99	37	328

*Isolated yield.

†Determined by chiral HPLC.

‡Diastereomeric ratio (dr) > 19:1 (determined by 1H NMR).

§Calculated conversion, C = ee/(ee + ee).

‖Selectivity factor (s) = ln[(1 –C) (1 − ee)]/ln[(1-C)(1 + ee)].

Substrate scope

With the optimized reaction conditions in hand, we next examined the scope of flavenes with various substituents, and the results were illustrated in Table 2. A very broad scope of substrates was studied, with substituents on both the C2-phenyl ring and the benzopyran moieties spanning a diverse range of sterically and electronically different groups. In general, the reactions afforded the flavan-3-ols 3a-3ai as single diastereomers in high yields (34 to 48%) with excellent enantioselectivities (95 to >99%, >99% for most substrates) and the recovered (R)-flavenes in 35 to 48% yields with 83 to >99% enantioselectivities (s factor, 170 to > 060). Substrates 1b-i bearing a wide range of electronically varied phenyls at the C2 position with different substituents proceeded with excellent chiral recognition. No significant steric effect was observed. The o-tolyl–substituted substrate rac-1j afforded the product 3j with the highest enantioselectivity (>99%) and a selective factor of 1060. Replacing the phenyl ring with 2-naphthyl moiety led to a similar result, producing the pi-extended 2-naphthyl group product 3m with >99% ee and recovered flavene 1m with >99% ee (s factor, >1060). Notably, the substrates containing heterocycles, such as 2-furyl or 2-thienyl, also worked well under the catalytic conditions and afforded the products 3n-o with 98 to 99% ee and the recovered substrates 1n-o with 92 to 96% ee. To our delight, the 2-alkyl-chromene 1p was successfully resolved under these reaction conditions, giving the product 3p in 34% yield and >99% ee and recovered starting material in 35% yield with >99% ee (s factor, >1060). The hydroboration of flavenes with an electron-donating group either at the C5, C6, C7, or C8 position (1q-1aa) was found successful and furnished the desired products in high yields (39 to 46%) with excellent enantioselectivities (83 to >99%). Besides, substrate 1ac with a hydroxyl substituent at C6 position could also be successfully converted into the corresponding product in excellent enantioselectivity (>99% ee). The racemic flavenes bearing an electron-withdrawing group at C6 or C8 position (1ad-ag) were well tolerated. High yields and excellent enantioselectivities were obtained for both products and recovered flavenes. The flavenes with ─F or ─Cl groups on both C2-phenyl ring and benzopyran also reacted smoothly, giving the products in high yields with excellent enantioselectivities (3ah and 3ai). The configuration of the recovered 1a was assigned as R configuration by comparison with the literature data of the known compound (), and the absolute configuration of product 3m was confirmed as (2R, 3S) by x-ray crystal structure analysis (). Notably, the opposite configuration of flavanols (2S, 3R)-3′ can also be accessed by using the ligand (S, S)-Ph-BPE (L1) under similar conditions (Table 2B, 3m′, 3o′, 3p′, 3ae′, and 3ag′).

Table 2.

Substrate scope.

Reaction conditions: CuCl (5.0 mol %), Ph-BPE (L1) (5.5 mol %), 1 (0.2 mmol), 2 (0.24 mmol), NaOtBu (10 mol%), MeOH (0.2 mmol), and THF (0.6 ml). Yields were isolated yields. Ees were determined by chiral HPLC. Diastereomeric ratio (dr) > 19:1 (determined by 1H NMR). Calculated conversion (conv), C = ee/(ee + ee). Selectivity factor (s) = ln[(1 –C)(1 − ee)]/ln[(1 –C)(1 + ee)].

Synthetic application

To demonstrate the potential applicability of this method, the resulting flavan-3-ol 3a was transformed into (2R, 3S)-2-phenylchroman-3-yl 3,4,5-trihydroxybenzoate 4 by esterification with tri-OBn gallic acid chloride, followed by hydrogenolysis, resulting in 63% yield over the two steps. Note that the product 4 was reported to show excellent anti-staphylococcal activity due to its ability to reverse methicillin resistance of the drug-resistant strains of Staphylococcus aureus (Fig. 2A) (). Furthermore, product 5 was obtained in 95% yield by the removal of methoxylmethyl (MOM) ester on the C2-phenyl of 3e in the presence of HCl and iPrOH without losing any enantioselectivities (Fig. 2B).

Fig. 2.

Synthetic application.

(A) Transformation of 3a. (B) Deprotection of MOM of 3e. rt, room temperature.

Plausible mechanism

A deuterium-labeling experiment was conducted to probe the reaction mechanism (Fig. 3). The borylation of rac-1a using CD3OD instead of CH3OH furnished the deuterium-labeled product 3a′ (44% yield, >99% ee, 90% D incorporation). The syn configuration between the hydoxyl group and the deuterium atom at the 4-position indicated that the Cu-(Bpin) complex underwent syn-addition to the C═C double bond of 2H-chromene rac-1a. On the basis of the aforementioned results and previous reports (, , , ), we propose a plausible reaction mechanism for the enantioselective borylation of 2-substituted 2H-chromenes by kinetic resolution (Fig. 3). The chiral Cu-alkoxide A generated in situ from (R, R)-Ph-BPE ligand, CuCl, and NaOtBu undergoes transmetalation with B2(pin)2 to afford nucleophilic Cu-B(pin) B species. The coordination of racemic flavene 1 to the chiral copper center B followed by the addition through a four-membered transition state C delivers the alkyl-Cu complex D. The benzylic carbon is more electronegative since it could be stabilized by the π* orbital of the aryl group. Last, the protonation of alkyl-Cu intermediate D by MeOH provides the product E and regenerates Cu-alkoxide.

Fig. 3.

Deuterium-labeling experiment and proposed catalytic cycle.

Anti-inflammatory efficiency of flavan-3-ols

Natural products or small molecules containing flavan skeleton have been reported to exhibit broad anti-inflammatory activity. To be noticed, most of the synthesized derivatives in Table 2 are new compounds through our data retrieval. Only four derivatives (3a, 3c, 3d, and 3aa) have been already described in (–). It is noteworthy that the anti-inflammatory capacity of all these compounds has not been investigated. To evaluate the anti-inflammatory effects of these newly synthesized flavan-3-ols, RAW 264.7 cells (mouse monocyte/macrophage cell) were used as the in vitro model. First, CCK-8 (Cell Counting Kit-8) assay was performed to test the effects of these newly synthesized flavan-3-ols on cell viability. RAW 264.7 cells were treated with the newly synthesized flavan-3-ols ranging from 25 to 100 μM for 24 hours. As shown in fig. S1, treatment with each compound at 25 and 50 μM for 24 hours had no effect on the cell viability of RAW 264.7 cell. Nevertheless, reduced viability was observed with some of the flavan-3-ols at 100 μM, such as compounds 3i, 3n, 3ad, and 3ai. Therefore, the final concentration of each compound used in the following experiments was no more than 100 μM, and compound concentrations at 25 and 50 μM were used in the further experiments.

The increased pro-inflammatory cytokine production in RAW 264.7 cells has been reported to play an influential role in the inflammation response (). To evaluate the anti-inflammatory effects of these flavan-3-ols in vitro, enzyme-linked immunosorbent assay (ELISA) kits were used to investigate their effects on the lipopolysaccharide (LPS)–induced pro-inflammatory cytokine production of RAW 264.7 cells. After treating with the flavan-3-ols (25 and 50 μM) and stimulated by LPS, the levels of three main pro-inflammation cytokines including IL-1β, IL-6, and TNF-α in the culture supernatant from RAW 264.7 cells were measured by ELISA kits.

As shown in Fig. 4, compared with LPS-stimulated group, most of the flavan-3-ols have inhibition effects on the secretion of IL-1β, IL-6, and TNF-α. We selected five final flavan-3-ols (3m, 3o, 3p, 3ae, and 3ag) that consistently showed higher inhibition rates (at least 20 to 60%) across all three pro-inflammatory cytokines tested and omitted those that showed lower inhibition rates (10 to 20%). All these results suggested that these five entities of flavan-3-ols would serve as the potential inflammation inhibitors.

Fig. 4.

The effect of flavan-3-ols on pro-inflammatory cytokine secretion in vitro.

RAW 264.7 cells were pretreated with each compound (25 and 50 μM) for 24 hours followed by LPS (1 μg/ml) stimulation for another 2 hours. The concentrations of IL-1β, IL-6, and TNF-α of compounds on the secretion of each pro-inflammatory cytokine in RAW 264.7 cell culture supernatants were measured by ELISA kits. (A) The amount of secreted IL-1β from LPS-stimulated RAW 264.7 cells treated with compounds 3a to 3ai. (B) The amount of secreted IL-6 from LPS-stimulated RAW 264.7 cells treated with compounds 3a to 3ai. (C) The amount of secreted TNF-α from LPS-stimulated RAW 264.7 cells treated with compounds 3a to 3ai. Data from three independent experiments were expressed as means ± SD. *P < 0.05 and **P < 0.01 compared with the LPS-stimulated group.

RNA-seq analysis of downstream genes regulated by flavan-3-ols

Given that compounds 3m, 3o, 3p, 3ae, and 3ag exhibited potent anti-inflammatory effects, we next examined flavan-3-ols–mediated transcriptional targets and pathways that could potentially be accounted for the inflammation response. RNA-seq analysis was performed to further study the effects of compounds 3m, 3o, 3p, 3ae, and 3ag on gene transcriptional pathways related to inflammation. RAW 264.7 cells were used in the RNA-seq analysis, in which cells were pretreated with each compound (50 μM) for 24 hours followed by LPS (1 μg/ml) stimulation for another 2 hours. After analyzing the whole genome sequencing results, we found that there were 1098 genes differentially regulated by compound 3ag as compared with the untreated LPS-stimulated group (Fig. 5A), of which 954 genes (49.8%) were down-regulated and 144 genes (7.5%) were up-regulated (Fig. 5, A and B, and table S1). Next, we focused on genes related to inflammatory reactions and inflammation-related diseases (inflammation-related genes) based on the gene and pathway classification of Kyoto Encyclopedia of Genes and Genomes (KEGG) database. As shown in Fig. 5C, among all these inflammation-related genes, 75 genes (listed on the right of Fig. 5C) were significantly changed by LPS stimulation as compared with the control group. In addition, as compared with the LPS-stimulated group, 40 genes were down-regulated by 50 μM compound 3ag, and the other four compounds (3m, 3o, 3p, and 3ae) also down-regulated parts of inflammation-related genes at 50 μM. However, the other four compounds down-regulated much fewer inflammation-related genes as compared to that of compound 3ag (Fig. 5C). Together, these results suggested that compound 3ag potentially exerts relatively stronger anti-inflammation effects by regulating a broader range of inflammatory genes and pathways.

Fig. 5.

RNA-seq analysis of flavan-3-ol 3ag–modulated genes.

(A) A volcano plot illustrating differentially regulated gene (DEG) expression from RNA-seq analysis between the LPS-stimulated group and the compound 3ag-treated group. The up-regulated and down-regulated genes are shown in orange and blue, respectively. Values are presented as the log2 of tag counts. (B) RNA-seq comparison revealed a total of 1915 genes expressed, of which 954 genes (49.8%) were down-regulated and 144 genes (7.5%) were up-regulated. (C) The hierarchical clustering of the whole genome analysis results showing the representative top 75 genes that were significantly altered in the expression level.

To further analyze the functional distribution of differentially expressed genes in relation to different biological processes, cellular components, and molecular functions,

Gene Ontology (GO) enrichment analysis was performed. The histogram of GO enrichment of differentially expressed genes was presented. As shown in Fig. 6, GO enrichment analysis demonstrated different categories of genes that were significantly down-regulated or up-regulated by the treatment of compound 3ag as compared with the LPS-induced group. In line with our ELISA results that showed reduction in pro-inflammatory cytokines (Fig. 4), the down-regulated genes were associated with inflammatory response (GO: 0006954) and cellular response to LPS (GO: 0071222) (Fig. 6 and table S2). These results suggested that compound 3ag shows strong inhibition effects on inflammation process and is closely involved in inflammation response.

Fig. 6.

GO functional clustering of genes that were down-regulated or up-regulated in different biological processes.

GO analysis result illustrates the genes that were significantly affected by compound 3ag treatment clustered in accordance with specific molecular functions (red), cellular components (green), and biological processes (blue). ERK1, extracellular signal–regulated kinase1.

Furthermore, to determine the biochemical metabolic pathways and signal transduction pathways involved in differentially expressed genes regulated by compound 3ag, KEGG pathway analysis was then performed. As shown in Fig. 7A, compared with the LPS-induced group, a number of genes that were down-regulated in compound 3ag-treated group were distributed in several main pathways closely related to inflammation process including IL-17 signaling pathway, PI3K-Akt signaling pathway, and TNF signaling pathway. These genes down-regulated by compound 3ag were closely related to several inflammatory diseases, such as rheumatoid arthritis and inflammatory bowel disease (table S3). Together, these results suggested that flavan-3-ol 3ag inhibited the transcription of genes closely involved in the process of inflammation and inflammation-related diseases, which is in line with the previous findings on the anti-inflammatory function of flavan skeleton compounds.

Fig. 7.

The KEGG pathway analysis of the differentially expressed genes.

(A) KEGG pathway analysis of the distribution of differentially expressed genes. Compared with the LPS-induced group, the distribution of differentially expressed genes regulated by compound 3ag distributed in different categories was shown. (B) The scatterplot showing KEGG rich distribution of differential genes. PPAR, peroxisome proliferator–activated receptor; MAPK, mitogen-activated protein kinase; STAT, signal transducers and activators of transcription; AGE-RAGE, advanced glycation end products, receptor for advanced glycation end products; FoxO, forkhead box O; AMPK, AMP-activated protein kinase.

To evaluate the influence of different configurations of the flavan-3-ols on the anti-inflammation activity, compounds (2, 3)-3m, (2, 3)-3o, (2, 3)-3p, (2, 3)-3ae, and (2, 3)-3ag and their enantiomers (2, 3)-3m′, (2, 3)-3o′, (2, 3)-3p′, (2, 3)-3ae′, and (2, 3)-3ag′ were selected. Then, these 10 compounds were evaluated for their anti-inflammation activity together with their racemic forms (-3m, -3o, -3p, -3ae, and -3ag). The inhibition rate of IL-1β secretion was applied to evaluate the anti-inflammation effect of each compound. Our results revealed that compounds 3m, 3o, 3ag, and 3ae showed higher inhibition rates than their enantiomers 3m′, 3o′, 3ag′, 3ae′, and racemic mixtures (Fig. 8). Meanwhile, compound (2, 3)-3p′ exhibited higher inhibition rate of IL-1β secretion than its racemic mixture (Fig. 8). These results strongly suggested that the chirality plays an important role in the recognition between the biologically active molecules and their targets, and enantio-enriched flavanols always show higher inhibition rates than their racemic mixtures.

Fig. 8.

The inhibition rates of compounds on IL-1β secretion.

RAW 264.7 cells were pretreated with each compound (50 μM) for 24 hours followed by LPS (1 μg/ml) stimulation for another 2 hours. The concentrations of IL-1β in RAW 264.7 cell culture supernatants were measured by ELISA kits, and the inhibition rate of compound on the secretion of each pro-inflammatory cytokine was calculated. Data from three independent experiments were expressed as means ± SD. *P < 0.05 and **P < 0.01 compared with the LPS-stimulated group.

DISCUSSION

A highly efficient kinetic resolution of chromenes was realized by Cu-catalyzed asymmetric hydroboration (s factor up to 1060, >99% ee for most substrates and products, exclusively trans products). This novel approach afforded a series of chiral flavan-3-ols containing two vicinal stereogenic centers. Then, the anti-inflammation effects of these diversified flavan-3-ols were further studied by the in vitro experiments. As we know, IL-1β, IL-6, and TNF-α are three key pro-inflammatory cytokines, and the overexpression of them is related to inflammation responses closely. The inhibitory effect of the newly synthesized compounds 3a and 3ai on the secretion of pro-inflammatory cytokines was investigated as a model system for the structure-activity relationship study. It was observed that the substituent at C2 position affected the activity of the compounds as 2-naphthyl–substituted derivative 3m and 2-furyl–substituted derivative 3o exhibited higher inhibition rates across all the pro-inflammatory cytokines tested. Among the compounds with ortho-, meta-, and para-methyl at 2-phenyl, compounds 3b, 3j, 3k, and 3l showed no ability to inhibit the secretion of pro-inflammatory cytokines. For compound 3c, when the methoxy group at the 2-phenyl has been substituted with a more bulky benzyloxy group, the corresponding compound 3d revealed a higher potency. The introduction of the halogen atoms (F, Cl, and Br) on the 2-phenyl, compounds 3f, 3g, and 3h, led to enhanced inhibitory potency. It is noteworthy that compound 3i, where an electronegative ─CF3 group was integrated, also maintained the similar inhibitory potency as 3f. Compound 3n, replacing the 2-phenyl with 2-thiophene, suppressed inhibitory activity.

On the other hand, substituent groups on C5, C6, C7, and C8 appear to be more critical for the activity, with different kinds of residues, such as electron-withdrawing (Cl and F) or electron-donating (OMe) groups. Among the series, the best results were obtained from compounds 3ae and 3ag, which showed higher inhibition rates than others across all three pro-inflammatory cytokines tested. It was observed that the introduction of chlorine atom significantly increased the inhibition efficiency of compounds 3ae, 3af, and 3ag. Yet, replacing chlorine with fluorine (3ai) had a detrimental effect on the activity. These studies revealed that the potent inhibition activity against secretion of pro-inflammatory cytokines of these flavonoids is close related to the presence of the flavan-3-ols pharmacophore.

In summary, we have developed a kinetic resolution of racemic 2-substitued 2H-chromenes by Cu-catalyzed asymmetric hydroboration for one-pot synthesis of chiral flavan-3-ols with high kinetic resolution factors (s up to 1060). A wide range of chiral flavan-3-ols were straightforwardly afforded with good yields and extremely high enantioselectivities and diastereoselectivities. These enantio-enriched flavanols showed remarkable inhibition on the expression of multiple inflammation genes and signaling pathways. Our in vitro experiments demonstrated that inflammation response was effectively suppressed by these novel enantio-enriched flavanols. RNA-seq experiments further revealed that the inflammatory inhibition potentially mediated through transcription regulation of IL-17 signaling pathway, PI3K-Akt signaling pathway, and TNF signaling pathway. These new entities of flavanols have excellent potential in serving as lead compounds for the treatment of inflammation diseases.

MATERIALS AND METHODS

All the air- or moisture-sensitive reactions and manipulations were performed under an atmosphere of argon by using standard Schlenk techniques and Drybox (Mikrouna, Supper 1220/750). 1H nuclear magnetic resonance (NMR) and 13C NMR spectra were recorded on a 400- or 500-MHz Bruker Avance spectrometer. CDCl3 was used as solvent. Chemical shifts (δ) were reported in parts per million with tetramethylsilane as the internal standard, and J values were given in hertz. The following abbreviations were used to explain the multiplicities: s, singlet; d, doublet; dd, double of doublets; t, triplet; q, quartet; m, multiplet. Flash column chromatograph was carried out using 200- to 300-mesh silica gel at medium pressure. High-resolution mass spectra (HRMS) were recorded on a liquid chromatography (LC)–time-of-flight spectrometer. Electrospray ionization (ESI)–HRMS data were acquired using a Thermo LTQ Orbitrap XL Instrument equipped with an ESI source. Optical rotation was obtained on a Rudolph Research Analytical (Atopol I). High-performance LC (HPLC) analysis was performed on Agilent 1260 series, and ultraviolet detection was monitored at 230 or 220 nm. Tetrahydrofuran was distilled over sodium. Melting points were measured on a MP-450 (Hanon) melting point apparatus and uncorrected.

Cu-catalyzed kinetic resolution of chromenes via asymmetric hydroboration

In a nitrogen-filled glovebox, a flame-dried screw-cap reaction tube equipped with a magnetic stir bar was charged with CuCl (0.01 mmol, 1.0 mg), (R, R)-Ph-BPE (0.012 mmol, 6.1 mg), and NaOtBu (0.02 mmol, 1.9 mg), and anhydrous THF (0.6 ml) was added. Then, the reaction mixture was stirred for 15 min. 2-Phenyl-2H-chromene 1a (0.20 mmol, 41.6 mg) and B2(pin)2 2 (0.24 mmol, 60.9 mg) were added. The Schleck reaction vial was sealed with a rubber plug and taken out of the glovebox. The tube was allowed to stir at −35°C for 5 min. Then, MeOH (0.2 mmol, 8 μl) was added. The resulting solution was allowed to stir at −35°C for 24 hours. NaBO3·H2O (0.5 mmol, 50.0 mg) and H2O (0.6 ml) were added. The resulting mixture was allowed to stir at room temperature for 5 hours. The reaction mixture was diluted with EtOAc (10 ml) and H2O (3 ml). The aqueous layer was extracted with EtOAc (2×, 10 ml). The organic layer was dried over anhydrous Na2SO4, filtered, and concentrated under vacuum. The residue was purified by column chromatography on silica gel to get the corresponding hydroboration product 3a and the recovered starting material (R)-1a. The ee values of 3a and 1a were determined by HPLC. Diastereomeric ratio was determined by 1H NMR.

Cell culture

The RAW 264.7 cell line was purchased from the Cell Bank of Shanghai Institute of Biochemistry and Cell Biology at the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% (v/v) fetal bovine serum, penicillin (100 U/ml), and streptomycin (100 U/ml) in a stable environment with 5% CO2 at 37°C. Before used in the following in vitro experiments, RAW 264.7 cells were pretreated with synthesized flavan-3-ol (25 and 50 μM) for 24 hours followed by LPS (1 g/ml) stimulation for another 2 hours.

Cell viability assay

The CCK-8 assay was used to evaluate the effect of each flavan-3-ol on the viability of RAW 264.7 cells. Cells were plated into 96-well plates at a density of 2 × 105 cells per well in medium and cultured overnight. For formal experiments, cells were treated with different concentrations of flavan-3-ols (25, 50, and 100 μM). After 24 hours, 20 μl of CCK-8 solution (catalog no. A311-01/02, Vazyme Biotech Co. Ltd., Nanjing, China) was added into the medium and incubated for 45 min. The absorbance was measured at 450 nm.

Enzyme-linked immunosorbent assay

RAW 264.7 cells were treated with these synthesized flavan-3-ols (25 and 50 μM) for 24 hours followed by LPS (1 μg/ml) stimulation for another 2 hours, and the culture supernatant was collected. The concentrations of IL-1β, IL-6, and TNF-α in the culture supernatant of RAW 264.7 cells were determined according to the manufacturer’s instructions of the Duo-set ELISA kits, purchased from R&D Systems Co. Ltd. (Minneapolis, MN, USA).

RNA-seq analysis

RAW 264.7 cells were treated with each flavan-3-ol (25 and 50 μM) for 24 hours followed by LPS (1 μg/ml) stimulation for another 2 hours. Cells were collected and then directly prepared for complementary DNA (cDNA) amplification and RNA-seq library construction. Total RNA was isolated with TRIzol reagent from each group. A cDNA library was prepared, and sequencing was performed according to the Illumina standard protocol developed by Suzhou GENEWIZ Biological Technology Co. Ltd. (https://genewiz.com). Raw reads from RNA-seq libraries were trimmed to remove the adaptor sequence and the reads with adaptor contaminants, low-quality reads (the mass value Q score < 5 of the base number accounts for more than 50%), and reads from N (N indicates the base information that cannot be determined), which is >10%. After filtering, reference genome and gene model annotation files were downloaded from a genome website browser (NCBI/UCSC/Ensembl). Indexes of the reference genome were built using Bowtie v2.0.6, and paired-end clean reads were aligned to the reference genome using TopHat v2.0.9. Bowtie was used for a BWT (Burrows-Wheeler transformer) algorithm for mapping reads to the genome, and TopHat can generate a database of splice junctions based on the gene model annotation file and thus achieve a better mapping result than other nonsplice mapping tools. For the quantification of the gene expression level, HTSeq v0.6.1 was used to count the read numbers mapped for each gene. The reads per kilobase per million mapped reads of each gene was calculated on the basis of the gene read counts mapped to this gene. A differential expression analysis was performed using the DESeq R package (version 1.10.1). For clustering, we clustered different samples to see the correlation using hierarchical clustering distance method with the function of heatmap, SOM (self-organization mapping), and k-means using silhouette coefficient to adapt the optimal classification with default parameter in R. The complete RNA-seq datasets in this manuscript captioned has been deposited at Gene Expression Omnibus database (http://ncbi.nlm.nih.gov/geo/) under accession ID GSE181052.

GO and KEGG enrichment analysis

GOSeq (v1.34.1) was used in identifying GO terms that annotate a list of enriched genes with a significant P adjusted value of less than 0.05. In addition, topGO was used to plot DAG.

KEGG is a collection of databases dealing with genomes, biological pathways, diseases, drugs, and chemical substances (http://en.wikipedia.org/wiki/KEGG). We used in-house scripts to enrich significant differential expression gene in KEGG pathways.

Statistical analysis

The data shown in the study were obtained from at least three independent experiments, and all data in different experimental groups were expressed as the mean SD. Statistical analyses were performed using a one-way analysis of variance, with post hoc analysis. Details of each statistical analysis are provided in the figure legends. Differences with P < 0.05 were considered statistically significant.