Introduction

Sustained hepatitis B virus (HBV) infections can cause chronic liver diseases, such as liver cirrhosis and hepatocellular carcinoma. According to the World Health Organization (WHO), 296 million people worldwide were chronically infected, and 820,000 deaths were caused by cirrhosis or hepatocellular carcinoma in 2019 [1]. Small molecule drugs clinically used to treat chronic HBV infection include nucleos(t)ide analogs, such as entecavir and tenofovir, which inhibit reverse transcription [2, 3]. Although nucleos(t)ide analogs reduce the risk of developing hepatocellular carcinoma, long-term treatment with nucleos(t)ide analogs often results in the emergence of drug-resistant viruses and lower treatment efficacies [4,5,6]. To overcome this problem, novel anti-HBV agents aimed at different molecular targets need to be developed.

We previously isolated several antiviral natural products from the culture broths of fungi isolated from seaweed, mosses, and plants [7,8,9,10,11,12]. Neoechinulin B [7], sulochrin [8] (known compounds) and peniciherquamide C (a novel compound) [9] were found to inhibit hepatitis C virus (HCV) infection. In addition, we discovered that vanitaracin A, a novel tricyclic polyketide, is an anti-HBV agent (Fig. 1) [10]. Vanitaracin A inhibits the viral entry process with an IC50 value of 0.61 µM without evident cytotoxicity (CC50 > 256 µM) in primary human hepatocytes. Our mechanistic study revealed that vanitaracin A directly binds to sodium taurocholate co-transporting polypeptide (NTCP), an HBV entry receptor. Importantly, vanitaracin A was also effective against infection by a clinically relevant entecavir-resistant HBV strain [13].

Fig. 1
figure 1

Structure of vanitaracin A (1)

The development of a synthetic approach to vanitaracin A and its derivatives is required to support a structure–activity relationship study; however, vanitaracin A possesses six stereogenic centers and only the syn OH-2/H3-15 relationship was determined from NOESY correlations between H-2 and H-4 in our previous study [10]. In the present study, we determined the absolute configuration of the secondary alcohol in vanitaracin A using a modified Mosher ester method. Since vanitaracin A is difficult to crystallize, we used the crystalline sponge method to determine the relative configurations of the stereogenic centers in the compound. The crystalline sponge method is an X-ray technique that analyzes the structure of a compound by absorbing the guest compound into the ordered cavity of a crystalline porous metal–organic coordination complex [14]. In addition to these analyses, NOESY experiments on vanitaracin A and NMR analyses of the crude side-chain carboxylic acid obtained by the alkaline hydrolysis of vanitaracin A revealed the absolute configurations of all six of its stereogenic centers.

Results and discussion

We previously determined the syn OH-2/H3-15 relationship in vanitaracin A from the NOESY correlation between H-2 and H-4 [10]. To determine the absolute configuration at C-2, we synthesized the α-methoxy-α-trifluoromethylphenylacetic acid (MTPA, Mosher) ester of 1 (Scheme 1). To that end, 1 was reacted with methyl iodide in the presence of K2CO3 to yield methyl ether 2, which was esterified with (R)- or (S)-MTPA chloride to afford (S)- or (R)-MTPA ester 3a or 3b. Differences in the 1H NMR chemical shifts of the two esters (Δδ; δ(S)δ(R)) revealed the absolute configuration at C-2 to be S (Fig. 2).

Scheme 1
scheme 1

Preparing (S)- and (R)-MTPA esters 3a and 3b via methyl ether 2

Fig. 2
figure 2

Δδ values for MTPA esters 3a and 3b

We used the crystalline sponge method to determine the configurations of the remaining stereogenic centers in vanitaracin A [14]. A single crystal of the porous [(ZnI2)3(tpt)2(solvent)x]n complex (crystalline sponge, tpt, tris(4-pyridyl)-1,3,5-triazine) was treated with vanitaracin A (1), which was absorbed into the crystal by the slow evaporation method [14, 15] and subjected to X-ray diffractometry. The structure of an asymmetric unit containing 1 and the crystalline sponge is shown in Fig. 3a. The relative configurations of C-2, C-4, C-11, and C-2′ were determined as shown in Fig. 3b, which suggests that the absolute configuration of 1 is 2S, 4R, 11R, 2′S. Unfortunately, we were unable to determine the configuration of C-9 using this method because C-16 and 9-OH are not distinguishable in the crystal structure. In addition, C-5′ and C-8′ were not distinguishable because the position of C-6′ was unclear owing to the various conformations of the side chain moiety; consequently, we were also unable to determine the configuration of C-4′. To determine the configurations of C-9 and C-4′, we subjected 1 to NOESY spectroscopy and analyzed the crude carboxylic acid obtained by the hydrolysis of 1 by NMR spectroscopy.

Fig. 3
figure 3

Crystal structure of the complex of 1 and the crystalline sponge. (a) Asymmetric unit containing 1 and the crystalline sponge (ORTEP drawing with 50% probability). (b) Crystal structure of 1 in ball-and-stick representation. (c) Absolute structure of 1 determined by crystalline sponge analysis

Recently, Chen et al. reported the structures of penijanthinones A and B that possess 2,4-dimethyl-4-hydroxycyclohexane-1,3-dione rings (Fig. 4a) [16]. They concluded that the two methyl groups in each structure are positioned on opposite faces of the ring because the relevant protons (H3-16 and H3-17) are not NOE correlated. The NOESY spectrum of vanitaracin A was acquired to reveal the configuration of C-9. No cross peaks were observed between H3-16 (δH 1.71) and H3-17 (δH 1.50) (Fig. 4a and S3), suggesting that the relative configurations of the two methyl groups are similar to those of penijanthinones A and B; hence, the absolute configuration of C-9 was assigned to be R. In addition, we acquired the electronic circular dichroism (ECD) spectrum of 1 (Fig. S4), which shows a negative Cotton effect at 296 nm and a positive Cotton effect at 245 nm, which is similar to the calculated ECD spectra of the 9R,11R-isomers of penijanthinone A [16].

Fig. 4
figure 4

Determining the configurations of C-9 and C-4′ in 1. a Structures of penijanthinone A and 1. No NOE correlation was observed between H3-16 and H3-17 in either compound. b Comparing the chemical shift of the C-3′ protons in 1 with those of the corresponding protons in (2R,4S)-2,4-dimethylhexanoic acid and (2S,4S)-2,4-dimethylhexanoic acid [17]. The difference in the chemical shifts (Δδ) of the C-3′ geminal protons was 0.58 ppm. c 1H-1H COSY spectrum of dimethylhexanoic acid obtained by the hydrolysis of 1

We compared the 13C chemical shifts of the 2,4-dimethylhexanoate unit in 1 with literature values for (2R,4S)-2,4-dimethylhexanoic acid and (2S,4S)-2,4-dimethylhexanoic acid (Table 1) to determine the configuration of C-4′ [17]. The chemical shifts of 1 are almost identical to those of (2S,4S)-2,4-dimethylhexanoic acid, whereas a difference was noted between the shift assigned to C-7′ in 1 (δ 17.8) and that of the analogous carbon in (2R,4S)-2,4-dimethylhexanoic acid (δ 16.9) (underlined in Table 1). In addition, H-3′ in 1 exhibits a 1H chemical shift that is in good agreement its counterpart in (2S,4S)-2,4-dimethylhexanoic acid (Fig. 4b). Furthermore, 1 was hydrolyzed with NaOH to generate 2,4-dimethylhexanoic acid, and the crude extract was directly analyzed using 1H-1H COSY spectroscopy (Fig. 4c), which revealed that H-2 in the carboxylic acid resonates at 1.73 ppm, identical to that reported for the 2S,4S-isomer. Based on these results, we propose that vanitaracin A has the 2′S,4′S configuration. Schmidt et al. reported a 1H NMR method for assigning the relative configuration of a 1,3-dimethylalkyl chain [18]. This method is based on the difference in the chemical shifts (Δδ) of the geminal methylene protons located between the two stereogenic centers. The relative configuration is anti when Δδ lies between 0.0 and 0.1 ppm, and is syn when Δδ > 0.4 ppm. The difference in the chemical shifts (Δδ) of the C-3′ geminal protons of vanitaracin A was determined to be 0.58 ppm (> 0.4 ppm), which suggests a syn relationship (Fig. 4b). Hence, we conclude that vanitaracin A has the 2S,4R,9R,11R,2′S,4′S absolute configuration.

Table 1 13C NMR data for hexanoate moiety in 1, (2 R,4 S)-2,4-dimethylhexanoic acid, and (2 S,4 S)-2,4-dimethylhexanoic acid in CDCl3

In summary, in this study we revealed the absolute configuration of each stereogenic center in vanitaracin A, which will further stimulate research into vanitaracin A and related natural products. The synthesis of vanitaracin A is expected to lead to the identification of more potent analogs and the development of novel anti-HBV agents. In this study, we determined configurations by combining the crystalline sponge method and NMR spectroscopy. Similar combinational methods have already been used to determine structures and configurations and are becoming standard in the field [19,20,21].

Materials and methods

General experimental procedure

Optical rotations were recorded on a JASCO P-2200 digital polarimeter (Jasco Corp., Tokyo, Japan) at room temperature. ECD spectra were recorded in MeOH at a concentration of 1.0 × 10−4 M and 23 °C using a JASCO J-820 CD spectrometer with 5-mm path-length quartz cuvettes. UV spectra were obtained using a UVmini-1240 spectrophotometer (Shimadzu Corp., Kyoto, Japan). Infrared (IR) spectra were recorded on a JASCO FT/IR-4600 spectrophotometer (Jasco Corp.) and reported in wavenumbers (cm−1). 1H and 13C NMR spectra were recorded on a Bruker 400 MHz spectrometer (Avance DRX-400; Bruker, Billerica, MA, USA) in CDCl3 and referenced against TMS and CDCl3, respectively. Chemical shifts are expressed in ppm and coupling constants (J) are expressed in Hz. Mass spectra were obtained on a JEOL JMS-700 high-resolution double-focusing mass spectrometer using fast atom bombardment (FAB), or a Sciex X500R quadrupole time-of-flight (QTOF) high-resolution mass spectrometer using electrospray ionization (ESI). Analytical TLC was performed using precoated silica gel 60 F254 plates (Merck, Darmstadt, Germany). Silica gel 60 N (Kanto Chemical, Tokyo, Japan) was used for silica-gel column chromatography.

Isolating vanitaracin A (1)

Vanitaracin A was isolated from the culture broth of the fungus Talaromyces sp., as described previously [10].

Vanitaracin A (1)

ECD (c 1.0 × 10−4, MeOH) Δε (nm) −7.9 (296), +6.5 (245); 1H NMR (400 MHz, CDCl3) δ 7.06 (1H, s, H-7), 6.47 (1H, brs, OH), 4.24 (1H, m, H-2), 3.74 (1H, dd, J = 4.6, 17.8 Hz, H-1), 3.44 (1H, brs, OH), 3.24 (1H, m, H-4), 2.75 (1H, m, H-1), 2.74 (1H, m, H-2′), 2.15 (1H, m, H-3), 1.76 (1H, m, H-3′), 1.73 (1H, m, H-3), 1.71 (3H, s, H-16), 1.51 (1H, m, H-4′), 1.50 (3H, s, H-17), 1.41 (3H, d, J = 7.1 Hz, H-15), 1.35 (1H, m, H-5′), 1.27 (3H, d, J = 7.0 Hz, H-7′), 1.18 (1H, m, H-3′), 1.16 (1H, m, H-5′), 0.93 (3H, d, J = 6.7 Hz, H-8′), 0.89 (3H, t, J = 7.4 Hz, H-6′); 13C NMR (100 MHz, CDCl3) δ 206.2 (C-10), 191.7 (C-12), 175.6 (C-1′), 158.7 (C-6), 143.9 (C-8), 138.5 (C-14), 129.2 (C-5), 120.2 (C-13), 110.2 (C-7), 83.3 (C-11), 76.1 (C-9), 66.2 (C-2), 40.8 (C-3′), 37.0 (C-3), 36.4 (C-2′), 36.0 (C-1), 34.9 (C-16), 31.9 (C-4′), 29.3 (C-5′), 27.1 (C-4), 23.2 (C-17), 22.1 (C-15), 19.2 (C-8′), 17.8 (C-7′), 11.1 (C-6′).

Methylating vanitaracin A

K2CO3 (4.8 mg) and MeI (0.2 ml) were added to a solution of vanitaracin A (5.2 mg) in acetone (1 ml) and the mixture was stirred at room temperature for 48 h. The reaction mixture was diluted with EtOAc (20 ml) and washed with H2O (3 × 10 ml). The crude product was purified by silica-gel column chromatography with hexane/EtOAc as the eluent (4:1 → 0:1) to give methyl ether 2 (2.6 mg, 48%).

Methyl ether 2

[α]18D –71 (c 0.12, CHCl3); UV λMeOH max nm (ε) 285 (10,000), 217 (13,000), 204 (14,000); IR νmax (film) cm−1 3462, 2961, 2929, 2875, 1733, 1699, 1585, 1460, 1273, 1105, 1034; HRMS (FAB) m/z 461.2537 [M + H]+ (calcd. for C26H27O7, 461.2539); 1H NMR (400 MHz, CDCl3) δ 7.16 (1H, s), 4.24 (1H, m), 3.96 (3H, s), 3.76 (1H, dd, J = 4.4, 17.5 Hz), 3.41 (1H, brs), 3.28 (1H, m), 2.77-2.69 (2H, m), 2.13 (1H, m), 1.81-1.71 (1H, m), 1.73 (3H, s), 1.50 (3H, s), 1.50 (1H, m), 1.37(3H, d, J = 7.0 Hz), 1.34-1.26 (2H, m), 1.27 (3H, d, J = 7.0 Hz), 1.21-1.13 (2H, m), 0.93 (3H, d, J = 6.6 Hz) 0.89 (3H, t, J = 7.4 Hz); 13C NMR (100 MHz, CDCl3) δ 206.5, 191.8, 175.5, 161.6, 144.2, 137.2, 130.9, 120.3, 104.8, 83.2, 76.3, 66.1, 55.6, 40.8, 37.0, 36.4, 35.9, 35.5, 31.9, 29.4, 27.0, 23.4, 22.4, 19.2, 17.9, 11.1.

The (R)- and (S)-MTPA esters of methyl ether 2

Triethylamine (0.01 ml), DMAP (0.5 mg), and (R)-MTPACl (0.01 ml) were added to a solution of 2 (2.0 mg) in CH2Cl2 (1.0 ml), and the mixture was stirred at room temperature for 1 h. The reaction mixture was diluted with EtOAc (20 ml) and washed with H2O (3 × 10 ml). The crude product was purified by silica-gel column chromatography using hexane/EtOAc as the eluent (4:1 → 1:1) to afford (S)-MTPA ester 3a (2.2 mg, 75%). (R)-MTPA ester 3b was prepared in a similar manner using (S)-MTPACl.

(S)-MTPA ester 3a

HRMS (ESI) m/z 677.2933 [M + H]+ (calcd. for C36H44F3O9, 677.2932); 1H NMR (400 MHz, CDCl3) δ 7.50-7.48 (2H, m, aromatic signals of MTPA), 7.38-7.32 (3H, m, aromatic signals of MTPA), 7.17 (1H, s, H-7), 5.62 (1H, m, H-2), 3.94 (3H, s, OMe), 3.66 (1H, dd, J = 5.2, 18.9, Hz, H-1), 3.42 (3H, s, OMe of MTPA), 3.25 (1H, m, H-4), 3.04 (1H, m, H-1), 2.75 (1H, m, H-2′), 2.10 (1H, m, H-3), 1.90 (1H, m, H-3), 1.77 (1H, m, H-3′), 1.72 (3H, s, H-16), 1.56-1.45 (1H, m, H-4′), 1.42 (3H, s, H-17), 1.40-1.30 (1H, m, H-5′), 1.27 (3H, d, J = 7.0 Hz, H-7′), 1.23-1.12 (2H, m, H-3′ and H-5′), 1.00 (3H, d, J = 7.1 Hz, H-15), 0.94 (3H, d, J = 6.6 Hz, H-8′), 0.90 (3H, t, J = 7.4 Hz, H-6′).

(R)-MTPA ester 3b

HRMS (FAB) m/z 677.2935 [M + H]+ (calcd. for C36H44F3O9, 677.2937); 1H NMR (400 MHz, CDCl3) δ 7.47-7.45 (2H, m, aromatic signals of MTPA), 7.35-7.33 (3H, m, aromatic signals of MTPA), 7.15 (1H, s, H-7), 5.68 (1H, m, H-2), 3.94 (3H, s, OMe), 3.61 (1H, dd, J = 4.8, 19.3 Hz, H-1), 3.43 (3H, s, OMe of MTPA), 3.29 (1H, m, H-4), 2.91 (1H, m, H-1), 2.74 (1H, m, H-2′), 2.16 (1H, m, H-3), 2.00 (1H, m, H-3), 1.78 (1H, m, H-3′), 1.71 (3H, s, H-16, 1.56-1.47 (1H, m, H-4′), 1.39-1.30 (1H, m, H-5′), 1.35 (3H, s, H-17), 1.27 (3H, d, J = 7.1 Hz, H-7′), 1.24-1.14 (2H, m, H-3´ and H-5′), 1.15 (3H, d, J = 7.1 Hz, H-15), 0.94 (3H, d, J = 6.6 Hz, H-8′), 0.90 (3H, t, J = 7.4 Hz, H-6′).

Structural analysis of vanitaracin A (1) using the crystalline sponge method

Crystalline sponge analysis was conducted by Mitsui Chemical Analysis & Consulting Service, Inc. (Chiba, Japan) as described previously [14, 15]. Briefly, a single crystal of the porous [(ZnI2)3(tpt)2(solvent)x]n complex (crystalline sponge, tpt, tris(4-pyridyl)-1,3,5-triazine) immersed in n-hexane (45 μl) in a vial was treated with a chloroform solution of 1 (1 μl; 1 μg/μl) for 18 h at 50 °C to enable the uptake of 1 into the pores of the crystalline sponge. Single-crystal X-ray diffractometry (Bruker AXS D8 VENTURE) was performed using a Cu Kα X-ray source. The sample crystal was cooled to 93 K using a cold nitrogen stream.

Crystallographic data for 1

C84.50H58I12N24O3Zn6, M = 3372.57, monoclinic C2, a = 34.3218(17) Å, b = 15.0772(6) Å, c = 28.5217(16) Å, β = 98.063(4)°, V = 14613.4(12) Å3, Z = 4, GoF=1.039, R1 = 0.1067, wR2 = 0.3263. Flack parameter (Parsons’ method): 0.329(7).

Crystallographic data were refined after determining the configurations of all stereogenic centers. The crystallographic information file (CIF) has been deposited with the Cambridge Crystallographic Data Center (2114744).

Hydrolysis of 1

NaOH (0.2 M, 200 μl) was added to a solution of 1 (2.3 mg) in MeOH (500 μl) and the mixture was stirred at room temperature for 3 h. The MeOH in the reaction mixture was removed under a stream of nitrogen gas, and the aqueous solution was washed with EtOAc. The aqueous layer was acidified with 0.1 M HCl to pH 2–3 and then extracted with EtOAc (3 ×). The EtOAc layers were combined and the solvent was removed under a stream of nitrogen gas. The presence of dimethylhexanoic acid was confirmed by ESI-MS (Fig. S9). The crude extract was dissolved in CDCl3 and subjected to 1H-1H COSY spectroscopy (Fig. S10).