Anthraquinones from the Aerial Parts of Rubia cordifolia with Their NO Inhibitory and Antibacterial Activities

The present study aimed to identify the composition of the aerial parts of Rubia cordifolia L. A chemical investigation on the EtOAc extracts from the aerial parts of Rubia cordifolia resulted in the isolation of four new anthraquinones, namely Cordifoquinone A–D (1–4), along with 16 known anthraquinones. Their structures were elucidated on the basis of NMR and HR-ESIMS data. All isolates were assessed for their inhibitory effects on NO production in LPS-stimulated RAW 264.7 macrophage cells. Compounds 1, 3 and 10 exhibited significant inhibitory activities with IC50 values of 14.05, 23.48 and 29.23 μmol·L−1, respectively. Their antibacterial activities of four bacteria, Escherichia coli (ATCC 25922), Staphylococcus aureus subsp. aureus (ATCC 29213), Salmonella enterica subsp. enterica (ATCC 14028) and Pseudomonas aeruginosa (ATCC 27853), were also evaluated. Our results indicated that the antibacterial activity of these compounds is inactive.


Introduction
Infectious diarrhea (ID) is a kind of diarrhea caused by multiple pathogens and factors [1]. In 2012, a systematic analysis published in The Lancet showed that infectious diarrhea was the second leading cause of death in children under five years of age worldwide [2]. According to the pathogenesis, it can be divided into inflammatory diarrhea and secretory diarrhea, the former mainly caused by bacterial infection [3]. When the bacteria infect the human body, the pathogens further invade the intestinal mucosa and cause an inflammatory response that can lead to diarrhea. Nitric oxide (NO) is a bioactive molecule with extensive and important biological regulatory functions, which plays an important role in inflammation, tumors, and the cardiovascular system [4,5]. NO has been shown to be synthesized through the L-arginine/iNOS/NO pathway. In this pathway, NO and the amino acid L-citrulline are synthesized from the amino acid L-arginine by inducible nitric oxide synthase (iNOS) [4]. The excessive production of NO plays an important role in inflammatory response. Wink et al. found that when inflammation occurs, immune cells produce large amounts of inducible iNOS, which further produces NO for the immune response [6,7]. Therefore, inhibition of NO production is a direct indicator to verify the anti-inflammatory activity of compounds.
Worldwide, it is common for people to use medicinal plants to treat diarrhea [8][9][10]. Researchers have been working to find natural products with antidiarrheal properties from medicinal plants [11][12][13]. Rubia cordifolia Linn. is a grassy climbing vine herb that belongs to the genus Rubia and is found in China, India, Korea, Japan and the Far East of Russia, growing on sparse forests, forest margins and grasslands [14]. It is a well-known medicinal plant in Traditional Chinese Medicine (TCM). Rubiae Radix et Rhizoma, the dried root and rhizome of R. cordifolia, was first recorded in "Shennong Bencaojing" and used for hemorrhage syndrome and arthralgia. The aerial parts of R. cordifolia used for hemorrhage syndrome and diarrhea were recorded in "Lüchanyan Materia Medica", another classic book on medicinal plants. In modern studies, R. cordifolia has been found to have a wide range of pharmacological activities. In 1983, Itokawa et al. found that a methanol extract of R. cordifolia had obvious anticancer activity [15]. In 2017, Gong et al. found that the water extract of R. cordifolia could treat senna leaf-induced diarrhea in mice [16]. In 2020, Wang et al. found that the ethanol extract of R. cordifolia could improve arachidonic acid metabolism by inhibiting COX-2 and cPLA2 [17]. Because of its excellent efficacy, R. cordifolia has been listed in the "Chinese Pharmacopeia" and is to this day [18]. Previous chemical studies revealed that the plant contains anthraquinones [19,20], naphthoquinones [20], terpenoids [21], cyclic hexapeptides [22] and lignans [23], some of which have antibacterial [24], anti-inflammatory [25] and antitumor activities [26].
At present, the research on R. cordifolia is mainly focused on the roots and rhizomes, and less on the aerial parts. Therefore, in order to find natural products with antibacterial or anti-inflammatory activities from this plant, we systematically studied the aerial parts of R. cordifolia. Our previous investigation identified eleven lignans from the aerial parts of R. cordifolia for the first time [23]. As a continuing investigation to find more undescribed and bioactive constituents from R. cordifolia, four new anthraquinones (1-4) along with 16 known anthraquinones (5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) were isolated from the aerial parts of R. cordifolia. Compounds 1-5, 9, 10, 13, 15-17 and 20 were isolated from this plant for the first time. This study describes the isolation, structure elucidation, NO inhibitory and antibacterial assay of these anthraquinones.

Results and Discussion
Compound 1 was obtained as a yellow powder, and the molecular formula was determined as C 16 7.71 (1H, dd, J = 7.9, 1.8 Hz)). The other benzene ring showed two orthodox-coupled aromatic proton resonances at (δ H 8.04 (1H, d, J = 8.5 Hz) and 7.38 (1H, d, J = 8.5 Hz)). Moreover, aromatic methyl group δH 2.55 (3H, s), aromatic methoxy group δ H 3.95 (3H, s) and phenolic hydroxyl group δ H 9.40 (1H, s) protons were also observed. The 13 C NMR exhibited 16 carbon signals, which consisted of 14 carbons of anthraquinone nucleus, a methoxy and a methyl. The HMBC ( Figure 1) correlations of H-7 with C-5, 5-OMe with C-5 and H-8 with C-6/C-9 revealed that the methoxy group and hydroxyl group were connected at C-5 and C-6, respectively. In addition, the methyl group was deduced to be located at C-2 based on the HMBC correlations of 2-Me with C-1/C-2/C-3 and H-1 with C-3/C-9. Thus, the structure of compound 1 was established as 2-methyl-5-methoxy-6-hydroxyanthraquinone and named Cordifoquinone A, as shown in Figure 1.
Compound 2 was obtained as a yellow powder, and the molecular formula was determined to be C 16 Figure 1) correlations from the aromatic methine protons to C-1/C-2/C-3 and from H-4 to C-2/C-10, together with the 1 H NMR data, implied the presence of a phenolic hydroxyl group at the C-1 and a hydroxymethyl group at the C-2 position. Moreover, the methoxy group was deduced to be located at C-7, and a phenolic hydroxyl group was deduced to be located at C-6, based on the HMBC correlations of the 7-OMe with C-7, H-5 with C-7/C-10 and H-8 with C-6/C-9. Therefore, the structure of compound 2 was confirmed as 1,6-dihydroxy-2-(hydroxymethyl)-7-methoxyanthraquinone and named Cordifoquinone B.     (Table 1), 3 and 2 were found to have similar structures, except that the phenolic hydroxyl group on the hydroxymethyl side was moved to the methoxy side. The 1 H NMR data exhibited resonances attributed to a 1, 3, 4-substituted benzene ring (δ H 8.12 (1H, d, J = 1.7 Hz), 8.06 (1H, d, J = 7.9 Hz) and 7.76 (1H, dd, J = 7.9, 1.7 Hz)), an aromatic proton singlet at δ H 7.19 (1H, s), an aromatic methine proton signals at δ H 4.67 (2H, s), an aromatic methoxy group δ H 3.83 (3H, s) and an aromatic hydroxyl group δ H 13.03 (1H, s). The methoxy group was deduced to be located at C-7 based on the HMBC (Figure 1) correlations of 7-OMe with C-7 and H-5 with C-7/C-10. Moreover, the HMBC correlations from the aromatic methine protons to C-1/C-2/C-3, from H-4 to C-2/C-10 and from H-1 to C-9 suggested that the hydroxymethyl group was directly connected to C-2. Based on the data mentioned above, the structure of compound 3 was defined to be 6,8-dihydroxy-2-(hydroxymethyl)-7-methoxyanthraquinone and named Cordifoquinone C.
Compound 4 was obtained as a yellow powder and had the same molecular formula of C 16 (Table 1) were very similar to those of compound 3. The difference between them was that two phenolic hydroxyl groups and the methoxy group position of compound 4 were connected to C-5/C-7 and C-6, respectively, which were supported by the HMBC (Figure 1) correlations from the aromatic methine protons to C-1/C-2/C-3, H-1 to C-9/C-3, 6-OMe to C-6 and H-8 to C-6/C-9. Accordingly, the whole structure of compound 4 was finally determined as 5,7-dihydroxy-2-(hydroxymethyl)-6-methoxyanthraquinone and named Cordifoquinone D.
Additionally, 16 known anthraquinones, pustuline [27], 5,7-dihydroxy-6-methoxy-2methylanthraquinone [28], soranjidiol [29], 1,6-dihydroxy-5-methoxy-2-methylanthraquinone [29], 1,5-dihydroxy-6-methoxy-2-methylanthraquinone [29], copareolatin 6-methylether [30], rubiadin [29], 2-methyl-1,3,6-trihydroxyanthraquinone [19], 7-hydroxy-2-(hydroxymethyl)-6methoxyanthraquinone [31], 1-hydroxy-2-(hydroxymethyl)anthraquinone [32], 1,6-dihydroxy-2-(hydroxymethyl)-5-methoxyanthraquinone [33], 1,6-dihydroxy-2-(hydroxymethyl)-5,7dimethoxyanthraquinone [34], 1,3-dihydroxy-2-hydroxymethyl-6-methoxyanthraquinone [35], 1,3-dihydroxy-2-(hydroxymethyl)-5,6-dimethoxyanthraquinone [36], 3,6-dihydroxy-2-(hydroxymethyl)-1-methoxyanthraquinone [37] and alizarin 1-methyl ether [38] were isolated from the aerial parts of R. cordifolia, as shown in Table 2. Their structures were elucidated by comparison of the NMR and ESI data with those reported in the literature. It is worth mentioning that anthraquinones in R. cordifolia are considered to be of the Rubia type; rings A and B are derived from horismite and α-ketoglutarate [39,40], whereas ring C is formed from isopentenyl diphosphate (IPP) [41]. Anthraquinone biosynthesized by this pathway has a methylation site in the C ring [42]. The aryl methyl group at site 2 can be further oxidized to hydroxymethyl. So, we hypothesized that compounds 1-20 might be synthesized by this pathway, as shown in Scheme 1.   (Table 3), respectively, which were more potent than the positive control. By comparing the structure and activity of compounds 1, 3, 10, we found that the orthoposition of methoxyl and phenolic hydroxyl groups in the A ring of anthraquinone was essential for the NO inhibitory activity, and the hydroxyl group at position C-6 or C-7 provided a greater contribution. It is known that the synthesis of NO requires the addition of superoxide anions [43]. So, we hypothesized that the hydroxyl group in anthraquinone may capture superoxide anions, which are necessary for the synthesis of NO. Compound 1 showed stronger NO inhibitory activity than compounds 3 and 10, indicating that the activity was not necessarily related to the number of hydroxyl groups but may be related to the position of hydroxyl groups. We hypothesized that it All isolated compounds 1-20 were evaluated for their inhibitory effects on NO production in LPS-activated RAW264.7 macrophage cells. Cell viability was first examined by MTT assay to exclude false positive results caused by the potential cytotoxicity of the tested compounds. NG-Methyl-L-arginine acetate salt, an inhibitor of NO synthase, was used as a positive control (IC 50 42.36 ± 2.47 µmol·L −1 ). As a result, compounds 1, 3 and 10 showed significant inhibitory effects against NO production, with IC 50 values of 14.05 ± 0.48, 23.48 ± 1.05 and 29.23 ± 0.34 µmol·L −1 (Table 3), respectively, which were more potent than the positive control. By comparing the structure and activity of compounds 1, 3, 10, we found that the orthoposition of methoxyl and phenolic hydroxyl groups in the A ring of anthraquinone was essential for the NO inhibitory activity, and the hydroxyl group at position C-6 or C-7 provided a greater contribution. It is known that the synthesis of NO requires the addition of superoxide anions [43]. So, we hypothesized that the hydroxyl group in anthraquinone may capture superoxide anions, which are necessary for the synthesis of NO. Compound 1 showed stronger NO inhibitory activity than compounds 3 and 10, indicating that the activity was not necessarily related to the number of hydroxyl groups but may be related to the position of hydroxyl groups. We hypothesized that it may be due to the formation of intramolecular hydrogen bonds between hydroxyl group at position C-5 or C-8 and ketone carbonyl group at position C-9 or C-10, which makes the compound structure more stable and reduces its ability to capture superoxide anions. Through this speculation, it is reasonable to explain why compound 3 showed slightly higher NO inhibitory activity than compound 10. In addition, all compounds were evaluated for their antibacterial activities against four bacteria: Escherichia coli (ATCC 25922), Staphylococcus aureus subsp. aureus (ATCC 29213), Salmonella enterica subsp. enterica (ATCC 14028), and Pseudomonas aeruginosa (ATCC27853). However, no antibacterial activity was observed for these compounds at concentrations below 128 µg·mL −1 in this bioassay.
Molecules 2022, 27, x FOR PEER REVIEW 6 of 12 may be due to the formation of intramolecular hydrogen bonds between hydroxyl group at position C-5 or C-8 and ketone carbonyl group at position C-9 or C-10, which makes the compound structure more stable and reduces its ability to capture superoxide anions. Through this speculation, it is reasonable to explain why compound 3 showed slightly higher NO inhibitory activity than compound 10. In addition, all compounds were evaluated for their antibacterial activities against four bacteria: Escherichia coli (ATCC 25922), Staphylococcus aureus subsp. aureus (ATCC 29213), Salmonella enterica subsp. enterica (ATCC 14028), and Pseudomonas aeruginosa (ATCC27853). However, no antibacterial activity was observed for these compounds at concentrations below 128 μg·mL −1 in this bioassay.

Plant Materials
The aerial parts of Rubia cordifolia Linn. Were collected in June 2020 from Huaiyuan County, Anhui Province, China, and authenticated by Prof. Shou-Jin Liu from Anhui University of Traditional Chinese Medicine (Anhui, China). A voucher specimen (No. 20200601QC) was deposited at the Chinese Medicine Resource Center, Anhui University of Traditional Chinese Medicine (ACM).

Extraction and Isolation
The air-dried and powdered aerial parts of R. cordifolia (22 kg) were extracted with 95% EtOH at room temperature. After removal of the solvent under reduced pressure, the EtOH extract (22 kg) was suspended in water and partitioned with EtOAc. The EtOAc fraction (1.6 kg) was dissolved in EtOAc and chromatographed on silica gel column chromatography (Si CC), eluted with petroleum ether−acetone (1:0 to 0:1) to yield eight fractions, Fr. A-H (86 g, 60 g, 130 g, 123 g, 105 g, 178 g, 168 g, and 208 g). Fr. E (105 g) was dissolved in acetone and subjected to Si CC eluted with a gradient of CHCl 3 -MeOH (1:0 to 0:1) to obtain four major fractions, Fr. E1 to Fr. E4 (30 g, 8 g, 14 g, and 26 g). The extraction and separation process is shown in Figure 2.

NO Inhibitory Assay
Murine macrophage cell line RAW264.7 was obtained from Shanghai Cell Bank of Chinese Academy of Sciences. RAW264.7 cells were seeded in 96-well cell culture plates (1.5 × 10 5 cells/well), stimulated with 1 µg·mL −1 LPS (Sigma, St. Louis, MO, USA) for 18 h, and treated with serial dilutions of the compounds with a maximum concentration of 50 µM in triplicate during this period. NG-Methyl-L-arginine acetate salt (L-NMMA, Sigma), a well-known nitric oxide synthase (NOS) inhibitor, was used as a positive control [44]. NO production in the supernatant was assessed by Griess reagents (Reagent A and Reagent B, respectively, Sigma). The absorbance at 570 nm was measured with a microplate reader (Thermo, Waltham, MA, USA). The viability of RAW264.7 cells was evaluated by the MTS assay simultaneously to exclude the interference of the cytotoxicity of the test compounds. All tests were performed in triplicate, and the results were expressed as IC 50 values.

Antibacterial Assay
Escherichia coli (ATCC 25922), Staphylococcus aureus subsp. aureus (ATCC 29213), Salmonella enterica subsp. enterica (ATCC 14028), and Pseudomonas aeruginosa (ATCC 27853) were obtained from China General Microbiological Culture Collection Center. The inoculated strains and suspension were cultured for 24 h and adjusted to 0.5 McFarlane standard turbidity. A 96-well plate was prepared by injecting 100 µL of Muller Hinton broth/Luria-Bertani into each well. A volume of 100 µL of stock solution from the initial prepared samples was added to the first well. Then, 100 µL from their two-fold serial dilutions were transferred into consecutive wells, and 100 µL of the inoculums was added to achieve a final inoculum concentration of 5 × 10 5 CFU/mL. The final volume in each well was 200 µL. Ceftazidime (Yuanye Co. Ltd., Shanghai, China) and sodium penicillin G (Labgic Co. Ltd., Beijing, China) were used as positive controls. After incubation at 37 • C for 24 h, growth was monitored by a Microplate Reader at 625 nm (PowerWave XS, BioTek, Winooski, VT, USA) [45].

Conclusions
Four new anthraquinones (1-4) and 16 known anthraquinones (5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) were isolated from a 95% EtOH extract of aerial prat from R. cordifolia and elucidated by UV, IR, HR-ESIMS and NMR spectroscopic data. Among the isolated compounds, compounds 1, 3 and 10 displayed NO inhibitory activity with IC 50 values of 14.05, 23.48 and 29.23 µmol·L −1 . The structure and activity relationship analysis revealed that the orthoposition of methoxyl and phenolic hydroxyl groups in the A ring of anthra-quinone was essential for the NO inhibitory activity. The hydroxyl group at position C-6 or C-7 may have promoted the NO inhibitory activity, whereas the hydroxyl group at position C-5 or C-8 had the opposite effect. In addition, this series of anthraquinones showed no activity in tests of antibacterial activity. This study proved the reasonableness of using the aerial part of R. cordifolia as a medicine to treat diarrhea in Traditional Chinese medicine. It also provides scientific evidence and a foundation for the understanding of the anti-inflammatory effects and further utilization of R. cordifolia. In future research, both the extract and the natural products of R. cordifolia can be considered a drug or a health supplement to treat diarrhea.
Supplementary Materials: The following supporting information can be downloaded online, Figure S1: 1 H, 13 C and HMBC NMR spectra of compound 1; Figure S2: 1 H, 13 C and HMBC NMR spectra of compound 2; Figure S3: 1 H, 13 C and HMBC NMR spectra of compound 3; Figure S4