Sesquiterpenoids from the Florets of Carthamus tinctorius (Safflower) and Their Anti-Atherosclerotic Activity

(1) Background: The florets of Carthamus tinctorius L. are traditionally used as a blood-activating drug and can be used for the treatment of atherosclerosis, but no compounds with anti-atherosclerotic activity have been reported. (2) Methods: This study investigated the chemical compounds from the florets of C. tinctorius. Comprehensive spectroscopic techniques revealed their structures, and ECD calculations established their absolute configurations. Nile Red staining, Oil Red O staining, and cholesterol assessment were performed on these compounds and their aglycones for the inhibitory activity against the formation of foam cells induced by oxidized low-density lipoprotein (ox-LDL) in RAW264.7 macrophages. In addition, RAW264.7 macrophages were tested for their anti-inflammatory activity by measuring the inhibition of NO production caused by LPS. (3) Results: Five new sesquiterpenoids (1–5) isolated from the florets of C. tinctorius were identified as (–)-(1R,4S,9S,11R)-caryophyll-8(13)-en-14-ol-5-one (1), (+)-(1R,4R,9S,11R)-caryophyll-8(13)-en-14-ol-5-one (2), (–)-(3Z,1R,5S,8S,9S,11R)-5,8-epoxycaryophyll-3-en-14-O-β-D-glucopyranoside (3), (+)-(1S,7R,10S)-guai-4-en-3-one-11-O-β-D-fucopyranoside (4), and (–)-(2R,5R,10R)-vetispir-6-en-8-one-11-O-β-D-fucopyranoside (5). All compounds except for compound 3 reduced the lipid content in ox-LDL-treated RAW264.7 cells. Compounds 3 and 4 and their aglycones were found to reduce the level of total cholesterol (TC) and free cholesterol (FC) in ox-LDL-treated RAW264.7 cells. However, no compounds showed anti-inflammatory activity. (4) Conclusion: Sesquiterpenoids from C. tinctorius help to decrease the content of lipids, TC and FC in RAW264.7 cells, but they cannot inhibit NO production, which implies that their anti-atherogenic effects do not involve the inhibition of inflammation.


Introduction
Cardiovascular diseases have recently topped the list of causes of diseases in general, posing an immediate danger to human health. Atherosclerosis (AS) is a major pathological basis for many cardiovascular and cerebrovascular diseases, and macrophages play a major role in the plaque formation process of AS. In the early stages of AS development, macrophages phagocytose oxidize more low-density lipoprotein (ox-LDL) than they can metabolize, causing their massive accumulation of lipids and turning into foam cells. Then, foam cells gather together under the vascular endothelium to form a raised plaque, which further narrows or blocks blood vessels, thereby promoting the development of AS [1,2]. Therefore, inhibition of macrophage foaminess can effectively treat AS. which further narrows or blocks blood vessels, thereby promoting the development of AS [1,2]. Therefore, inhibition of macrophage foaminess can effectively treat AS.
Recently, it has been demonstrated that many traditional Chinese medicines (TCMs), especially the blood-activating and stasis-removing medicines, are effective in treating AS [3][4][5]. The florets of Carthamus tinctorius L. (safflower), a well-known TCM widely cultivated in Xinjiang and Sichuan, are described as an "essential medicine for promoting blood circulation" in Ben Cao Gang Mu. Modern pharmacological studies have shown that safflower is effective in treating cardiovascular diseases and inhibiting the development of AS [6][7][8][9]. Additionally, safflower is utilized as food coloring, functional food, and feedstuff to supple fatty acids, improve hair health, extend endurance, and reduce wrinkles [10][11][12][13][14]. Up to now, phytochemical studies on safflower have revealed more than 200 chemical compounds, mainly including terpenoids, flavonoids, alkaloids, organic acids, and polyacetylenes [15][16][17][18]. However, fewer than 10 sesquiterpenoids have been reported [18]. In this study, five new sesquiterpenoids (compounds 1-5) were identified from safflower, including caryophyllane-type, guaiane-type, and vetispirane-type sesquiterpenoids ( Figure 1). Interestingly, compounds 4 and 5 are rare sesquiterpenoid fucopyranosides. To our knowledge, only a small number of fucopyranosides have been discovered in nature, and most of them exist as flavone glycosides [19,20], triterpenoid saponins [21,22], and steroidal saponins [23]. Their inhibitory actions on ox-LDL-induced lipid accumulation were explored by Nile Red staining, Oil Red O staining, and cholesterol ELISA testing to evaluate their anti-atherosclerotic activity.

Structure Elucidation of Compounds 1-5
Compound 1 is a colorless oil, and its molecular formula was determined as C15H24O2 with four degrees of unsaturation according to HR-ESI-MS at m/z 259. 1666 (Table 2) and DEPT data of compound 1 detected 15 carbon signals in total, including two methyls, seven methylenes (one oxygenated and one olefinic), three methines, and three quaternary carbons (one ketonic and one olefinic). Based on these data, compound 1 is a caryophyllane-type sesquiterpenoid

Structure Elucidation of Compounds 1-5
Compound 1 is a colorless oil, and its molecular formula was determined as C 15 (Table 2) and DEPT data of compound 1 detected 15 carbon signals in total, including two methyls, seven methylenes (one oxygenated and one olefinic), three methines, and three quaternary carbons (one ketonic and one olefinic). Based on these data, compound 1 is a caryophyllane-type sesquiterpenoid similar to gibberosin P, but the hydroxymethyl functionality at C-11 in compound 1 takes the place of a 4-hydroxyvaleryl group in gibberosin P [24]. The planar structure of compound 1 was confirmed with the assistance of 1 H-1 H COSY and HMBC correlations as shown in Figure 2. In particular, the hydroxymethyl unit is located at C-11 based on the HMBC Nutrients 2022, 14, 5348 3 of 13 cross-peaks from H 2 -14 to C-1, C-10, C-11, and C-15. An NOESY experiment was applied to identify the relative configuration of compound 1. Correlations of H-9/H 3 -15 and H-1/H-4 and H 2 -14 ( Figure 3) suggested that H-9 and Me-11 were co-facial, whereas H-1, H-4, and hydroxymethyl-11 occupied the opposite face. By comparing the computed electronic circular dichroism (ECD) data of compound 1 with the experimental ECD data (Figure 4), the absolute configuration of compound 1 was revealed to be 1R,4S,9S,11R. Therefore, compound 1 was determined to be (-)-(1R,4S,9S,11R)-caryophyll-8(13)-en-14-ol-5-one.      Compound 3 has a molecular formula of C21H34O7, as determined by the HR-ESI-MS. The 1 H, 13 C, and DEPT NMR data of compound 3 suggested that it is also a caryophyllanetype sesquiterpenoid. The signals for an anomeric proton [δH 4.24 (d, J = 7.8 Hz)] and an     . Experimental and calculated ECD spectra of compounds 1-2 and 3a-5a. Based on the HR-ESI-MS data of compound 2, its chemical formula is similar to that of compound 1. The 1 H, 13 C, and 2D NMR spectra of compounds 2 and 1 showed that the planar structures of these two compounds are identical. Further analysis revealed that compound 2 was a C-4 epimer of compound 1 based on the NOESY correlations of H 2 -14/H-1, H-1/H 3 -12, and H-9/H 3 -15 ( Figure 3). Moreover, ECD calculations showed that the absolute configuration of compound 2 is 1R,4R,9S,11R ( Figure 4). Therefore, compound 2 was identified as (+)-(1R,4R,9S,11R)-caryophyll-8(13)-en-14-ol-5-one.  13 C, and DEPT NMR data of compound 3 suggested that it is also a caryophyllanetype sesquiterpenoid. The signals for an anomeric proton [δ H 4.24 (d, J = 7.8 Hz)] and an anomeric carbon (δ C 104.7) revealed the existence of a β-glucosyl [25], indicating that compound 3 is a caryophyllane glycoside. In addition, the 1 H and 13 C NMR data (Tables 1 and 2 , and an olefinic quaternary carbon (δ C 140.8) in the aglycone unit. Thus, the existence of an epoxy moiety and a trisubstituted double bond was deduced based on the molecular formula. Comparison of the NMR data between compound 3 and 5α,8α-epoxycaryophyll-3-ene [26] suggested that an additional β-glucose moiety and an oxygenated methene group in compound 3 replaced a methyl group (C-14) in 5α,8αepoxycaryophyll-3-ene, and this deduction was confirmed by the HMBC correlation from H-1 to C-14.
In the NOESY spectrum of compound 3, cross-peaks of H 2 -14/H-1 and H-9/H 3 -15 demonstrated the same orientation of H 2 -14 and H-1, whereas H-9 and H 3 -15 had the opposite orientation. The orientation of the 5-O-8 bridge was established by an NOE signal of H-9/H 3 -13. In addition, an NOE interaction between H-3 and H 3 -12 disclosed a Z-geometry for ∆ 3 . To verify the absolute configuration of compound 3, it was subjected to enzymatic hydrolysis, which yielded an aglycone 3a and a D-glucose. Then, ECD calculation was carried out to resolve the absolute configuration of 3a. As depicted in Figure 4, the calculated ECD spectrum of (1R,5S,8S,9S,11R)-3a and the experimental ECD spectrum of 3a were in good agreement. Therefore, it was concluded that compound 3 is The molecular formula of compound 4 was identified as C 21 .3)]. Comprehensive analysis of the 1 H-1 H COSY, HSQC, and HMBC spectra indicated that the planar structure of compound 4 is similar to (1R,7R,10S)-11-O-β-D-glucopyranosyl-4-guaien-3-one [27], except for the glycosyl unit C-11. The glycosyl unit in compound 4 was further determined to be β-D-fucosyl due to the coupling constants (J 1 ,2 = 7.8 Hz, J 2 ,3 = 9.6 Hz, J 3 ,4 = 3.6 Hz, and J 4 ,5 ≈ 0 Hz), together with the 13   The HR-ESI-MS of compound 5 revealed that it is a structural isomer of compound 4. Comparative analysis of their 1 H and 13 C NMR data exhibited that they share similar functional groups such as an α,β-unsaturated ketone unit, four methyl groups, and a β-D-fucosyl moiety. However, as shown in Figure 2, 2D NMR data analysis pointed out that compound 5 is a vetispirane-type sesquiterpenoid [29]. The NOESY correlations of H 3 -15/H-1a, H-10/H 2 -4, and H-2/H 3 -15 ( Figure 3) determined the relative structure of compound 5. Similar to compound 4, the aglycone 5a was produced by enzymatic hydrolysis of compound 5, and 2R,5R,10R configuration was assigned to 5a according to ECD calculations (Figure 4). Therefore, the structure of compound 5 was established as (−)-(2R,5R,10R)-vetispir-6-en-8-one-11-O-β-D-fucopyranoside.

Anti-Atherosclerotic Activity of Compounds 1-5 and Aglycones 3a-5a
The anti-atherosclerotic activity of the isolates (1)(2)(3)(4)(5) and their aglycones (3a-5a) was evaluated by detecting their inhibitory effects against RAW264.7 cell foaminess induced by ox-LDL (Yiyuan Biotechnologies, Guangzhou, China) [1,30]. First, a CCK-8 assay was applied to evaluate the effects of the eight compounds on RAW264.7 cell viability, and all of the compounds were found to be noncytotoxic. Then, they were tested for anti-macrophage foaming activity by reducing the content of lipid in RAW264.7 cells treated by ox-LDL. As depicted in Figure 5, all compounds, except for compound 3, showed significant activity by Nile Red staining. In particular, the effects of compounds 1, 2, 3a, 4a, 5, and 5a were equal to or superior to the effect of the positive control (25 µM of simvastatin). Interestingly, all of the aglycones (3a-5a) had better effects than their glycosides (3)(4)(5), which suggests that glycosidation of the sesquiterpenoids of safflower may result in a decrease in their anti-atherosclerotic activity. Although compound 4 showed weak activity in Nile Red staining, a significant effect was observed for compound 4 in Oil Red O staining ( Figure 6). Specifically, compared with the control group, the model group collected more Oil Red O-stained lipid. In the drug group, compound 4 significantly decreased Oil Red O-stained lipid accretion.
To further explore the anti-atherosclerotic activity of compounds 3, 3a, 4, and 4a, the total cholesterol (TC) and free cholesterol (FC) levels in foamy macrophages were assessed by ELISA (Shanghai Keshun Science and Technology Co., Ltd., Shanghai, China). As shown in Figure 7, compounds 3, 3a, 4, and 4a significantly decreased the levels of TC and FC in ox-LDL-treated RAW264.7 cells in a dose-dependent way (p < 0.01 vs. model group). In addition, all compounds were investigated for their ability to inhibit lipopolysaccharideinduced NO production in RAW264.7 cells. The results showed that none of the compounds had an anti-inflammatory effect. Thus, it can be inferred that the safflower sesquiterpenoids do not exert their anti-atherogenic effects through the anti-inflammatory pathway [31]. Unfortunately, no further investigations on the molecular mechanisms were carried out due to the limited sample quantities of the isolates. sesquiterpenoids do not exert their anti-atherogenic effects through the anti-inflammatory pathway [31]. Unfortunately, no further investigations on the molecular mechanisms were carried out due to the limited sample quantities of the isolates.

General Experimental Procedures
General experimental instruments and materials are shown in Text S1, Supplementary Materials.

Plant Material
The florets of C. tinctorius were gathered from Luole Village, Sichuan Province, China. The sample was authenticated by Dr. Ji-hai Gao (Chengdu University of TCM, Sichuan, China). A voucher specimen (No. CT-20180608) was placed at the State Key Laboratory of Southwestern Chinese Medicine Resources, Chengdu University of TCM.

Extraction and Isolation
The florets of C. tinctorius (50 kg) were decocted twice with 12 times the quantity of H2O for 1 h each time. The extract was concentrated under reduced pressure in a big rotary evaporator (10 L) at 55 °C to produce a residue (16 kg). The residue was separated on a D-  of compounds 3, 3a, 4, and 4a on the intracellular content of FC. Results are expressed as the mean ± SD (n = 3). Model group was subject to 75 µg/mL ox-LDL. ## p < 0.01 vs. control group; ** p < 0.01 vs. model group.

General Experimental Procedures
General experimental instruments and materials are shown in Text S1, Supplementary Materials.

Plant Material
The florets of C. tinctorius were gathered from Luole Village, Sichuan Province, China.

Cell Viability Assay
RAW264.7 cells were grown in DMEM supplemented with 10% FBS under a humidified environment at 37 • C and 5% CO 2 . The cells were planted in 96-well plates (3 × 10 4 cells/well) and incubated with compounds 1-5 and 3a-5a (25,50, and 100 µM) for one day. Then, the cells were treated by CCK-8 solution (10 µL/well) and incubated for 1 h. Under 450 nm, the absorbance of each well was tested to evaluate the viability of the cells.

Nile Red Staining
RAW264.7 cells were placed in 96-well plates (3 × 10 4 cells/well). Then, the cells were treated with ox-LDL (75 µg/mL), together with compounds 1-5 and 3a-5a (25,50, and 100 µM). After incubation for 24 h, the cells were kept for 30 min at 37 • C with 4% paraformaldehyde, then rinsed with PBS. The cells were hatched with freshly prepared Nile Red staining solution (2 µg/mL) for 30 min. Finally, PBS was used to wash the cells, and the fluorescence intensity was measured at 530 nm/590 nm.

Oil Red O Staining
RAW264.7 cells were placed in 24-well plates (3 × 10 5 cells/well). Then, the cells were treated with ox-LDL (75 µg/mL) and compound 4 (25, 50, and 100 µM). After incubation for 24 h, the cells were kept for 30 min at 37 • C with 4% paraformaldehyde, followed by rinsing with PBS. The cells were washed with 50% isopropanol for 20 s, and incubated with fresh-filtered Oil Red O solution (60% saturated Oil Red O/40% deionized water) for 30 min. The cells were washed with alternate rinses of 50% isopropanol and PBS to remove the floating colors and photographed for observation by fluorescence microscopy. Finally, the cells were washed with isopropanol, and the absorbance was observed at 550 nm.

ELISA
RAW264.7 cells (1 × 10 6 /well) were placed into six-well culture plates and treated by the above method. The contents of TC and FC were measured using the ELISA kits.

Conclusions
In conclusion, five new sesquiterpenoids were isolated from safflower, including caryophyllane-type, guaiane-type, and vetispirane-type sesquiterpenoids. The sesquiterpenoid types characterized herein have not been reported from C. tinctorius, and compounds 4 and 5 are rare sesquiterpenoid fucopyranosides. The sesquiterpenoids showed significant anti-atherosclerosis activity by decreasing the contents of lipid, TC, and FC in ox-LDL-treated RAW264.7 cells. Altogether, these new sesquiterpenoids not only enrich the diversity of sesquiterpenoids in safflower but also explain the effective material basis of safflower in treating AS.

Data Availability Statement:
The data presented in this study are available in the Supplementary Materials or can be provided by the authors.

Conflicts of Interest:
The authors declare no conflict of interest.