Neocornuside A–D, Four Novel Iridoid Glycosides from Fruits of Cornus officinalis and Their Antidiabetic Activity

Four previously undescribed iridoid glycosides neocornuside A–D (1–4), along with six known ones (5–10), were isolated from Cornus officinalis fruit. Their structures were elucidated by extensive spectroscopic (NMR, UV, IR, and MS) analysis and comparison with data reported in the literature. All isolates were assessed for their antidiabetic activity on the relative glucose consumption in insulin-induced insulin-resistant HepG2 cells. The results showed that compounds 1, 3, and 7 exhibited significant antidiabetic activities with EC50 values of 0.582, 1.275, and 0.742 μM, respectively. Moreover, compounds 1, 3, and 7 could improve the ability of 2-NBDG uptake of insulin-induced HepG2 cells.

Diabetes mellitus (DM) is a major health problem for the people all over the world. In recent years, the global incidence of diabetes has increased rapidly due to many factors such as the improvement of living standards, changes in dietary structure, increasingly stressful rhythm of life, a less active and more sedentary lifestyle, and diabetes has become the third most serious chronic disease threatening human health after tumors and cardiovascular disease [14,15]. It has been linked to oxidative stress, which arises mainly through oxidation, oxidative degradation of glycated proteins, and nonenzymatic protein glycation [14]. Plant products and their derivatives have been widely accepted to possess many pharmacological activities, such as anti-inflammatory, antimicrobial, anticancer and antidiabetic activity. Coptis chinensis Franch (Ranunculaceae) polysaccharide (CCPW) can produce antidiabetic activity in rats with T2DM through its antioxidative effect, which is closely related to the JNK/IRS1/PI3K pathway [16]. Three polysaccharides were extracted from Suillellus luridus (Suilu.A, Suilu.C, and Suilu.S) which exhibited significant antidiabetic activity in diabetic mice induced by streptozotocin [17]. Moreover, the methanolic extract of Geigeriaalata has antidiabetic activity and it is suggested that this antidiabetic activity is due to enhanced insulin secretion, modulation of β-cell function, and improvement of antioxidant status [18]. Thus, traditional medicines have been proved to be a vital source of future drugs to prevent and treat many diseases, including diabetes mellitus. Iridoids are one of the major and characteristic ingredients of Cornus officinalis. The previous study proved the reasonableness of using iridoids isolated from Cornus officinalis to treat diabetes [15,19]. C. officinalis extracts and pure compounds could ameliorate diabetes-associated damages and complications. Oral administration of loganin and morroniside decreased fasting blood glucose levels in diabetes mellitus mice. Ursolic acid exhibited the highest reactive oxygen species scavenging activity and α-glucosidase inhibitory activity [7]. In addition, loganic acid (LA) exhibited antioxidant properties in relation to STZ-induced DM. It may indicate LA as one of the plant components in the development of new drugs that will treat metabolic and functional disorders in leukocytes under diabetes [20]. In recent years, there is growing interest in the utilization of natural products as potential therapeutic agents for treating DM. Therefore, in order to find natural products with antidiabetic activity from this plant, we systematically studied the fruits of Cornus officinalis. In our recent study, four new iridoid glycosides (1)(2)(3)(4) and six known ones (5)(6)(7)(8)(9)(10) were obtained from Cornus officinalis ( Figure 1). In addition, the in vitro antidiabetic activity of the isolated compounds was evaluated.
Molecules 2022, 27, x FOR PEER REVIEW 2 of 12 antidiabetic activity in diabetic mice induced by streptozotocin [17]. Moreover, the methanolic extract of Geigeriaalata has antidiabetic activity and it is suggested that this antidiabetic activity is due to enhanced insulin secretion, modulation of β-cell function, and improvement of antioxidant status [18]. Thus, traditional medicines have been proved to be a vital source of future drugs to prevent and treat many diseases, including diabetes mellitus. Iridoids are one of the major and characteristic ingredients of Cornus officinalis. The previous study proved the reasonableness of using iridoids isolated from Cornus officinalis to treat diabetes [15,19]. C. officinalis extracts and pure compounds could ameliorate diabetes-associated damages and complications. Oral administration of loganin and morroniside decreased fasting blood glucose levels in diabetes mellitus mice. Ursolic acid exhibited the highest reactive oxygen species scavenging activity and α-glucosidase inhibitory activity [7]. In addition, loganic acid (LA) exhibited antioxidant properties in relation to STZ-induced DM. It may indicate LA as one of the plant components in the development of new drugs that will treat metabolic and functional disorders in leukocytes under diabetes [20]. In recent years, there is growing interest in the utilization of natural products as potential therapeutic agents for treating DM. Therefore, in order to find natural products with antidiabetic activity from this plant, we systematically studied the fruits of Cornus officinalis. In our recent study, four new iridoid glycosides (1-4) and six known ones (5-10) were obtained from Cornus officinalis ( Figure 1). In addition, the in vitro antidiabetic activity of the isolated compounds was evaluated.

Structure Elucidation
Compound 1 was assigned a molecular formula of C22H32O14, as determined from HRESIMS (m/z: 543.1689 [M+Na] + ) and 13 Table 1). The sugar moiety of compound 1 was determined as D-glucose by chiral-HPLC analysis after acid hydrolysis. The above information suggested 1 to be a iridoid glucoside, which was similar to loganin [21] 3.64) were determined to be linked to C-13 and C-16 by HMBC correlations of methylene protons at δ H 2.72, 2.63 and methoxy protons at δ H 3.64 to C-13 (δ C 169.6), and C-16 (δ C 173.2), respectively. Finally, HMBC correlations from H-7 (δ H 5.04) to C-13 (δ C 169.6), suggested that the oxygenated methine H-7 was connected to C-13 ( Figure 2). In the NOESY spectrum, the correlations between H-1 and H-6 (δ H 1.67)/H-8 indicated that H-8 was α-oriented, Me-10 was β-oriented; the correlations between H-6 (δ H 1.67) and H-7 indicated that H-7 was α-oriented. Meanwhile, the correlations from H-5 to H-9 and H-6 (δ H 2.13) confirmed that H-9 and H-5 were β-oriented ( Figure 3). Thus, compound 1 was elucidated as shown in Figure 1, and named neocornuside A. More details are shown in Supplementary Materials. as D-glucose by acid hydrolysis and chiral-HPLC analysis. Thus, the structure of compound 2 was defined as shown in Figure 1, and named neocornuside B. More details are shown in Supplementary Materials. The molecular formula of compound 3, C34H50O20, was determined based on its HRESIMS at m/z 801.2795 [M+Na] + and 13 C NMR data ( Table 2). The NMR data of 3 were consistent with those of cornuside L [21], except for the chemical shifts of C-7 (3, δC 79.5; cornuside L,δC 75.1), C-8 (3, δC 44.3; cornuside L,δC 42.5), and C-10 (3, δC 17.6; cornuside L,δC 13.8). All of the aforementioned information and signals of HMBC, COSY, and HSQC confirm the planar structure of 3 was identical to cornuside L. However, the NOESY cor-   Figure 3.

Structure Elucidation
The key NOESY correlations of compounds 1-4.
The molecular formula of compound 4 was confirmed as C35H52O21 with 10 degrees of unsaturation on the basis of HRESIMS (m/z: 831.2913 [M+Na] + ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.

Cell Viability of Compounds 1-10 in Insulin-Induced HepG2 Cells
Compounds 1-10 had no cytotoxic effect on the cell viability of insulin-induced HepG2 cells in the concentration of 10 μM, which was observed by CCK-8 assay ( Figure  4). Insulin has mitogenic and anti-apoptotic properties, which promote the progression and metastasis of many types of cancer cells [26,27]. Thus, the compounds and insulin promoted the proliferation of HepG2 cells (Figure 4), and clinical management to counteract insulin resistance and subsequent hyperinsulinemia should be taken to prevent the development of hepatocellular carcinoma (HCC) [28].  Table 1). The evident difference was that the methoxy signal H-17 (δ H 3.70) was connected to C-15 (δ C 40.0) in compound 2 but not C-14 (δ C 68.8). The deduction can be further supported by the HMBC correlations between H-17 (δ H 3.70) and C-15 (δ C 40.0) ( Figure 2). The sugar moiety in 2 was also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Thus, the structure of compound 2 was defined as shown in Figure 1, and named neocornuside B. More details are shown in Supplementary Materials.
The molecular formula of compound 3, C 34 H 50 O 20 , was determined based on its HRESIMS at m/z 801.2795 [M + Na] + and 13 C NMR data ( Table 2). The NMR data of 3 were consistent with those of cornuside L [21], except for the chemical shifts of C-7 (3, δ C 79.5; cornuside L,δ C 75.1), C-8 (3, δ C 44.3; cornuside L,δ C 42.5), and C-10 (3, δ C 17.6; cornuside L,δ C 13.8). All of the aforementioned information and signals of HMBC, COSY, and HSQC confirm the planar structure of 3 was identical to cornuside L. However, the NOESY correlations (  (Table 2). Its 1D NMR data ( Table 2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2 (4, δ C 73.5; cornuside I,δ C 83.1) and C-3 (4, δ C 86.8; cornuside I,δ C 76.6). The C-3 of the 7β-O-methylmorroniside unit was linked to C-7 of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7 at δ H 4.80 and C-3 at δ C 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.

Cell Viability of Compounds 1-10 in Insulin-Induced HepG2 Cells
Compounds 1-10 had no cytotoxic effect on the cell viability of insulin-induced HepG2 cells in the concentration of 10 µM, which was observed by CCK-8 assay (Figure 4). Insulin has mitogenic and anti-apoptotic properties, which promote the progression and metastasis of many types of cancer cells [26,27]. Thus, the compounds and insulin promoted the proliferation of HepG2 cells (Figure 4), and clinical management to counteract insulin resistance and subsequent hyperinsulinemia should be taken to prevent the development of hepatocellular carcinoma (HCC) [28].  The molecular formula of compound 4 was confirmed as C35H52O21 with 10 degrees of unsaturation on the basis of HRESIMS (m/z: 831.2913 [M+Na] + ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.
Along with the above new compounds, six known iridoid glycosides were isolated from fruits of Cornus officinalis and identified as 8-epiloganic acid (5)   ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.
Along with the above new compounds, six known iridoid glycosides were isolated from fruits of Cornus officinalis and identified as 8-epiloganic acid (5)   ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.
Along with the above new compounds, six known iridoid glycosides were isolated from fruits of Cornus officinalis and identified as 8-epiloganic acid (5)   ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.
Along with the above new compounds, six known iridoid glycosides were isolated from fruits of Cornus officinalis and identified as 8-epiloganic acid (5)   ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.
Along with the above new compounds, six known iridoid glycosides were isolated  ) and 13 C NMR (Table2). Its 1D NMR data (Table2) were similar to those of cornuside I [21], with the difference of chemical shifts of C-2′ (4, δC 73.5; cornuside I,δC 83.1) and C-3′ (4, δC 86.8; cornuside I,δC 76.6). The C-3′ of the 7β-O-methylmorroniside unit was linked to C-7″ of the α-morroniside unit by an ether linkage based on the HMBC correlations between H-7″ at δH 4.80 and C-3′ at δC 86.8. The sugar moieties in 4 were also identified as D-glucose by acid hydrolysis and chiral-HPLC analysis. Consequently, the structure of compound 4 was determined and named neocornuside D, as shown in Figure 1.

Glucose Consumption of Compounds 1-10 in Insulin-Induced HepG2 Cells
Compounds 1-10 were evaluated for their antidiabetic activity against insulin resistant HepG2 cells. The results in Figure 5 showed that compounds 1, 3, and 7 significantly increased the relative glucose consumption in insulin-induced HepG2 cells (p< 0.05 or p< 0.01). Other compounds could increase the relative glucose consumption, but there

Glucose Consumption of Compounds 1-10 in Insulin-Induced HepG2 Cells
Compounds 1-10 were evaluated for their antidiabetic activity against insulin resistant HepG2 cells. The results in Figure 5 showed that compounds 1, 3, and 7 significantly increased the relative glucose consumption in insulin-induced HepG2 cells (p < 0.05 or p < 0.01). Other compounds could increase the relative glucose consumption, but there was no statistical significance compared with the model group (p > 0.05). Thus, we focused on the antidiabetic activity of compounds 1, 3, and 7. The results showed that compounds 1, 3, and 7 displayed significant antidiabetic activity with EC 50 values of 0.582, 1.275, and 0.742 µM (Table 3), respectively, which indicate that these effective compounds may improve the insulin resistance in HepG2 cells and could provide reference for the development and application of C. officinalis to treat DM. By comparing the structure and activity of compounds 1, 3, and 7, we found that the cyclopentane-type iridoid structural unit may be essential for the antidiabetic activity. Compound 1 showed stronger antidiabetic activity than compounds 3 and 7, indicating that the activity may be related to the relative configuration at position H-7. Moreover, we speculated that compound 3 had an extra α-morroniside unit, resulting in lower antidiabetic activity of compound 3 than compound 7. However, the results showed no significant antidiabetic activity for compounds 2, 5, 8, and 10, therefore the antidiabetic activity may be related to the relative configuration at position H-7, the side chain attached to C-7, and whether the carbonyl group at C-11 forms an ester. In conclusion, further structure-activity relationship remains to be clarified in future research.

Effect of Compounds 1, 3, and 7 on Glucose Uptake in Insulin-Induced HepG2 Cells
To investigate whether iridoid glycosides could promote glucose uptake in HepG2 cells, the uptake of 2-NBDG was evaluated by HepG2 cells treated with different concentrations (5, 10 and 20 μM) of test compounds 1, 3, and 7. The results showed that 2-NBDG uptake in HepG2 cells was significantly decreased after exposed to insulin (Figure 6 and  7). However, compounds 1, 3, and 7 improved the ability of 2-NBDG uptake in insulininduced HepG2 cells (Figures 6 and 7). Therefore, compounds 1, 3, and 7 efficiently alleviated the HepG2 cells injury induced by insulin, which present potential anti-diabetic effects. Figure 5. Effect of compounds 1-10 on the relative glucose consumption in insulin-induced HepG2 cells (x ± s, n = 4). The impact of compounds 1-10 on the relative glucose consumption measured by a glucose assay kit. # p < 0.05, versus control group; * p < 0.05 or ** p < 0.01, versus insulin group.

Effect of Compounds 1, 3, and 7 on Glucose Uptake in Insulin-Induced HepG2 Cells
To investigate whether iridoid glycosides could promote glucose uptake in HepG2 cells, the uptake of 2-NBDG was evaluated by HepG2 cells treated with different concentrations (5, 10 and 20 µM) of test compounds 1, 3, and 7. The results showed that 2-NBDG uptake in HepG2 cells was significantly decreased after exposed to insulin (Figures 6 and 7). However, compounds 1, 3, and 7 improved the ability of 2-NBDG uptake in insulin-induced HepG2 cells (Figures 6 and 7). Therefore, compounds 1, 3, and 7 efficiently alleviated the HepG2 cells injury induced by insulin, which present potential anti-diabetic effects.
To investigate whether iridoid glycosides could promote glucose uptake in HepG2 cells, the uptake of 2-NBDG was evaluated by HepG2 cells treated with different concentrations (5, 10 and 20 μM) of test compounds 1, 3, and 7. The results showed that 2-NBDG uptake in HepG2 cells was significantly decreased after exposed to insulin (Figure 6 and  7). However, compounds 1, 3, and 7 improved the ability of 2-NBDG uptake in insulininduced HepG2 cells (Figures 6 and 7). Therefore, compounds 1, 3, and 7 efficiently alleviated the HepG2 cells injury induced by insulin, which present potential anti-diabetic effects.  cells, the uptake of 2-NBDG was evaluated by HepG2 cells treated with different concentrations (5, 10 and 20 μM) of test compounds 1, 3, and 7. The results showed that 2-NBDG uptake in HepG2 cells was significantly decreased after exposed to insulin (Figure 6 and  7). However, compounds 1, 3, and 7 improved the ability of 2-NBDG uptake in insulininduced HepG2 cells (Figures 6 and 7). Therefore, compounds 1, 3, and 7 efficiently alleviated the HepG2 cells injury induced by insulin, which present potential anti-diabetic effects.

Cell Viability Assay
CCK-8 assay was used to evaluate cell viability of HepG2 cells. After treatments, added 10 µL of CCK-8 reagent and incubated at 37 • C for 1 h. The optical density (OD) value of every well was measured at 450 nm using a microplate spectrophotometer.

Glucose Consumption Assay
The HepG2 cells (1 × 10 5 cells/mL) cultured in 96-well plates were treated with insulin and test compounds as previously described. The culture medium was collected, and the glucose concentrations were measured using the same method [29] with a glucose assay kit. The glucose content of the experimental group medium was subtracted from the glucose content of the original DMEM medium to afford the glucose consumption (GC). The relative glucose consumption (RGC) was calculated by the following formula: RGC = GC/OD.

Glucose Uptake Assay
Glucose uptake rate was measured using 2-NBDG, according to the previously reported method [29]. The HepG2 cells were seeded at 1 × 10 5 cells/mL in 6-well plates at 37 • C for 24 h in a humidified atmosphere of 5% CO 2 . The cells were pre-incubated with various concentrations of test compounds 1, 3, and 7 (5, 10, and 20 µM). After 24 h, 2-NBDG (25 µM) was added in incubation at 37 • C for 1 h. The cells were collected and washed with PBS, and then resuspended in PBS. The cell fluorescence intensity was detected by flow cytometry with excitation wavelength of 488 nm and emission wavelength of 530 nm. The results were analyzed by the software FlowJo 10.8.

Acid Hydrolysis of Compounds 1-4
Each one (1.0 mg) of the compounds 1-4 was dissolved in 2 M HCl-H 2 O (2.5 mL) and heated at 80 • C for 3 h. The reaction mixture was extracted with EtOAc. The aqueous layer was evaporated under vacuum, diluted repeatedly with H 2 O, and evaporated in vacuo to furnish a neutral residue. The residue was dissolved in MeOH (1.5 mL) and analyzed by HPLC equipped with a chiral column (CHIRALPAK AD-H, 5 µm, 4.6 × 250 mm) and an evaporative light scattering detector using n-hexane-EtOH (82:18; v/v) as the mobile phase (0.5 mL/min). For all of the selected compounds, the sugars were found to be D-glucoses by comparing its retention time with that of D-glucose (21.413 and 22.554 min) and L-glucose (22.099 and 23.648 min).

Conclusions
In summary, the chemical composition of Cornus officinalisfruit was further investigated, leading to the isolation of four new iridoid glycosides, neocornuside A-D (1-4), together with six known compounds (5-10). Among the isolated compounds, no cytotoxic effect was seen on the cell viability of insulin-induced HepG2 cells in the concentration of 10 µM, and compounds 1, 3, and 7 displayed significant antidiabetic activity with EC 50 values of 0.582, 1.275, and 0.742 µM, respectively, which was proven to have the potential to ameliorate the glucose uptake of insulin-induced HepG2 cells in doses of 10, 5, and 20 µM, respectively. These effective compounds may represent promising natural antidiabetic compounds for the treatment of DM. It also provided scientific evidence and a foundation for the understanding of the antidiabetic effects and further utilization of Cornus officinalis. In future research, enrichment of the active compounds should be performed for in vivo validation. On the other hand, other compounds will be investigated further for their potential activity using a Surface Plasmon Resonance (SPR) technique for expanding the scope of application.