Flavonoids and its derivatives from Callistephus chinensis flowers and their inhibitory activities against alpha-glucosidase.

Inhibitors of carbohydrate-hydrolysing enzymes play an important role for the treatment of diabetes. One of the therapeutic methods for decreasing of postprandial hyperglycemia is to retard absorption of glucose by the inhibition of carbohydrate- hydrolysing enzymes, such as α-glucosidase, in the digestive organs. To investigate the therapeutic potential of compounds from natural sources, Callistephus chinensis flowers (CCF) were tested for inhibition of α-glucosidase, and acarboes was used as the positive control. The 70 % ethanol extract of CCF exhibited significant α-glucosidase inhibitory activities with IC50 value of 8.14 μg/ml. The stepwise polarity fractions of CCF were tested further for in vitro inhibition of α-glucosidase. The ethyl acetate (EtOAc) fraction exhibited the most significant inhibitory activity. Eight pure compounds, apigenin, apigenin-7-O-β-D- glucoside, kaempferol, hyperin, naringenin, quercetin, luteolin, and kaempferol-7-O-β-D- glucoside, were isolated (using enzyme assay-guide fractionation method) from the EtOAc fraction. Among these, quercetin was the most active one (IC50 values 2.04 μg/ml), and it appears that the inhibiting percentages are close to acarbose (IC50 values 2.24 μg/ml), the positive control, on α-glucosidase inhibition. HPLC/UV analysis indicated that the major components of CCF are kaempferol, hyperin and quercetin. The presented results revealed that CCF containing these eight flavonoids could be a useful natural source in the development of a novel α-glucosidase inhibitory agent against diabetic complications.

Introduction

Over the past several decades, the prevalence of diabetes mellitus has increased dramatically worldwide (Gorelick et al., 2011[9]). According to some estimates, India, China, and United States will have the largest number of diabetic people by the year 2030 (Wild et al., 2004[36]). The reasons for the rising prevalence are often linked with an increase in both obesity and ageing as well as genetic factors (Hui et al., 2010[14]). Without proper and timely treatments, diabetes can result in many complications including hyperglycemia, diabetic ketoacidosis, cardiovascular disease and chronic renal failure (Yu et al., 2013[42]). Accordingly, active and adequate treatments are essential for diabetes.

Adjustment of blood sugar is an effect approach and very important for diabetes control. The ideal final result for diabetes mellitus is to prevent the excessive rise of postprandial blood glucose. α-Glucosidas, wildly existed in the brush border surface of intestinal cells, is essential for carbohydrate digestion due to monosaccharides are only readily absorbed from the intestine. Other carbohydrates must be degraded enzymatically before they can be absorbed (Kim, 2013[20]). The inhibitors of α-glucosidase could delay in the digestion of ingested carbohydrates and consequently lead to suppress the postprandial blood glucose (Heacock et al., 2005[13]).Therefore, α-glucosidase inbibitors have promise as therapeutic agents for type 2 non-insulin-dependent diabetes mellitus (Baron, 1998[4]). However, current glucosidase inhibitions such as acarbose and miglitol for diabetes management are not fully satisfactory and prolonged use often causes some serious side effects, particularly with diarrhea (Lebovitz, 1997[22]) and corresponding intestinal pain and flatulence (Fujisawa et al., 2005[8]). Randomized controlled trials with glucosidase inhibitions report these gastrointestinal side effects as the most common reason for noncompliance and early subject withdrawal (Neuser et al., 2005[31]). Consequently, researchers have increasingly focused on finding more effective and safe α-glucosidase inhibitors from natural materials to develop physiological functional food to treat diabetes (Bhandari et al., 2008[5]; Liu et al., 2011[24]; Matsuura et al., 2002[28]; Nishioka et al., 1998[33]; Wang et al., 2010[35]), such as polyphenols from fruits (Jiang et al., 2013[16]), isoflavones from soybean (Niamnuy et al., 2011[32]), and hydrolysate from sardine muscle (Matsui et al., 1996[27]).

Callistephus chinensis, which belongs to the Asteraceae family, is extensively grown in both the northern and southern parts of China. Since its flowers possess high ornamental value, C. chinensis are now widely cultivated in many countries. Several studies have demonstrated that the Asteraceae family display strong biological activity, such as anti-diabetes, antioxidant effects and inhibitory effects against bacteria and viruses, and at the same time can also prevent indigestion, pneumonia, hepatitis and tumors (Cheng et al., 2005[6]; Gorzalczany et al., 2013[10]; Kim et al., 2013[19]; Matsuda et al., 2002[26]; Zhu et al., 2005[43]; Yan et al., 1999[37]). However, to our knowledge, no study has been done to determine the anti-diabetic activity of Callistephus chinensis flower (CCF). Thus, this present study investigates the chemical constituents of CCF and their anti-diabetic activity and aim to find new natural source for the development of α-glucosidase inhibitory agent for dealing with diabetic complications.

Materials and Methods

Plant materials and chemicals

The flowers of Callistephus chinensis were collected form Inner Mongolia Autonomous Region, Republic of China, in 2012 and were kept in desiccators after they had been air dried. All the solvents for the extraction were purchased from Qingdao Haiyang Chemical Co., Ltd. 1H- and 13C-NMR spectra were recorded on a Bruker AVANCE 600 NMR spectrometer (Rheinstetten, Germany). Reagents and solvents including α-glucosidase from recombinant Saccharomyces cerevisiae (expressed in unspecified host), α-amylase from Bacillus licheniformis, 4-nitrophenyl α-D-glucopyranoside, starch, potassium phosphate buffer, and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Acarbose, used as a standard, was produced by the Bayer Company in Germany.

Preparation of extracts

C. chinensis flowers (CCF, 8 kg) were ground to fine a powder and then extracted with 8 L of 70 % ethanol for 2.5 h under reflux at 50 °C. After extraction, the solvent was removed by vacuum filtration, and the 70 % ethanol extract (265 g) was collected. The aqueous solution was extracted followed by n-hexane, CH2Cl2, EtOAc and n-BuOH to get layer of EtOAc (73.5 g).

Isolation of active compounds from EtOAc-soluble fraction

The EtOAc fraction (70 g) was subjected to purification by silica gel (1200 g) column chromatography using CH2Cl2:MeOH solvent system with increased polarity (from 0:100 to 100:0, v/v). The eluent was collected into ten fractions. Fraction 4 (1.2 g) was applied to a Shephadex LH-20 column and eluted with MeOH to give compound 3 (37 mg) and 5 (14 mg) after recrystallisation with MeOH. Fraction 6 (4.7 g) was again purified by silica gel column to yield compound 1 (22 mg) and 6 (112 mg). Fraction 9 (9.2 g) was further purified by silica gel column and Sephadex LH-20 column to yield compound 2 (17.0 mg), 4 (24.0 mg) and 8 (7.8 mg).

Assay for α-glucosidase inhibitory activity

α-Glucosidase inhibitory activity of samples was performed using a previously described method (Kim et al., 2004[21]; Yu et al., 2013[42]), with slight modification. Thirty microliter of the α-glucosidase solution (2 units mL-1, 0.1 mol L-1 potassium phosphate buffer, pH 6.8) was pre-mixed with 20 μL of the purified compound at different concentrations (in 1 % DMSO) and incubated at 37.5 °C for 5 min, 150 μL of p-nitrophenyl glucopyranoside (pNPG, 10 mM) and 800 μL of 0.1 M potassium phosphate buffer were added to the mixture to start the reaction. After another thirty minutes' incubation at 37.5 °C, 2 mL of 1 M Na2CO3 was added to per test tube for termination (Yu and Yin, 2011[40][41]). The absorbance of the sample at 405 nm was measured to quantify the amount of released product (p-nitro phenol) used by microplate Reader (varioskan flash-3001, Thermo Scientific, USA). Three parallel operations were done. Acarbose was used as positive control. Inhibition of α-glucosidase activity by each of the compound was expressed relative to control (no inhibitor).

Sample preparation and quantitative analysis of flavonoids in CCF

To quantify the amount of compounds 1-8 in CCF extract, 2 mg of each extract was dissolved in 1 mL MeOH. The resultant solutions were filtered through a Whatman 0.45 μm PVDF syringe filter (Cat No. 6779, NJ, USA) prior to HPLC. The HPLC separation of compounds 1-8 in CCF extracts for quantitative analysis was performed using a reverse phase column (Discovery C18, 5 μm, 250 × 4.6 mm, Chuangxintongheng, China) and a mobile phase phase consisted of acetonitrile and water (v/v). The gradient solvent system was 13:87, initially, and was increased in linear gradients to 45: 65 over 65 min. The flow rate was kept constant at 1.0 mL/min, and the eluent was monitored by UV absorbance at 210 nm.

Limit of detection and quantification of flavonoids in CCF

In general, limit of detection (LOD) and limit of quantification (LOQ) were used to validate the HPLC method. Determination of LOD and LOQ values were usually based on the linear regression equation and calculated by signal-to-noise ratio of ≥ 3 and ≥ 10, respectively.

Calibration curves and statistical analysis

The calibration curves and statistical analysis were referred to a previously described method (Mok et al., 2013[29]) with slight modification. Each of eight pure isolated compounds were prepared in MeOH (2 mg/mL) and repeatedly mixed with the same solvent. The concentration of compounds 1-8 was verified by comparing the individual peak areas of each sample to those of the corresponding standards. Calibration curves were obtained based on peak area (Y) as vertical coordinates and concentration (X, mg/mL) as horizontal axis. Data were expressed as mean ± S.D. of three replicate determinations for each sample with different concentrations (n = 5). The inhibitory activities of the samples are described as inhibitory concentration 50 % (IC50) and calculated using SPSS program. Statistical significance was calculated by one-way analysis of variance (ANOVA) method and Dunnett's test.

Results and Discussion

α-Glucosidase activity of the extract and fractions from CCF

The different fractions partitioned from the 70 % ethanol extract were tested for the presence of inhibitory activity against α-glucosidase, and the results are summarized in Table 1(Tab. 1) and Figure 1(Fig. 1). As the EtOAc fraction showed obviously higher inhibitory activities than did the 70 % ethanol extract against α-glucosidase, we supposed that the main active compositions of the 70 % ethanol extract of CCF were basically in the EtOAc fraction. Meanwhile, the data shown that the EtOAc fraction exhibited significant α-glucosidase inhibition, with an IC50 value of 2.97 μg/mL. In previous papers, flavonoid compounds from the leaves of fruits and grape seeds have been of increasing interest due to their anti-diabetic properties (Wang et al., 2010[35]; Han et al., 2012[11]; Kar et al., 2009[17]; Montagut et al., 2010[30]). Although extracts of the Asteraceae family plants have been reported to exhibit significant hepatic protective activity (Jeong et al., 2013[15]), few studies have reported on the anti-diabetic activities of CCF.

Table 1

Table 1: IC50 values of the fractions and compounds 1-8 from CCF for inhibition of α-glucosidase

Figure 1

α-Glucosidase inhibition by extract, fractions, compounds 1-8 and acarbose:

(a) α-glucosidase inhibition by extract, fractions and acarbose. (b) α-glucosidase inhibition by compounds 1-8 and acarbose. (c) α-glucosidase inhibition by compound 6, EtOAc fraction and acarboes

Isolation of active compounds and structural determination

The EtOAc soluble portion of the 70 % ethanol extract from CCF was then subjected to repeated rounds of silica gel and Shephadex LH-20 chromatography, resulting in the isolation of compounds 1-8. The structural determination of isolated compounds was confirmed by a combination of 1H-, 13C-NMR (Table 2(Tab. 2)). The singlet signals at δ 12.02-12.91 observed in the 1H-NMR spectra of compounds 1-8 showed 5-OH of an A-ring in their structures. The signals of typical structures in flavonoids were shown in 13C-NMR spectra and compared with values in the literature (Agrawal, 1989[1]; Agrawal and Bansal, 1989[2]; Andersen and Markham, 1989[3]; Crow et al., 1986[7]; Mok et al., 2013[29]; Markham et al., 1978[25]; Park et al., 2006[34]; Harborne and Williams, 1993[12]; Yasukawa et al., 1989[38]). Based on the spectroscopic data obtained, the chemical structures of the purified compounds were identified as apigenin (1), apigenin-7-Oglucoside (2), kaempferol (3), hyperin (4), naringenin (5), quercetin (6), luteolin (7), kaempferol-7-O-β-D-glucoside (8) (Figure 2(Fig. 2)). Although common flavonoids have been isolated from various plants, to our knowledge, this is the first report on the isolation of phytochemical constituents from CCF.

Table 2

13C-NMR spectral data for compounds 1-8 (DMSO-d) from CCF

Figure 2

Structures of compounds 1-8 from CCF

Inhibition of α-glucosidase activity

Compounds 1-8 were tested for their inhibitory activity against α-glucosidase at four different concentrations. Comparing the activities of isolated compounds with those of the fractions, in particular, our results shown that the quercetin (6) was the main active constituent of CCF, it appears that the inhibiting percentages of isolated compounds are close to acarbose, the positive control, on α-glucosidase inhibition. The IC50 values were shown in Table 1(Tab. 1) and the inhibiting percentages of the fractions and isolated compounds were shown in Figure 1(Fig. 1). In addition, many flavonoids and phenol constituents with strong inhibitory activity toward α-glucosidase have been previously reported (Kawanishi et al., 2003[18]; Lee and Kim, 2008[23]; Yawadio et al., 2007[39]). All the eight compounds isolated from CCF are the typical structures in flavonoids, but among them, only the flavonols (compounds 3, 6 and 8) showed higher α-glucosidase inhibitory activity than the flavonoid. Di-hydroxy flavonoids in a B-ring (compounds 4 and 6) showed more significant inhibitory activity against α-glucosidase than other mono-hydroxy flavonoids. The most interesting is that compound 6 containingdi-hydroxyl groups B-rings and 3-OH in the skeleton exhibited the highest α-glucosidase inhibitory activity (IC50 values 2.04 µg/mL).

Quantitative analysis, LOD, and LOQ of flavonoids in CCF

The concentrations of compounds 1-8 in CCF were determined by HPLC/UV. The HPLC chromatograms of the mixture of eight pure compounds and EtOAc-soluble extract are shown in Figure 3(Fig. 3). The calibration equations and retention times are shown in Table 3(Tab. 3). The retention time of the expected peak for each of the compounds (1-8) purified from CCF was the same as that of the corresponding standard compounds. Kaempferol (3) and quercetin (6) were the major flavonols in CCF (Figure 3(Fig. 3)). The LOD and LOQ were 0.05-3.65 and 0.19-12.20 μg/mL determined at the signal-to-noise ratio S/N of 3 and 10, respectively, in CCF (Table 4(Tab. 4)). The HPLC profiles of the EtOAc fraction indicated that kaempferol (3) and quercetin (6) are the major components, and they play a crucial role in α-glucosidase inhibition.

Figure 3

(a) HPLC chromatograms of the mixture of eight pure compounds. (b) HPLC chromatograms of EtOAc-soluble extract. (c) Quantities of compounds 1-8 in CCF

Table 3

Linearity of standard curves and sensitivity of compounds 1-8 in CCF

Table 4

Quantities of compounds 1-8 in CCF

Conclusions

This is the first study on the utilization of CCF as an anti-diabetic source. Our data suggest that CCF extract could be used as a viable alternative to pharmaceutical inhibitors for the inhibition of α-glucosidase activity, since CCF is well tolerated and relatively inexpensive. It is a good source of anti-diabetic agents, to which flavonols make an important contribution. The data also pointed out that CCF containing these flavonoids could be a useful natural source for the development of α-glucosidase inhibitory agent for dealing with diabetic complications. Further investigations will focus on clarifying the anti-diabetes activity mechanisms of the isolated compounds.

Acknowledgements

The research was supported by E&T modern center for Natural Products of Liaoning Province of China (No. 2008402021) and Construction of R&D institute of state original new drug at Benxi of Liaoning Province (2009ZX09301-012-105B). We are glad to acknowledge the Analytical Center of Shenyang Pharmaceutical University for NMR measurements.