Printer Friendly

Aldose reductase inhibition of a saponin-rich fraction and new furostanol saponin derivatives from Balanites aegyptiaca.

ABSTRACT

Background: Balanites aegyptiaca Del. (Zygophyllaceae) fruits are used to treat hyperglycemia in Egyptian folk medicine and are sold by herbalists in the Egyptian open market for this purpose. Nevertheless, the fruits have not yet been incorporated into pharmaceutical dosage forms. The identity of the bioactive compounds and their possible mechanisms of action were not well understood until now.

Purpose: Aldose reductase inhibitors are considered vital therapeutic and preventive agents to address complications caused by hyperglycemia. The present study was carried out to identify the primary compounds responsible for the aldose reductase inhibitory activity of Balanites aegyptiaca fruits.

Study design: The 70% ethanolic extract of Balanites aegyptiaca fruit mesocarp and its fractions were screened for inhibition of the aldose reductase enzyme. Bio-guided fractionation of the active butanol fraction was performed and the primary compounds present in the saponin-rich fraction (D), which were responsible for the inhibitory activity, were characterized. HPLC chromatographic profiles were established for the different fractions, using the isolated compounds as biomarkers.

Methods: Aldose reductase inhibition was tested in vitro on rat liver homogenate. The butanol fraction of the 70% ethanolic extract was fractionated using vacuum liquid chromatography (VLC, RP-18 column). The most active sub-fraction D, which was eluted with 75% methanol, was subjected to preparative HPLC to isolate the bioactive compounds.

Results: The butanol fraction displayed inhibitory activity against the aldose reductase enzyme ([IC.sub.50] = 55.0 [+ or -] 6 [micro]g/ml). Sub-fraction D exhibited the highest inhibitory activity ([IC.sub.50] = 12.8 [+ or -] 1 [micro]g/ml). Five new steroidal saponin derivatives were isolated from this fraction. The isolated compounds were identified as compound 1a/b, a 7:3 mixture of the 25R:25S epimers of 26-O-[beta]-D-glucopyranosyl-furost-5-ene-3,22,26-triol 3-O-[[alpha]-L-rhamnopyranosyl-(1 [right arrow] 3)- [beta]-D-glucopyranosyl-(1 [right arrow] 2)]- [alpha]-L- rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D-glucopyranoside; compound 3, 26-O-[beta]-D-glucopyranosyl-(25R)-furost-5-ene-3,22,26-triol 3-O-[ [beta]-D-glucopyranosyl-( 1 [right arrow] 2)]- [alpha]-L-rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D-glucopyranoside; compound 3, 26-O-D- glucopyranosyl-(25R)-furost-5,20-diene-3,26-diol 3-O-[[alpha]-L-rhamnopyranosyl-(1 [right arrow] 3) - [beta]-D-glucopyranosyl-(1 [right arrow] 2)]- [alpha]-L-rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D-glucopyranoside; compound 4, 26-O-[beta]-D-glucopyranosyl (25R)-furost-5,20-diene-3,26-dioI 3-0-[ [beta]-D-glucopyranosyl-(1 [right arrow] 2)]- [alpha]-L-rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D-glucopyranoside; and compound 5, which is the 25S epimer of compound 4, by using various spectroscopic methods [MS,1D and 2D NMR(HSQC, HMBC, DQF-COSY, HSQC-TOCSY)]. Compounds 1a/b, 2,3,4,5 exhibited highly significant aldose reductase inhibitory activities ([IC.sub.50] values were 1.9 [+ or -] 0.2, 1.3 [+ or -] 0.5, 5.6 [+ or -] 0.2, 5.1 [+ or -] 0.4, 5.1 [+ or -] 0.6 [micro]M, respectively) as compared to the activity of the reference standard quercetin ([IC.sub.50] = 6.6 [+ or -] 0.3 [micro]M).

Conclusion: The aldose reductase inhibitory activity of Balanites fruits is due to the steroidal saponins present. HPLC chromatographic profiles of the crude butanol fraction and its 4 sub-fractions showed that the most highly bioactive fraction D contained the highest amount of steroidal saponins (75%) as compared to the 21% present in the original butanol fraction. The isolated furostanol saponins proved to be highly active in an in vitro assay.

Keywords:

Balanites aegyptiaca

Zygophyllaceae

Diabetes

Furostane

Aldose reductase

Preparative HPLC

Introduction

Aldose reductase is a key enzyme in the polyol pathway, which catalyzes the conversion of glucose into sorbitol in cases of hyperglycemia (Brownlee 2001). An increase in polyol pathway flux leads to the accumulation of sorbitol in the retina, nerves, and kidney and results in the development of diabetic complications such as retinopathy, neuropathy, and nephropathy (Kador et al. 1980). Thus, aldose reductase inhibitors could be used for the treatment and prevention of such diabetic complications (Kawanishi et al. 2003).

The dried fruits of Balanites aegyptiaca Del. (Zygophyllaceae), popularly known as desert dates, are used as an oral antihyperglycemic in Egyptian folk medicine (Kamel 1991). The bittersweet mesocarp of the fruit is eaten for this purpose. The fruit is an edible drupe about the size of a plum (Tackholm 1974). It consists of a thin, hard epicarp, a dark brown mesocarp, and a hard endocarp that encloses the seed (Beka et al. 2011). Several in vivo studies have supported the antidiabetic activity of the mesocarp of the fruit (Gad et al. 2006; Kamel 1991). A previous in vitro study was carried out by members of our group to better understand the mechanism of action of this antihyperglycemic activity. Different extracts of Balanites fruit (made by extracting with methanol, cold water, hot water, and 70% ethanol) exhibited comparable effects to that of insulin when examining the glucose uptake occurring in peripheral C2C12 skeletal muscle cells. Moreover, dichloromethane- and ethyl acetate- soluble fractions of these extracts (i.e., fractions containing compounds of low to middle polarity) exhibited 1.5 fold the activity of insulin, while a polar, saponin-rich fraction exhibited lower activity in the same tissues (Abdel Motaal et al. 2012). Several furostane and pregnane saponins have been isolated from the mesocarp of the fruit (Farid et al. 2002; Hosny et al. 1992; Kamel 1998; Kamel and Koskinen 1995; Liu and Nakanishi 1982).

To investigate the possible mechanisms of action of bioactive constituents of the fruits, with the long-term goal to identify compounds that can potentially be used in therapy to treat Diabetes mellitus, the aldose reductase inhibitory activities of the extracts and fractions of Balanites fruit were examined. Isolation and structural elucidation of the bioactive lead compounds from the sub-fraction identified as most active was carried out, and HPLC profiles of the active fraction and its sub-fractions were established.

Materials and methods

Plant material

The fruits of Balanites aegyptiaca were brought from Aswan and were authenticated by Dr. M. Gebali (Plant Taxonomy and Egyptian Flora Department, National Research Center, Giza, Egypt). The plant name was checked using The Plant List (http://www.theplantlist.org/). A voucher specimen (voucher no. 22) was deposited at the herbarium of the Faculty of Pharmacy and Drug Technology, Heliopolis University, Cairo, Egypt.

General

Silica gel H (Merck, Darmstadt, Germany) and Lichroprep RP18 silica gel (15-25 [micro]m, Merck, Darmstadt, Germany) were used when performing vacuum liquid chromatography (VLC). Thin layer chromatography (TLC) was performed on silica gel C18 pre-coated plates (Sigma-Aldrich) using 75 and 80% aqueous methanol as development systems. The chromatograms were monitored under UV 254 and 366 nm and visually inspected after spraying with p-anisaldehyde/sulfuric acid spray reagent (after heating at 100[degrees]C, 1 min). Acetonitrile, methanol (HPLC grade), DL-glyceraldehyde, the reduced form of nicotinamide adenine dinucleotide phosphate potassium salt (NADPH), and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich (Vienna, Austria) for the aldose reductase activity assessment and phytochemical profiling. Sodium phosphate dibasic and sodium phosphate monobasic were purchased from Scharlau (Hamburg, Germany) and quercetin dihydrate ([greater than or equal to] 98%) was obtained from Roth (Karlsruhe, Germany). All other chemicals and reagents were of analytical grade. NMR: [sup.1]H and two-dimensional NMR experiments with Varian- Unity- Inova-600 spectrometer; [delta] in ppm, J in Hz. Mass spectra: ESI platform LCZ (Micromass) according to Farid et al. 2002.

Extraction and fractionation

Two kilogram of dried Balanites fruit were macerated in 70% ethanol. Combined extracts were concentrated in vacuo, resulting in 320 g of dried extract. 132 g of this extract were suspended in 300 ml distilled water and successively partitioned against solvents of increasing polarity. The different fractions were dried in vacuo giving 0.41, 1.87, 0.85, 8.5, and 80 g of n-hexane (HxFr), methylene chloride (MeFr), ethyl acetate (EtFr), n-butanol saturated with water (BuFr), and remaining water (RwFr) fractions, respectively (Fig. 6). The BuFr (8.5 g) was chromatographed using normal-phase VLC (7x7 cm, silica gel H, 100 g). Gradient elution was performed using mixtures of chloroform/methanol, yielding 7 fractions, which were concentrated in vacuo and stored until further fractionation. About 1.6 g BuFr-5 (6.6 g, eluted with CH[Cl.sub.3] : MeOH, from 40: 10 to 60 : 90) was further chromatographed using reverse-phase VLC (4 x 8 cm, Lichroprep RP-18, 50 g). Gradient elution was performed with methanokwater (60:40 to 100:0,5% steps, 100 ml fractions). Four sub-fractions (A-D) were formed on the basis of TLC and analytical HPLC analysis. This procedure was repeated two more times with the remaining extract (2.0 g per column). The combined fraction D (400 mg, eluted with 75% methanol/[H.sub.2]O) was chosen for further investigation (Fig. 6).

Isolation of compounds using preparative HPLC

Fraction D was dissolved in 75% aqueous methanol and injected (200 mg per injection) for analysis with preparative HPLC (Sepserv C18 column, 10 pm, 20 x 250 mm, Sepserv, Berlin, Germany) using a Varian HPLC system (Varian Inc., Palo Alto, CA, USA) equipped with Dynamax Model SD-1 and Varian Model SD-1 gradient pumps. Detection was performed using a model UV-1 absorbance detector (monitoring at 205 nm), and using a mixture of acetonitrile (A) and water (B) as the mobile phase. Elution was carried out at a flow rate of 11 ml/min as follows: 0-15 min, 20:80 A:B; 15-35 min, 20:80 A:B to 35:65 A:B; 35-65 min, 35:65 A:B; 65-70 min, 35:65 A:B to 100:0 A:B; 70-80 min, 100:0 A:B; 80-85 min, re-equilibration to starting conditions. This fractionation yielded compound 1a/b (5 mg, [R.sub.t] = 35 min), compound 2 (6 mg, [R.sub.t] = 36 min), compound 3 (5.4 mg, [R.sub.t] = 49 min), compound 4 (6.7 mg, [R.sub.t] = 51 min), and compound 5 (4.2 mg, [R.sub.t] = 52 min) (Figs. 3 and 6).

Aldose reductase inhibition assay

Glucose and lipid homeostasis takes place in the liver. It was found to have moderate levels of the aldose reductase enzyme (Clements et al. 1969; Markus et al. 1983; Scuric et al. 1998; Wirth and Wermuth 1985). For the aldose reductase inhibition assay, a liver weighing 8.7 g was isolated from the rat by cervical dislocation using ether as an anaesthetic. This research was conducted in accordance with the internationally accepted principles for laboratory animal use and care as found in the European Community guidelines (EEC Directive of 1986; 86/609/EEC). The liver was chopped and homogenized in 10 volumes of sodium phosphate buffer (pH 6.2) using a homogenizer (Heidolph-RZR 2014, Germany). 100 ml of sodium phosphate buffer contained Na[H.sub.2]P[O.sub.4] x [H.sub.2]O, 1.1247 g and [Na.sub.2]HP[O.sub.4], 0.2626 g. After centrifugation at 9000 rpm for 30 minutes in a cooling centrifuge (Hettich, Universal-320R, USA), the supernatant, which contained the aldose reductase enzyme was collected (liver homogenate). All procedures were carried out at 4[degrees]C. Balanites test samples (extracts, fractions, or isolated compounds) and quercetin (reference standard, [greater than or equal to] 98%) were dissolved in 50% aqueous DMSO and the stock solutions were diluted in phosphate buffer (pH 6.2) to yield various concentrations. The final concentrations of Balanites samples in the final reaction mixture ranged from 10 to 500 [micro]g/ml. Different volumes of liver homogenate ranging from 50 to 300 ul in the 1.0 ml reaction mixture were tested. DL-glyceraldeyde (0.1 to 1 mM) and the reduced form of NADPH (0.01 to 0.05 mM) were freshly prepared using sodium phosphate buffer (pH 6.2) in the 1.0 ml reaction mixture. Liver homogenate and DL-glyceraldeyde were mixed together without any inhibitor and the reaction was initiated by the addition of NADPH. Sodium phosphate buffer (pH 6.2) was used to fill the reaction volume up to 1.0 ml. Aldose reductase activity was assayed spectrophotometrically (Secomam Anthelie spectrophotometer, Dathelie program version 4.11, Ales Cedex, France) by measuring the decrease in the absorption of NADPH at 340 nm. The difference in the optical density was recorded after 15 min. The maximum decrease in absorbance occurred when 1.0 ml reaction mixture containing 150 pi liver homogenate, 1 mM DL-glyceraldeyde and 0.05 mM NADPH was examined over a period of 60 min. The percent inhibitions of the Balanites test samples were calculated, assuming that the aldose reductase activity in the absence of the inhibitor was 100%. All determinations were performed in triplicates. The concentration of each test sample that caused 50% inhibition of the enzyme activity ([IC.sub.50]) was determined from the curve established by plotting the concentration against the percent inhibition.

HPLC analysis of active fractions

An analytical HPLC method was developed for the analysis of the active BuFr and its sub-fractions A-D. An Agilent Technologies 1100 series HPLC was used, equipped with an Agilent 1200 series G1322A quaternary pump and degasser, and a G1314A variable wavelength detector. Isolated compounds and active fractions were injected on to a Lichrospher 100 RP-18, 5 [micro]m, 250 x 4 mm column (Merck, Germany) equipped with a 5 [micro]m, 10 x 4 mm guard column and maintained at a temperature of 25[degrees]C. The mobile phases used were acetonitrile (solvent A) and water (solvent B). Gradient elution was carried out at a flow rate of 0.5 ml/min (0-5 min, 10:90 A:B; 5-10 min, 10:90 to 30:70 A:B; 10-15 min, 30:70 A:B; 15-20 min, 30:70 to 90:10 A:B; 20-35 min, 90:10 to 100:0 A:B). The injection volume was 20 [micro]l and UV detection was performed at 205 nm.

Results and discussion

Aldose reductase inhibition

The 70% ethanolic extract of B. aegyptiaca and its successive fractions; HxFr, MeFr, EtFr, BuFr and RwFr were screened at different concentrations for their aldose reductase inhibitory activities on rat liver homogenate. The 70% ethanol, MeFr, and EtFr extracts showed moderate inhibitory activities ([IC.sub.50] = 401.0 [+ or -] 1,374.0 [+ or -] 1 and 372.0 [+ or -] 2 [micro]g/ml, respectively). The BuFr and RwFr showed higher activities ([IC.sub.50] = 55.0 [+ or -] 2,155.0 [+ or -] 6 [micro]g/ml, respectively). Our previous work revealed that the less polar fractions of Balanites fruit, namely the dichloromethane- and ethyl acetate- soluble extract portions (standardized to contain 0.031 and 0.239% of rutin, 0.007 and 0.004% of isorhamnetin, respectively), displayed significantly high insulin-like activities in peripheral tissues (Abdel Motaal et al. 2012). Here, the more polar fractions of Balanites fruit (BuFr and RwFr), which contain mainly saponins, displayed higher aldose reductase inhibitory activity. The BuFr which was chosen for further fractionation using VLC (Fig. 6) was about 7 times as active as the crude ethanolic extract and other active fractions (MeFr and EtFr), and contained fewer sugars and water-soluble impurities than the remaining water fraction (RwFr). Fractionation of BuFr yielded four subfractions (A-D), whereby D showed the highest inhibitory activity ([IC.sub.50] = 12.8 [+ or -] 1 [micro]g/ml) as compared to B ([IC.sub.50] = 216 [+ or -] 7 [micro]g/ml) and to [IC.sub.50] = 2.0 [+ or -] 8 [micro]g /ml, the activity of the reference standard quercetin. Sub-fractions A and C showed no activity under our experimental conditions (Fig. 1). The activity of sub-fraction D was substantially higher than that of BuFr, indicating that the inhibitory bioactive compounds were enriched in this fraction.

Fraction D was subjected to chromatography using a preparative HPLC column, whereby five steroidal saponins, compounds 1a/b, 2-5, were isolated (Figs. 3 and 6). Compounds la/b, 2,3,4, 5 exhibited highly significant aldose reductase inhibitory activities ([IC.sub.50] values were 1.9 [+ or -] 0.2,1.3 [+ or -] 0.5, 5.6 [+ or -] 0.2, 5.1 [+ or -] 0.4, 5.1 [+ or -] 0.6 [micro]M, respectively) as compared to the activity of the reference standard quercetin ([IC.sub.50] = 6.6 [+ or -] 0.3 [micro]M) (Fig. 2).

Stuctural elucidation of isolated bioactive compounds

The NMR spectra of the five steroidal saponins (compounds 1a/b, 2, 3, 4, 5) showed four quaternary carbon resonances corresponding to C-5, C-10, C-13 and C-22 of a furostan (Table 1) (Staerk et al. 2006). The appearance of quaternary carbon resonances at 110.7 and 110.8 ppm were characteristic of the 22-OH in compounds 1a/b and 2. While in compounds 3-5, olefinic signals appeared at 103.7-103.6 (C-20) and 152.5-152.5 (C-22), respectively, and the 21-Me appeared at 11.6 ppm, suggesting [[DELTA].sup.20(22)] furostanoids (Agrawal 2005). The carbohydrate region of the HSQC spectra were practically identical for compounds 1a/b and 3, the NMR data indicated the presence of two terminal 6-desoxy hexoses, one terminal hexose and two substituted hexoses. Compounds 2,4 and 5 again showed very similar HSQC correlations for the carbohydrate moieties. In comparison to compounds 1a/b and 3, one terminal 6-desoxy hexose moiety was missing in 2, 4 and 5. Homonuclear (COSY) and heteronuclear (HSQC, HMBC) correlations enabled the assignment of the entire set of [sup.1]H and [sup.13]C resonances and identifying the five new furostanol saponin derivatives (1a/b-5). Acid hydrolysis and GC were used to identify the sugar constituents of the isolated compounds. The compounds were separately acid hydrolyzed and the resulting sugar moieties were extracted, derivatized and analyzed using GC along side with authentic sugars. The absolute configuration of the sugars were determined according to the method described by Hara et al. (1987). Based on the retention time of the authentic sugars, D-glucose and L-rhamnose were identified as the sugar moieties present (Hara et al. 1987).

Compound 1a/b, was obtained as a white amorphous powder. Its molecular formula was inferred as [C.sub.57][H.sub.94][O.sub.27] from the ESI-MS spectrum: a peak at m/z 1210.6 corresponding to [[M-H].sup.-], suggesting that the molecular weight of compound 1a/b is 1211.34. The [sup.1]H and [sup.13]C NMR data (Tables 1 and 2) were consistent with a saponin previously isolated from B. aegyptiaca fruit (Staerk et al., 2006), except for switching the sites of attachment of the glucopyranosyl (Glcp") and rhamnopyranosyl (Rhap'") residues and the presence of an extra rhamnopyranosyl (Rhap""). This was confirmed through examination of HMBC correlations. The anomeric proton H-1' ([delta] 4.90) of the first glucose was correlated with C-3 ([delta] 78.1) of the aglycone, and the aglycone proton H-3 ([delta] 3.86) showed a correlation to the anomeric carbon C-1' ([delta] 99.7). The resonance of C-2' ([delta] 81.2) showed a correlation to H-1" ([delta] 5.03) and the proton H-2' ([delta] 4.16) was correlated with C-1" ([delta] 104.6) establishing the attachment point of the second glucose (Glcp"). Because the anomeric protons of both the terminal rhamnoses showed the same [sup.1]H NMR shift value ([delta] 6.19), the observed correlations to C-4' ([delta] 77.3) and C-3'' ([delta] 83.2) proved that the glucoses were substituted by rhamnoses. The assignment of the individual rhamnoses to these positions was done by HMBC correlations between H-3'' ([delta] 4.31) and C-1'"' (102.7) and H-4' ([delta] 4.17) and C-1'" ([delta] 101.5), respectively. The fifth sugar moiety showed an HMBC correlation of its anomeric proton ([delta] 4.81) to C-26 ([delta] 75.4) of the aglycone.

The two sets of [sup.1]H NMR chemical shift values for the methylene group C-26 (3.95, 3.62 and 4.06, 4.53) suggested that the saponin was actually a mixture of the 25R and 25S epimeric hemiacetals. Both approaches, comparison of the [sup.1]H NMR chemical shift values with data from the same kind of saponins recorded under the same conditions (Farid et al. 2002) and using the chemical shift difference of the diastereomeric protons at C-26 (Agrawal 2005) revealed that compound 1 was a 7:3 mixture of the 25R and 25S epimers. Compound 1a/b was, therefore, identified as a mixture (7:3, 25R.25S epimers) of 26-O-[beta]-D-glucopyranosyl-furost-5-ene-3,22,26-triol 3-O-[[alpha]-L-rhamnopyranosyl-(1[right arrow]-3)- [beta]-D-glucopyranosyl-(1[right arrow]2)]-[alpha]-Lrhamnopyranosyl-(1[right arrow]4)-[beta]-D-glucopyranoside (Fig. 4).

Compound 2 was obtained as a white amorphous powder. Its molecular formula was inferred as [C.sub.51][H.sub.84][O.sub.23] from the ESI-MS spectrum: a peak at m/z 1064.54 corresponding to [[M-H].sup.-], suggesting that the molecular weight of compound 2 is 1065.2. Its [sup.1]H and [sup.13]C NMR data were similar to that of compound 1a/b, except for the absence of the fourth rhamnose moiety which was confirmed by the presence of 4 anomeric protons only (Table 2). Again, a complete set of HMBC correlations was found for compound 2: H-1' ([delta] 4.92) showed a correlation to C-3 ([delta] 78.2), H-V ([delta] 5.09) to C-2' ([delta] 82.0), H-1'" ([delta] 6.20) to C-4' ([delta] 77.6). These correlations defined the structure of the glycosidic side-chain attached to C-3 of the aglycone. The fourth sugar moiety, a [beta]-glucose, showed an HMBC correlation of its anomeric proton ([delta] 4.78) to C-26 ([delta] 75.2) of the aglycone. The differences in the [sup.1]H NMR chemical shift of the [H.sub.2]-26 geminal protons ([[DELTA].sub.ab] = 3.92 - 3.63 = 0.29 ppm) and comparison with the chemical shift values in literature (Agrawal 2005; Farid et al. 2002) suggested the presence of a pure 25R epimer. Compound 2 was identified as 26-O-[beta]-D-glucopyranosyl(25R)-furost-5-ene-3,22,26-triol 3-O-[ [beta]-D-glucopyranosyl-(1[right arrow]2)]-[alpha]-L-rhamnopyranosyl-(1[right arrow]4)-[beta]-D-glucopyranoside (Fig. 4).

Compound 3 was obtained as a white amorphous powder. Its molecular formula was inferred as [C.sub.57][H.sub.92][O.sub.26] from the ESI-MS spectrum: a peak at m/z 1192.59 corresponding to [[M-H].sup.-], suggesting that the molecular weight of compound 3 is 1193.32. The [sup.1]H and [sup.13]C NMR chemical shifts ([delta] ppm) of its carbohydrate moieties were consistent with those of compound 1a/b (Table 2), suggesting the same attachments of the four sugar moieties which was confirmed through HMBC correlations. In addition, a 2D HSQC-TOCSY experiment was recorded for compound 3 in order to confirm the carbon resonance assignments of the glucose moieties. The differences in the [sup.1]H NMR chemical shift of the [H.sub.2]-26 geminal protons ([[DELTA].sub.ab] = 3.93 - 3.63 = 0.30 ppm) suggested the presence of a pure 25R epimer (Agrawal 2005). Compound 3 was thus identified as 26-O-[beta]-D-glucopyranosyl-(25R)-furost-5,20-diene-3,26-diol 3-O-[[alpha]-L-rhamnopyranosyl-(1 [right arrow] 3)- [beta]-D-glucopyranosyl-(1 [right arrow] 2)] - [alpha]-L-rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D-glucopyranoside (Fig. 4).

Compounds 4 was obtained as white amorphous powder. Its molecular formula was inferred as [C.sub.51][H.sub.82][O.sub.22] from the ESI-MS spectra; a peak at m/z 1046.53 corresponding to [[M-H].sup.-], suggesting that the molecular weight of compound 4 is 1047.18.

Compounds 5 was obtained as white amorphous powder. Its molecular formula was inferred as [C.sub.51][H.sub.82][O.sub.22] from the ESI-MS spectra: a peak at m/z 1046.53 corresponding to [[M-H].sup.-], suggesting that the molecular weight of compound 5 is 1047.18.

The [sup.1]H and [sup.13]C NMR data of compounds 4 and 5 were identical except for the [sup.1]H NMR chemical shifts of the [H.sub.2]-26 geminal protons. They were 3.96, 3.63 and 4.06, 3.51 in compounds 4 and 5, respectively, denoting that 5 is the 25S epimer of 4 (25R) (Agrawal 2005) (Table 1). The NMR data of their sugar moieties were very similar to those of compound 2 (Table 2) and the HMBC spectra showed the same correlations as found for compound 2. Thus compound 4 was identified as 26-O-[beta]-D-glucopyranosyl-(25R)-furost-5,20-diene-3,26-diol 3-O-[ [beta]-D-glucopyranosyl-(1 [right arrow] 2)] -[alpha]-L-rhamnopyranosyl-(1 [right arrow] 4)-[beta]-D- glucopyranoside, and compound 5 as its 25S epimer (Fig. 4).

HPLC chromatographic profiles

HPLC fingerprints/chromatographic profiles were established for characterization of the active fraction BuFr and its sub-fractions A, B, C and D. Isolated steroidal saponins were dissolved in methanol (HPLC grade, Sigma-Aldrich) and injected separately. The BuFr and subfraction A were analyzed at a concentration of 5 mg/mL methanol and sub-fractions B-D, at 1 mg/mL. The isolated compounds were identified in each chromatogram whenever possible, whereas compounds 1a/b and 2 gave one sharp peak at 22.4 min, compound 3, a peak at 26.7 min, and the two epimers 4 and 5 appeared as one sharp peak at 28.5 min (Fig. 5). Comparing the chromatographic profiles in Fig. 5, sub-fraction D was identified as being rich in the steroidal saponins (75%, at [R.sub.t] 22.4-28.5 min) as compared to BuFr (21% saponins) and sub-fraction B (10% saponins). Also sub-fractions A and C, which gave no activity, were either devoid of steroidal saponins or contained traces of them, respectively. A positive correlation could be drawn between the aldose reductase inhibitory activity and the presence of steroidal saponins in the mixture. These results were not entirely unexpected. A previous in vivo study revealed that a mixture of saponins from Balanites aegyptiaca displayed antidiabetic activities when tested in STZ-induced diabetic mice (Kamel 1991). These findings are also not restricted to Balanites; triterpenoid saponins are listed as bioactive consitutents of extracts of several medicinal plants used in Traditional Chinese Medicine, which display plasma glucose lowering activity when tested in STZ-induced diabetic mice (Fie et al. 2011). It is also of interest to note that furostanol saponins from fenugreek (Trigonella foenum-graeceum), a medicinal plant that is used to treat various diseases, including hypercholesterolemia and obesity, modulate disaccharidase and glycogen enzymatic activities when tested in vivo in rats and have been shown to increase the hepatic glycogen content, lower blood glucose levels and lead to improved results in the oral glucose tolerance test (OGTT) (Hamden et al. 2010).

Conclusion

The saponin rich fraction D of the butanol extract of Balanites aegyptiaca fruits was found to possess significant aldose reductase inhibitory activity. These results raise the possibility that B. aegyptiaca or some of its components may act, at least in part, as aldose reductase inhibitors. These findings are of significant importance, because furostanol saponins have rarely been previously reported to display aldose reductase inhibitory activity. Five new furostanol saponin derivatives isolated from fraction D showed substantially higher activities than the crude butanolic extract and were highly active compared to the reference compound, quercetin. These isolated compounds may represent promising lead structures for novel oral aldose reductase drug development, depending on their bioavailability and safety profiles, which will be tested in appropriate future in vivo and clinical studies. The findings for Trigonella foenum-graeceum, however, indicate that the furostanol saponins isolated from Balanites aegyptiaca may have more than one potential mechanism of action contributing to the anti-diabetic activity, and that further studies on the disaccharidase and glycogen enzymatic activity modulation would be useful.

Conflict of interest

No conflicts are known.

ARTICLE INFO

Article history:

Received 5 January 2015

Revised 30 April 2015

Accepted 28 May 2015

Abbreviations: MeFr, methylene chloride fraction; EtFr, ethyl acetate fraction; BuFr, n-butanol saturated with water fraction; RwFr, remaining water fraction.

Acknowledgments

This work was carried out in frame of an FP7 project, Marie Curie Action, PIRSES-GA-2008-230816. The project title was "Natural antidiabetic and anti-hypertensive drugs (NAAN)". Rudolf Bauer acted as project coordinator, and Amira Abdel Motaal acted as the Egyptian team leader.

References

Abdel Motaal, A., Shaker, S., Haddad, P.S., 2012. Antidiabetic activity of standardized extracts of Balanites aegyptiaca fruits using cell-based bioassays. Pharmacognosy J. 4,20-24.

Agrawal, P.K., 2005. Assigning stereodiversity of the 27-Me group of furostane-type steroidal saponins via NMR chemical shifts. Steroids 70,715-724.

Beka, R.G., Guiama, V.D., Delmont, A., Donn, P., Slomianny, M.-C, Libouga, D.G., Mbofung, C.M., Guillochon, D., Vercaigne-Marko, D., 2011. Glycosyl part identified within Balanites aegyptiaca fruit protease. Int.J. Biol. Macromol. 49,397-401. Brownlee, M., 2001. Biochemistry and molecular cell biology of diabetic complications. Nature 414,813-2820.

Clements, R.S., Jr Weaver, J.P., Winegrad, A.I., 1969. The distribution of polyol: NADP oxidoreductase in mammalian tissues. Biochem. Biophys. Res. Commun. 37, 347-353.

Farid, H., Haslinger, E., Kunert, 0., Wegner, C., Hamburger, M., 2002. New steroidal glycosides from Balanites aegyptiaca. Helv. Chim. Acta 8,1019-1026.

Gad, M.Z., ElSawalhi, M.M., Ismail, M.F., ElTanbouly, N.D., 2006. Biochemical study of the antidiabetic action of the Egyptian plants Fenugreek and Balanites. Mol. Cell. Biochem. 28,1173-1183.

Hamden, K., Jaouadi, B., Salami, T., Carreau, S., Bejar, S., Elfeki, A., 2010. Modulatory effect of fenugreek saponins on the activities of intestinal and hepatic disaccharidase and glycogen and liver function of diabetic rats. Biotechnol. Bioprocess Eng. 15, 745-753.

Hara, S., Okabe, H., Mihashi, K., 1987. Gas-liquid chromatographic separation of aldose enantiomers as trimethylsilyl ethers of methyl 2-(polyhydroxyalkyl)-thiazolidine-4(R)-carboxylates). Chem. Pharm. Bull. 35,501-506.

He, K., Li, X., Chen, X., Ye, X., Huang, J., Jin, Y., Li, P., Deng, Y., Jin, Q., Shi, Q., Shu, H., 2011. Evaluation of antidiabetic potential of selected traditional Chinese medicines in STZ-induced diabetic mice. J. Ethnopharmacol. 137,1135-1142.

Hosny, M., Khalifa, T., Calis, L, Wright, A.D., Sticher, 0., 1992. Balanitoside, a furostanol glycoside, and 6-methyldiosgenin from Balanites aegyptiaca. Phytochemistry 31, 3565-3569.

Kador, P.F., Kinoshita. J.H., Tung, W.H., Chylackjr., L.T., 1980. Differences in the susceptibility of various aldose reductases to inhibition. II. Inves. Ophthalmol. Vis. Sci. 19, 980-982.

Kamel, M.S., 1991. Studies on Balanites aegyptiaca fruits, an antidiabetic Egyptian folk medicine. Chem. Pharm. Bull. 39,1229-1233.

Kamel, M.S., 1998. A furostanol saponin from fruits of Balanites aegyptiaca. Phytochemistry 48, 755-757.

Kamel, M.S., Koskinen, A., 1995. Pregnane glycosides from fruits of Balanites aegyptiaca. Phytochemistry 40, 1773.

Kawanishi, K., Ueda, H., Moriyasu, M., 2003. Aldose reductase inhibitors from the nature. Curr. Med. Chem. 10, 1353-1374.

Liu, H.W., Nakanishi, K., 1982. The structures of balanitins, potent molluscicides isolated from Balanites aegyptica. Tetrahedron 8,513-519.

Markus, H.B., Raducha, M., Harris, H., 1983. Tissue distribution of mammalian aldose reductase and related enzymes. Biochem. Med. 29,31-45.

Scuric.Z., Stain, S.C., Anderson, W.F., Hwang, J.J., 1998. New member of aldose reductase family proteins overexpressed in human hepatocellular carcinoma. Hepatology 27, 943-950.

Stark, D., Chapagain, B.P., lindin, T., Wiesman, Z., Jaroszewski, J.W., 2006. Structural analysis of complex saponins of Balanites aegyptiaca by 800 MHz [sup.1]H NMR spectroscopy. Magn. Reson. Chem. 44,923-928.

Tackholm, V., 1974. Students' Flora of Egypt, second ed. Cooperative Printing Company, Beirut.

Wirth, H.P., Wermuth, B., 1985. Immunohistochemical localisation of aldehyde and aldose reductase in human tissues. Prog. Clin. Boil. Res. 174,231-239.

Amira Abdel Motaal (a,b) *, Hesham El-Askary (a), Sara Crockett (c), Olaf Kunert (d), Basma Sakr (b), Sherif Shaker (b), Alice Grigore (e), Radu Albulescu (e), Rudolf Bauer (c)

(a) Faculty of Pharmacy, Cairo University, Kasr-El-Ainy St, Cairo 11562, Egypt

(b) Faculty of Pharmacy and Drug Technology, Heliopolis University, 2834 El Horreya, Cairo, Egypt

(c) Institute of Pharmaceutical Sciences. University of Graz, A-8010 Graz, Austria

(d) Institute of Pharmaceutical Chemistry and Pharmaceutical Technology, University of Graz, A-8010 Graz, Austria

(e) National Institute for Chemical--Pharmaceutical Research and Development (ICCF), Bucharest, Romania

* Corresponding author at: Pharmacognosy Department, Faculty of Pharmacy, Cairo University, Kasr-El-Ainy St., Cairo 11562, Egypt. Tel.: +201221770823; fax: +20226579360.;

E-mail addresses: a_motaal@hotmail.com, amira.abdelmotaal@hu.edu.eg (A. Abdel Motaal), heshamaskary@yahoo.com (H. El-Askary), sara.crockett@uni-graz.at (S. Crockett), olaf.kunert@uni-graz.at (O. Kunert), basma.sakr@hu.edu.eg (B. Sakr), sherif.shaker@hu.edu.eg (S. Shaker), alicella77@yahoo.com (A. Grigore), radu_al@hotmail.com (R. Albulescu), rudolf.bauer@uni-graz.at (R. Bauer).

http://dx.doi.org/10.1016/j.phymed.2015.05.059

Table 1
[sup.1]H and [sup.13]C NMR chemical shifts ([delta] ppm) of
the aglycone of compounds 1-4 in ([D.sub.5]) pyridine at
25[degrees]C, at 600 ([sup.1]H) and 150 MHz ([sup.13]C);
Si[Me.sub.4] used as internal standard, J in Hz.

                  1a/b                        2

                  [[delta]   [[delta]         [[delta]   [[delta]
                  .sub.C]    .sub.H]          .sub.C]    .sub.H]

[CH.sub.2] (1)    37.4       1.77, 0.99       37.4       1.74, 0.99
[CH.sub.2] (2)    29.9       2.11, 1.86       30.0       2.09, 1.85
H-C (3)           78.1       3.86             78.2       3.85
[CH.sub.2] (4)    38.8       2.75, 2.70       38.9       2.74, 2.69
C (5)             140.8      --               140.9      --
H-C (6)           121.6      5.31 (brs)       121.7      5.30 (brs)
[CH.sub.2] (7)    32.2       1.91, 1.49       32.2       1.92, 1.46
H-C (8)           31.6       1.6              31.7       1.57
H-C (9)           50.2       0.92             50.3       0.90
C (10)            37.0       --               37.1       --
[CH.sub.2] (11)   21.0       1.47, 1.47       21.0       1.45, 1.45
[CH.sub.2] (12)   39.8       1.77, 1.15       39.9       1.77, 1.13
C (13)            40.4       --               40.6       --
H-C (14)          56.4       1.09             56.6       1.08
[CH.sub.2] (15)   32.3       1.98, 1.45       32.5       2.02, 1.46
H-C (16)          80.9       4.93             80.3       4.92
H-C (17)          63.7       1.94             63.8       1.93
Me (18)           16.2       0.90 (s)         16.3       0.90 (s)
Me (19)           19.1       1.06 (s)         19.3       1.05 (s)
H-C (20)          40.5       2.25             40.6       2.24
Me (21)           16.2       1.31             16.3       1.32
                             (d, 6.7)                    (d, 6.7)
C (22)            110.7      --               110.8      --
[CH.sub.2] (23)   37.0       2.07, 2.00       37.1       2.03, 2.00
[CH.sub.2] (24)   28.2       2.04, 1.70       28.3       2.04, 1.68
H-C (25)          34.2       1.93             34.2       1.92
[CH.sub.2] (26)   75.4       4.06, 4.53 and   75.2       3.92, 3.63
                             3.95, 3.62
Me (27)           17.2       1.03             17.2       1.00
                             (d, 6.6)                    (d, 6.7)

                  3                       4

                  [[delta]   [[delta]     [[delta]   [[delta]
                  .sub.C]    .sub.H]      .sub.C]    .sub.H]

[CH.sub.2] (1)    37.4       1.75, 0.99   37.3       1.73, 0.96
[CH.sub.2] (2)    30.1       2.09, 1.85   30.0       2.11, 1.87
H-C (3)           78.1       3.83         77.9       3.86
[CH.sub.2] (4)    38.9       2.74, 2.68   38.8       2.75, 2.72
C (5)             140.9      --           140.7      --
H-C (6)           121.7      5.32 (brs)   121.7      5.29 (brs)
[CH.sub.2] (7)    32.4       1.88, 1.51   32.3       1.97, 1.49
H-C (8)           32.3       1.53         nda        nd
H-C (9)           50.4       0.91         50.3       0.89
C (10)            37.1       --           37.1       --
[CH.sub.2] (11)   21.2       1.47, 1.42   21.2       1.46, 1.40
[CH.sub.2] (12)   39.6       1.77, 1.18   39.5       1.75, 1.14
C (13)            433        --           43.3       --
H-C (14)          54.8       0.90         54.8       0.88
[CH.sub.2] (15)   34.3       2.13, 1.52   nd         nd
H-C (16)          84.4       4.81         84.2       4.80
H-C (17)          64.5       2.47         64.3       2.45
Me (18)           14.0       0.74(s)      13.8       0.75 (s)
Me (19)           19.1       1.05 (s)     19.2       1.07 (s)
H-C (20)          103.7      --           103.6      --
Me (21)           11.6       1.65         11.6       1.65
                             (d, 6.6)                (d, 6.6)
C (22)            152.5      --           152.5      --
[CH.sub.2] (23)   23.7       2.25, 2.23   23.6       2.23, 2.20
[CH.sub.2] (24)   31.4       1.84, 1.48   31.5       1.83, 1.48
H-C (25)          33.4       1.95         33.4       1.95
[CH.sub.2] (26)   74.9       3.93, 3.63   75.0       3.96, 3.63

Me (27)           17.2       1.03         17.1       1.03
                             (d. 6.7)                (d. 6.7)

Compound 5 showed the same NMR data of the aglycone as
compound 4, except for [CH.sub.2](26) was at 4.06, 3.51.

(a) nd. not determined.

Table 2
[sub.1]H and [sub.13]C NMR chemical shifts ([delta] ppm) of the
carbohydrate moieties of compounds 1-4 in ([D.sub.5]) pyridine
at 25[degrees]C, at 600 ([sup.1]H) and 150 MHz ([sup.13]C);
Si[Me.sub.4] used as internal standard; J in Hz.

                         1a/b                  2

                                    [[delta]              [[delta]
                         [[delta]   .sub.H]    [[delta]   .sub.H]
                         .sub.C]    (J, Hz)    .sub.C]    (J, Hz)

Glcp' (a)    H-C(1')     99.7       4.90       100.1      4.92
                                    (d, 7.2)              (d, 7.6)
             H-C(2')     81.2       4.16       82.0       4.18
             H-C(3')     77 3       4.17       77.3       4.18
             H-C(4')     77 J       4.17       77.6       4.23
             H-C(5')     76.0       3.77       76.1       3.82
             CH2(6')     61.7       4.43,      62.1       4.46, ndc
                                    4.40

Glcp"        H-C(1")     104.6      5.03       104.9      5.09
                                    (d, 7.5)              (d, 7.7)
             H-C(2")     75.0       4.01       74.9       4.03
             H-C(3")     83.2       4.31       77.3       4.18
             H-C(4")     69.2       4.21       71.2       4.23
             H-C(5")     78.1       3.88       78.3       3.95
             CH2(6")     61.7       4.37,      62.1       4.43, nd
                                    4.30

Rhap"' (b)   H-C(1"')    101.5      6.19       101.6      6.20
                                    (brs)                 (brs)
             H-C(2"')    72.3       4.64       72.3       4.69
             H-C(3"')    72.5       4.52       72.6       4.53
             H-C(4"')    73.9       4.28       74.2       4.28
             H-C(5"')    69.3       4.88       69.6       4.89
             Me(6"')     18.4       1.74       18.5       1.74
                                    (d, 6.6)              (d, 6.6)
Rhap""       H-C(1"")    102.7      6.19
                                    (brs)
             H-C(2"")    72.2       4.69
             H-C(3"")    72.5       4.52
             H-C(4"")    73.9       4.28
             H-C(5"")    69.7       4.93
             Me(6"")     18.3       1.68
                                    (d, 6.6)

Glcp""'      H-C(1""')   104.8      4.81       104.8      4.78
                                    (d. 7.2)              (d, 7.9)
             H-C(2""')   75.0       3.99       75.1       3.98
             H-C(3""')   78.4       4.18       78.5       4.18
             H-C(4""')   71.7       4.20       71.7       4.18
             H-C(5""')   78.1       3.92       78.3       3.90
             CH2(6""')   62.8       4.53,      62.8       4.51,
                                    4.36                  4.34

                         3                     4

                                    [[delta]              [[delta]
                         [[delta]   .sub.H]    [[delta]   .sub.H]
                         .sub.C]    (J, Hz)    .sub.C]    (J, Hz)

Glcp' (a)    H-C(1')     99.9       4.89       99.7       4.96
                                    (d, 7.9)              (d, 7.2)
             H-C(2')     81.4       4.15       81.7       4.23
             H-C(3')     77.2       4.15       77.3       4.23
             H-C(4')     77.5       4.16       77.6       4.23
             H-C(5')     76.1       3.77       76.1       3.87
             CH2(6')     61.7       4.46,      62.1       4.46, nd
                                    4.41

Glcp"        H-C(1")     104.6      5.01       105.1      5.12
                                    (nd)                  (d, 7.4)
             H-C(2")     75.1       4.00       75.1       4.03
             H-C(3")     83.4       4.28       77.3       4.18
             H-C(4")     69.3       4.18       71.0       4.29
             H-C(5")     78.1       3.88       78.3       3.98
             CH2(6")     61.8       4.37,      62.1       4.34, nd
                                    4.28

Rhap"' (b)   H-C(1"')    101.7      6.15       101.5      6.27
                                    (brs)                 (brs)
             H-C(2"')    72.2       4.67       72.2       4.76
             H-C(3"')    72.4       4.53       72.5       4.59
             H-C(4"')    74.0       4.26       73.9       4.33
             H-C(5"')    69.3       4.86       69.4       4.95
             Me(6"')     18.4       1.73       18.5       1.77
                                    (d, 6.6)              (d. 6.7)
Rhap""       H-C(1"")    102.8      6.15
                                    (brs)
             H-C(2"")    72.3       4.72
             H-C(3"")    72.4       4.51
             H-C(4"")    74.0       4.29
             H-C(5"")    69.9       4.92
             Me(6"")     18.6       1.71
                                    (d. 6.7)

Glcp""'      H-C(1""')   104.7      4.86       104.7      4.85
                                    (d, 7.5)              (d, 7.5)
             H-C(2""')   75.1       4.01       75.0       4.02
             H-C(3""')   78.4       4.18       78.5       4.24
             H-C(4""')   71.8       4.17       71.4       4.24
             H-C(5""')   78.1       3.93       78.3       3.98
             CH2(6""')   62.9       4.52,      62.8       4.57,
                                    4.35                  4.40

Compound 5 showed the same NMR data of
carbohydrate moieties as compound 4.

(a) Glcp, [beta]-D-glucopyranosyl.

(b) Rhap, [alpha]-L-rhamnopyranosyl.

(c) nd, not determined.
COPYRIGHT 2015 Urban & Fischer Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Motaal, Amira Abdel; Askary, Hesham El-; Crockett, Sara; Kunert, Olaf; Sakr, Basma; Shaker, Sherif;
Publication:Phytomedicine: International Journal of Phytotherapy & Phytopharmacology
Article Type:Report
Date:Aug 15, 2015
Words:6887
Previous Article:Different anticancer effects of Saxifragifolin A on estrogen receptor-positive and estrogen receptor-negative breast cancer cells.
Next Article:Two triterpeniods from Cyclocarya paliurus (Batal) Iljinsk (Juglandaceae) promote glucose uptake in 3T3-L1 adipocytes: the relationship to AMPK...
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters