Phenolic Substances from - American Chemical Society

Article pubs.acs.org/jnp

Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX

Phenolic Substances from Ocimum Species Enhance GlucoseStimulated Insulin Secretion and Modulate the Expression of Key Insulin Regulatory Genes in Mice Pancreatic Islets Livia Marques Casanova,†,‡ Wenqian Gu,‡ Sônia Soares Costa,† and Per Bendix Jeppesen*,‡ †

Instituto de Pesquisas de Produtos Naturais, Centro de Ciências da Saúde, Bloco H, Cidade Universitária, Universidade Federal do Rio de Janeiro, 21 941-902 Rio de Janeiro, RJ, Brazil ‡ Department of Clinical Medicine, Aarhus University Hospital, Aarhus University, Tage-Hansens Gade 2, 8000 Aarhus C, Denmark ABSTRACT: Ocimum gratissimum and Ocimum basilicum are plants ethnopharmacologically used to treat diabetes mellitus, a life-threatening disease that affects millions of people worldwide. In order to further understand their antidiabetic potential, which has been previously demonstrated in animal models, we aimed to investigate the acute and chronic effects of major phenolic substances from both plants on insulin secretion and gene expression in pancreatic islets isolated from NMRI mice. Insulin secretion was measured after acute (1 h) and long-term (72 h) incubation of islets with one of four cinnamic acid derivatives (caftaric, caffeic, chicoric, and rosmarinic acids) or a C-glucosylated flavonoid (vicenin-2). All substances acutely enhanced glucose-stimulated insulin secretion (GSIS) from islets at concentrations from 10−10 to 10−6 M. They also increased GSIS after chronic incubation (10−8 M). None of them increased insulin secretion in the presence of low glucose concentration. Furthermore, these substances markedly changed the gene expression profile of key insulin regulatory genes INS1, INS2, PDX1, INSR, IRS1, and proliferative genes as well as glucose transporter 2 (GLUT2), in treated islets. Thus, they may play an important role in diabetes treatment. This is the first report on the insulin-secretory activity of caftaric acid, rosmarinic acid, and vicenin-2.

D

L. and Ocimum basilicum L., have been studied by our group recently.34,35 Aqueous leaf extracts of these plants have shown important hypoglycemic activity in streptozotocin (STZ)induced diabetic mice. Caffeic acid and its derivatives (caftaric, chicoric, and rosmarinic acids) are major phenolic compounds of both extracts. The C-glucosylated flavonoid vicenin-2 is a major compound of O. gratissimum extract; however, its presence was not detected in O. basilicum in our previous study. While chicoric acid presented moderate acute hypoglycemic activity in STZ-induced diabetic mice at a low dose (3 mg/kg), the other phenolic substances had no activity under the same conditions. A possible synergistic effect was postulated to explain the hypoglycemic activity of these extracts.34,35 However, little is known about how these phenolic substances may contribute to this effect and the possible mechanisms involved. Thus, further studies are necessary to clarify this issue, enabling not only the comprehension of the hypoglycemic activity of these species but also the identification of compounds potentially useful to treat DM. The aim of the present study was to investigate whether phenolic substances from O. basilicum and O. gratissimum have beneficial effects on insulin secretion, a potential mechanism

iabetes mellitus (DM) is a life-threatening disease that has reached epidemic proportions worldwide. According to the International Diabetes Federation, 415 million people were diabetic in 2015, and this number is expected to reach 642 million in 2040.1 Current DM therapies focus on the control of glycemic levels in order to avoid diabetic vascular complications. However, despite the many different drugs available for clinical use, these drugs can produce important side effects and are unable to prevent the progression of the disease or the development of diabetic complications.2−5 Taking this into consideration, it is highly relevant to search for new antidiabetic therapies. Medicinal plants and plant-derived substances are a promising source of new drugs to treat DM.4,6−17 Phenolic substances are widely distributed in the plant kingdom and are common dietary components. Many phenolic substances, especially flavonoids, phenolic acids, and stilbenes, have shown antidiabetic activity through mechanisms such as inhibition of glucose absorption in the gut, increase in insulin sensitivity in peripheral tissues, and stimulation of insulin secretion.18−24 The Ocimum L. genus (Lamiaceae) includes some species ethnopharmacologically used to treat DM in Africa and Asia.9,25−27 Their hypoglycemic activity was demonstrated by various in vivo studies.28−33 Two of them, Ocimum gratissimum © XXXX American Chemical Society and American Society of Pharmacognosy

Received: August 16, 2017

A

DOI: 10.1021/acs.jnatprod.7b00699 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 1. Major compounds from phenolic fractions of O. basilicum (OB-Ph) and O. gratissimum (OG-Ph). (A and B) Chromatograms (330 nm) of OB-Ph and OG-Ph (5 mg/mL), respectively; caftaric acid (peak 1; 12.1 min), vicenin-2 (peak 2; 15.1 min), caffeic acid (peak 3; 15.4 min), chicoric acid (peak 4; 17.3 min), rosmarinic acid (peak 5; 24.5 min.). (C) Chemical structures of caftaric acid (1), vicenin-2 (2), caffeic acid (3), chicoric acid (4), and rosmarinic acid (5). (D) Amount of caftaric, caffeic, chicoric, and rosmarinic acids and vicenin-2 are expressed in mg of substance per g of lyophilized OB-Ph and OG-Ph material.

Ph fractions or their pure phenolic substances. The exposure of islets to high glucose concentration significantly stimulated insulin secretion in all experiments when compared to islets exposed to low glucose concentration. Islets were incubated with OB-Ph and OG-Ph fractions at concentrations of 5, 0.05, and 0.0005 μg/mL in the presence of high glucose concentration (16.7 mM). Both fractions were able to acutely enhance insulin secretion at all tested concentrations in comparison with 16.7 mM glucose control. Insulin secretion was increased by 2.4-fold (p < 0.001) at 5 μg/ mL, 2.2-fold (p < 0.001) at 0.05 μg/mL, and 2.3-fold (p < 0.001) at 0.0005 μg/mL of OB-Ph. An increase of 2.7-fold (p < 0.0001) in insulin secretion was produced by OG-Ph at the highest concentration (5 μg/mL), while the concentrations of 0.05 and 0.0005 μg/mL both caused a 1.7-fold (p < 0.05) increase in insulin secretion. These results are depicted in Figure 2A and B. The bioactivity of phenolic fractions from O. basilicum and O. gratissimum revealed to be a promising result even at low doses. Therefore, the next step of the study was to assay their single phenolic compounds. Mice islets were incubated with caftaric, caffeic, chicoric, and rosmarinic acids, as well as vicenin-2 at concentrations of 10−6, 10−8, and 10−10 M in the presence of high glucose concentration. All of them significantly increased insulin secretion, as can be seen in Figure 2C−G. Caftaric and rosmarinic acids were the most active compounds. Insulin secretion was increased by 5.5-fold (p < 0.0001), 5.1-fold (p < 0.0001), and 2.8-fold (p = 0.0002) by 10−6, 10−8, and 10−10 M caftaric acid in comparison with 16.7 mM glucose control, respectively. At the concentrations of 10−6, 10−8, and 10−10 M, rosmarinic acid increased insulin secretion by 4.9-fold (p < 0.001), 2.8-fold (p < 0.001), and 3.3-

through which their plant extracts may exert hypoglycemic activity. In order to achieve this goal, we analyzed ex vivo the acute and chronic effect of major phenolic components of aqueous leaf extracts of these species on insulin secretion and gene expression of isolated pancreatic islets from mice.

■

RESULTS AND DISCUSSION Phenolic Composition of O. basilicum and O. gratissimum Extracts. Fractionation of O. basilicum (OB) and O. gratissimum (OG) aqueous leaf extracts through gel filtration chromatography led to two enriched-phenolic fractions, OB-Ph and OG-Ph, respectively. Each fraction corresponds to approximately 12% of the respective crude extract. The analysis of OB-Ph and OG-Ph fractions by HPLCDAD enabled the identification and quantification of five phenolic substances in both fractions: caftaric acid, vicenin-2, caffeic acid, chicoric acid, and rosmarinic acid, as depicted in Figure 1A−D. These phenolic substances were previously reported for the two species studied.34−36 Rosmarinic and chicoric acids were the major compounds of OB-Ph, while caftaric acid, chicoric acid, and vicenin-2 were the major compounds of OG-Ph (Figure 1D). Vicenin-2 was also detected in OB-Ph; however, it was not previously detected in crude O. basilicum aqueous extracts, probably due to its very low concentration.35 Acute Dose-Dependent Insulin Secretion Studies. For acute insulin secretion assays, isolated pancreatic islets from Naval Medical Research Institute (NMRI) female mice were incubated for 1 h in low and high glucose concentrations (3.3 and 16.7 mM, respectively) in the absence of the compounds. In parallel, pancreatic islets were incubated in high glucose concentration (16.7 mM) in the presence of OB-Ph and OGB



Article

Figure 2. Acute effects on insulin secretion from murine pancreatic islets after 60 min incubations with phenolic-enriched fractions from O. basilicum (OB-Ph; A), O. gratissimum (OG-Ph; B), caftaric acid (C), caffeic acid (D), chicoric acid (E), rosmarinic acid (F), or vicenin-2 (G). Islets were incubated for 60 min in 3.3 or 16.7 mM glucose without compounds (control) or in 16.7 mM glucose in the presence of the phenolic fractions at 0.0005, 0.05, or 5 μg/mL (A and B) or in the presence of the isolated phenolic substances at 10−10, 10−8, or 10−6 M concentrations (C−G). The medium was harvested and insulin content measured. Insulin levels are expressed as fold change of insulin secretion from 16.7 mM glucose control. Number of replicates per treatment = 24. Data are presented as mean ± SEM; *p < 0.05 in comparison with 16.7 mM glucose control.

The results are depicted in Figure 3A−D. None of the substances increased insulin secretion at low or normal glucose concentrations (3.3 and 6.6 mM, respectively), as can be seen in Figure 3A and B. Only rosmarinic acid potentiated insulin secretion at 11.1 mM glucose (68% increase; p < 0.01). At this glucose concentration, caffeic acid also showed a tendency to increase insulin secretion, however not statistically significant, as can be seen in Figure 3C. At 16.7 mM glucose, all compounds potentiated insulin secretion, in accordance with previous assays (Figure 3D). Chronic Insulin Secretion Studies. In order to assess the potential chronic effects of the five phenolic substances from OB and OG, pancreatic islets were incubated with these substances for 72 h at 10−8 M. After the incubation period, islets were stimulated for 60 min with 16.7 mM glucose. As expected, islet insulin release increased in the presence of this

fold (p < 0.001), respectively. Chicoric acid was also highly active at the concentration of 10−6 M (p < 0.001), increasing insulin secretion by 4.1-fold; however, the activity was much lower at the concentrations of 10−8 and 10−10 M, which produced a 1.9-fold (p < 0.01) and 1.8-fold (p < 0.01) increase in insulin secretion, respectively. Vicenin-2 produced a similar effect at all concentrations tested, increasing insulin secretion by 2.3-fold (p < 0.001) at 10−6 and 10−8 M and by 2.5-fold (p < 0.001) at 10−10 M. Caffeic acid was the less active compound. It significantly increased insulin secretion only at the concentrations of 10−6 and 10−8 M by 2-fold (p < 0.05) and 1.7-fold (p < 0.05), respectively. Acute Glucose-Dependent Insulin Secretion Studies. The acute glucose-dependent stimulatory effects of these five substances (10−6 M) on insulin release were subsequently assayed at glucose concentrations of 3.3, 6.6, 11.1, or 16.7 mM. C



Article

Figure 3. Glucose-dependent acute effects of phenolic substances from O. gratissimum and O. basilicum on insulin secretion from murine pancreatic islets after 60 min of incubation with 3.3 mM (A), 6.6 mM (B), 11.1 mM (C), or 16.7 mM glucose (D). Islets were incubated for 60 min in four different glucose concentrations in the presence or absence of caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, and vicenin-2 at 10−6 M concentration, following which the supernatant was harvested and insulin content measured. Insulin levels are expressed in ng/mL. Number of replicates per treatment = 16. Data are presented as mean ± SEM; *p < 0.05 in comparison with control.

high glucose concentration when compared with 3.3 mM glucose. Long-term (72 h) incubation of islets with all five substances resulted in a significant increase in insulin secretion at high glucose concentration (16.7 mM), while no effect was observed at low glucose concentration (3.3 mM), as depicted in Figure 4A−E. The caffeic acid derivativescaftaric, chicoric, and rosmarinic acidsexerted the most prominent stimulatory activity, enhancing insulin secretion by 3.2-fold (p < 0.001), 3.5-fold (p < 0.0001), and 3.4-fold (p < 0.0001), respectively, in comparison with 16.7 mM control. Caffeic acid itself was much less active, enhancing insulin secretion by 1.5-fold (p < 0.05). Vicenin-2 also potentiated insulin secretion, increasing it by 2.2fold (p < 0.001). Gene Expression Analysis. Gene expression analysis of eight genes related to regulation of GSIS and beta-cell function were studied after chronic treatment of islets (72 h) with the phenolic substances at 10−8 M, as illustrated in Figure 5. The analysis revealed a significant increase in insulin gene expression (INS1 and INS2; Figure 5A and B) in islets treated with caftaric, chicoric, and rosmarinic acids, while no effect was observed for those treated with caffeic acid and vicenin-2. The increase in insulin expression was consistent with the pronounced increase in glucose-stimulated insulin secretion (GSIS) observed under long-term treatment with these three substances, as shown above. Caffeic, chicoric, and rosmarinic acids enhanced the expression of the transcription factor PDX1 (Figure 5C), while the other substances had no effect on the abundance of

this factor. PDX1 is a glucose-responsive regulator of insulin gene expression that plays a critical role in maintaining the function of beta-cells.37,38 The increase in insulin gene expression produced by chicoric and rosmarinic acids correlates well with the increase in the expression of this transcription factor. However, caftaric acid had no effect on PDX1 abundance, even though it stimulated insulin expression. Caffeic acid, on the other hand, had no effect on insulin gene expression, even though it produced a pronounced increase in PDX1. Other transcription factors (e.g., Nkx2.2, Pax6, Foxa2, and Nkx6.1) regulate insulin gene expression and beta-cell function and thus could account for the expression profile observed for these substances.38 The expression of insulin receptor (INSR; Figure 5D) was enhanced in islets treated with all the substances, except caffeic acid. On the other hand, none of the substances affected the expression of glucagon receptors (GCGR; Figure 5F). Among the phenolic substances tested, only chicoric and rosmarinic acids significantly enhanced the expression of IRS1 (Figure 5E), a protein that is part of the insulin signaling pathway. Insulin acts in an autocrine manner in beta-cells to stimulate insulin gene expression and to regulate beta-cell mass through enhanced proliferation and decreased apoptosis.39 IRS-1 plays important biological functions for both metabolic and mitogenic (growth promoting) pathways: mice deficient in IRS1 have only a mild diabetic phenotype, but a pronounced growth impairment; that is, IRS-1 knockout mice reach only 50% of the weight of normal mice.40 D



Article

disappears from mature beta-cells and becomes an adult islet alpha-cell specific factor.43,44 The results of the gene expression analysis are summarized in Table 1. Potential Antidiabetic Activity of Phenolic Substances from Ocimum sp. To the best of our knowledge, this is the first report on the enhancing potential of caftaric acid and vicenin-2 on glucose-stimulated insulin secretion. Chicoric acid was previously demonstrated to enhance insulin secretion in rat pancreatic islets and in INS-1 cells.45 In this report, the substance acutely increased islet insulin secretion at the glucose concentration of 8.3 mM. In the present study, chicoric acid enhanced insulin secretion only at 16.7 mM glucose, and no increase was detected at 6.6 or 11.1 mM glucose. This divergence may be due to differences in the concentrations used. While we tested chicoric acid at 10−6 M to 10−10 M, Tousch et al.45 used a higher concentration (50 μg/ mL), near 10−4 M. Caffeic acid was previously found to increase insulin secretion in INS-1 cells under hyperglycemic conditions during acute and chronic exposure at concentrations of 10−6 to 10−10 M.21,46 We found similar results in mice islets, further confirming the potential of caffeic acid as a GSIS enhancer. In one of the previous studies, the gene expression of INS-1 by cells chronically treated with caffeic acid in the presence of normal glucose concentration was analyzed. Caffeic acid was shown to enhance insulin genes expression (INS1 and INS2). In the present study, the same effect was not observed in islets. However, similarly to our findings, the substance enhanced PDX1 expression.21 Rosmarinic acid was able to enhance serum insulin levels in alloxan-induced diabetic mice after 8 weeks of treatment (20 mg/mL). However, it was not clear if this effect was due to insulin secretagogue activity or the recovery/restoration of beta-cell mass after damages induced by alloxan.47 Interestingly, none of the substances exhibited acute stimulatory activity at low and normal glucose levels. In addition, islets chronically incubated with these phenolics for 72 h did not secrete more insulin than control at low glucose concentration. The absence of stimulatory activity at low/ normal glucose levels is an interesting feature, as it is likely to reduce their potential to induce insulin secretion when it is not necessary to decrease the glycemic levels. In previous studies by our group, the acute hypoglycemic activity of these substances was evaluated in STZ-induced diabetic mice at doses of 1.5 to 3 mg/kg. At the experimental conditions used, only chicoric acid was active.34,35 However, as this animal model relies on destruction of beta-cells to exert symptoms of DM, it is not adequate for detecting the activity of potential insulin secretion enhancers.48 Therefore, the hypoglycemic effect produced by chicoric acid may have been due to different mechanisms, for example producing an increase in muscle glucose uptake, as observed by Azay-Milau et al. (2013).49 Furthermore, it is possible that the doses used were too low and time of treatment was too short for the observation of hypoglycemic effects. Caffeic and rosmarinic acids, for instance, have exhibited in vivo hypoglycemic activity in studies using different animal models and/or experimental conditions.50−54 The mechanism by which these phenolic substances exert an insulin-secretory effect is not clear. Phenolics are reputed for their antioxidant activity, especially those bearing a catechol moiety, such as the four cinnamic acid derivatives in this study.55,56

Figure 4. Chronic effects on insulin secretion from mice pancreatic islets after 72 h of incubation with caftaric acid (A), caffeic acid (B), chicoric acid (C), rosmarinic acid (D), or vicenin-2 (E). Islets were incubated for 72 h in media containing 11.1 mM glucose in the absence or presence of the isolated phenolic substances at 10−8 M. After that, islets were treated for 60 min with either 3.3 or 16.7 mM glucose, after which the medium was harvested and insulin content measured. Insulin levels are expressed as fold change in insulin secretion from 16.7 mM glucose control. Number of replicates per treatment = 24. Data are presented as mean ± SEM; *p < 0.05 in comparison with respective control.

The treatment with rosmarinic and caffeic acids increased the GLUT2 (Figure 5G) expression, suggesting that the glucose sensitivity is increased. GLUT2 is the glucose transporter with the lowest affinity and the highest capacity for glucose, and this allows beta-cells to take up glucose effectively only in times of plenty, when insulin release is needed.41 The increased expression of GLUT2 may lead to an increased first phase insulin response, being thereby a very important regulator of glucose homeostasis.42 Additionally, caffeic acid increased the expression of MAFB (Figure 5H), while the other substances had no effect on this transcription factor. MAFB is detected in alpha- and beta-cells during developmental stages. However, it E



Article

Figure 5. Effects of the phenolic substances from O. gratissimum and O. basilicum on islet gene expression. mRNA abundance of INS1, INS2, PDX1, INSR, IRS1, GCGR, GLUT2, and MAFB in pancreatic islets incubated for 72 h with caftaric acid (CFA), caffeic acid (CA), chicoric acid (CHA), rosmarinic acid (RA), or vicenin-2 (V) at 10−8 M was studied by real time RT-PCR using TaqMan assays. Triplicate samples were taken for each treatment, and the samples were measured in triplicates. Gene expression was normalized to 18S rRNA. Difference in the mRNA abundance was calculated compared with untreated control. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 in comparison with respective control.

various phenolic substances, including caffeic acid, resulted in protection against oxidative stress. However, other intracellular or extracellular mechanisms that could play an important role for the improvement of insulin secretion by phenolic substances are possible. For instance, quercetin was shown to stimulate insulin secretion probably through KATP channel inhibition and simultaneous stimulation of voltage-sensitive Ca2+ channels in cell membrane.59 Besides the insulinotropic effect demonstrated in the present study, the phenolic substances from OB and OG may exert antidiabetic activity by additional potential mechanisms, as suggested by some previous studies discussed below. Caffeic acid was shown to in vivo regulate hepatic enzymes important to carbohydrate metabolism.53 In addition, it was able to attenuate insulin resistance in cultured hepatocytes and stimulate glucose uptake in 3T3-L1 preadipocytes and L6 myocytes.49,60,61 Rosmarinic acid was also shown to regulate key carbohydrate-metabolizing enzymes in the liver in vivo and to enhance insulin sensitivity.52,62,63 It also reduced insulin resistance in the skeletal muscle of rats through AMPK activation.64 Additionally, it was shown to in vivo modulate the trafficking of glucose

Table 1. Genes Upregulated in Islets Treated for 72 h with Phenolic Substances from O. gratissimum and O. basilicum (Cells Marked in Gray Represent Upregulated Genes; GCGR Is Not Shown, as the Tested Substances Had No Effect on Its Expression)

a

CFA: caftaric acid. bCA: caffeic acid. cCHA: chicoric acid. dRA: rosmarinic acid; eV: vicenin-2.

As beta-cells are especially vulnerable to oxidative stress, their antioxidant activity may be related to the insulinotropic effect.57 Lapidot58 showed that the preincubation of beta-cells with F



Article

together and lyophilized, resulting in a crude phenolic material (OB-B, 11.8 g). OB-B was divided in four aliquots (approximately 3 g), and each aliquot was injected in a Sephadex G-10 column (28.5 × 2.6 cm, water/20% ethanol/50% ethanol). The Sephadex G-10 column chromatography of each aliquot afforded again a nonphenolic fraction (water) and a phenolic fraction (water/20% ethanol/50% ethanol). The phenolic fractions were pooled and lyophilized (OB-B2, 2.4 g). OB-B2 was injected into a Sephadex LH-20 column (49 × 2.0 cm, water/20% ethanol/50% ethanol/100% ethanol), which afforded once more a nonphenolic fraction (water) and 2.3 g of a phenolic fraction (water/20% ethanol/50% ethanol/100% ethanol), which was named OB-Ph. This fraction was used for the islet assays. OG extract was submitted to the same fractionation process, using the same chromatographic columns described above. A sample of OG (20.4 g) was resuspended in distilled water (1.0 L) and precipitated with ethanol (1.0 L). The supernatant was concentrated in a waterbath and was subsequently lyophilized (15.0 g). Fractionation of the supernatant in a Sephadex G-25 column in four stages afforded a nonphenolic fraction and a phenolic-containing fraction (OG-B, 12.7 g). OG-B was submitted to a G-10 column in four stages, yielding a nonphenolic fraction and a phenolic fraction (OG-B2; 3.2 g). OG-B2 was injected into a Sephadex LH-20 column, which yielded a nonphenolic fraction and 2.1 g of a phenolic fraction, which was named OG-Ph. This fraction was used for the pancreatic islet assays. HPLC-DAD Analyses. A sample of each extract (10 mg) and fraction (5 mg) was diluted in a mixture (1 mL) of distilled water and acetonitrile (10:1). The injection volume was 20 μL. The mobile phase consisted of water containing 0.1% formic acid (eluent A) and acetonitrile containing 0.1% formic acid (eluent B). The samples were run for 50 min at 1 mL/min, and the absorbance was monitored between 200 and 500 nm. The gradient used was as follows: 0−10 min (100−80% A), 10−20 min (80−78% A), 20−30 min (78−75% A), 30−35 min (75−70% A), 35−40 min (70−50% A), 40−45 min (50− 30% A), and 45−50 min (30−0% A), as described by Casanova et al.34 The constituents of the extract and fractions were identified by comparing the retention time and UV spectrum of peaks with those of reference samples and confirmed by co-injection with these reference standards. Cinnamic acid derivatives were quantified by plotting the peak areas at 330 nm against the calibration curves of chicoric acid (y = 65061.5x − 368526.6; R2 = 0.997), rosmarinic acid (y = 36567.6x − 193620.8; R2 = 0.999), caffeic acid (y = 98160.6x − 3734.1; R2 = 0.999), and vitexin (y = 26093.2x − 1316.3; R2 = 0.997). The former was used for vicenin-2 quantification after mass correction. Caftaric acid quantification was based on the caffeic acid calibration curve with proper mass correction. Each analysis was carried out in triplicate. Animals. For ex vivo studies with pancreatic islets, 125 Naval Medical Research Institute (NMRI) female mice (Taconic, Ry, Denmark) were used, all weighing 22−25g. A 12 h light/dark cycle was used. This study was approved by the Danish Council for Animal Experiments. Pancreatic Islet Isolation. Islets were isolated by the collagenase digestion technique from adult female NMRI mice (Taconic) weighing 22 to 25 g, as described previously.69 In brief, the animals were anaesthetized with pentobarbital (50 mg/kg intraperitoneally) and submitted to midline laparotomy. The pancreas was retrogradely filled with 3 mL of ice-cold Hanks balanced salt solution (HBSS) (Sigma-Aldrich) supplemented with 0.3 mg/mL collagenase (Boehringer Mannheim GmbH). The pancreas was subsequently removed and incubated for 19 min at 37 °C. After rinsing in HBSS, the islets were handpicked under a stereomicroscope and immediately transferred to plates with 10 mL of Roswell Park Memorial Institute (RPMI) 1640 containing 11.1 mM glucose supplemented with penicillin G, 10% inactivated fetal calf serum, and 2.06 mM Lglutamine (all Gibco/Invitrogen). The isolated islets were incubated overnight at 37 °C and 95% normal atmosphere/5% CO2. Incubation of Pancreatic Islets for Acute Dose-Dependent Exposure Assay. In the acute insulin secretion study, the phenolic fractions of O. basilicum and O. gratissimum as well as their major phenolic substances (caftaric acid, caffeic acid, chicoric acid, rosmarinic

transporters SGLT1 in enterocyte brush-border membrane, thus influencing oral glucose absorption.50 Chicoric acid, in its turn, was shown to inhibit in vitro the enzyme tyrosine phosphatase 1B (PT1B), a negative regulator of the insulin signaling pathways, and to stimulate the AMPkinase pathway, which plays an important role in glucose metabolism.65,66 Finally, vicenin-2 was shown to in vitro inhibit PT1B and alpha-glucosidase enzymes and to enhance glucose uptake by 3T3-L1 preadipocytes.67,68 However, studies on biological activities of caftaric acid are scarce. This is, as far as we know, the first study to report the antidiabetic potential of this substance. Concluding Remarks. The major phenolic substances from O. basilicum and O. gratissimum, namely, caffeic acid and its derivatives caftaric acid, chicoric acid, and rosmarinic acid as well as the flavonoid vicenin-2, seem to possess short- and longterm beneficial effects on glucose-stimulated insulin secretion without stimulating insulin release at low glucose concentration. This study is the first to demonstrate the effects of caftaric acid, rosmarinic acid, and vicenin-2 on insulin secretion and islet gene expression. Further research is necessary to confirm our findings in vivo and to characterize the mechanisms of action of these compounds. Our findings, however, reinforce the potential of O. basilicum and O. gratissimum as antidiabetic agents.

■

EXPERIMENTAL SECTION

General Experimental Procedures. Size-exclusion chromatography was performed on Sephadex LH-20 gel (25−100 nm, SigmaAldrich, St. Louis, MO, USA), Sephadex G-25 (17−132 μm, Pharmacia Fine Chemicals, Uppsala, Sweden), or Sephadex G-10 (40−120 μm, Sigma-Aldrich). The fractionation steps were monitored by TLC on silica 60 F254 using n-butanol/acetic acid/water (BAW, 3:1:1), visualized under UV, and revealed with cerium sulfate solution or NP/PEG. Caffeic acid and chicoric acid were purchased from Sigma-Aldrich. Rosmarinic acid was purchased from MP Biomedicals (Santa Ana, CA, USA). Caftaric acid and vicenin-2 were isolated from O. gratissimum L. leaves as described elsewhere.34 HPLC-DAD analyses were performed on a Shimadzu liquid chromatograph LC20AT with a diode-array wavelength SPD-M20A detector (Laboratório de Produtos Naturais, UFRJ, Macaé Campus), using a Merck reverse-phase column C-18 (5 μm, 250 mm, 4 mm). Plant Material. O. gratissimum and O. basilicum leaves were collected in the blooming season from specimens cultivated in Barra do Piraı ́, RJ (Brazil), from June to October 2014. O. basilicum and O. gratissimum specimens were identified by Prof. Dr. Élide Pereira dos Santos (UFPR, Brazil) and Prof. Dr. Rosana Conrado Lopes (UFRJ, Brazil), respectively. Voucher specimens (RFA 39921, O. basilicum; RFA35593, O. gratissimum) were deposited at the herbarium of the Institute of Biology (UFRJ, Brazil). Extraction and Fractionation. Fresh leaves from both species O. gratissimum (907 g) and O. basilicum (932 g)were triturated in a food processor and boiled with distilled water (10% w/v) for 10 min. The extracts of OB and OG were filtered and lyophilized, affording 41.7 and 39.5 g of dried material, respectively. The fractionation process of these extracts was based on precipitation followed by gel filtration chromatography, as described below. A sample of OB extract (21.6 g) was resuspended in distilled water (1.0 L) and precipitated with the addition of ethanol (1.0 L). The supernatant was concentrated in a water-bath until complete evaporation of the ethanol and was subsequently frozen and lyophilized, affording a dried material (14.0 g). It was then divided in four aliquots of approximately 3.5 g. Each aliquot was subjected to Sephadex G-25 column chromatography (19.0 × 5.5 cm, water/20% ethanol). Each chromatographic process yielded a nonphenolic fraction (water) that eluted first and a phenolic-containing fraction (water/20% ethanol). Phenolic fractions of each column were pooled G



Article

acid, and vicenin-2) were tested. OB-Ph and OG-Ph were dissolved in sterile water at a stock concentration of 500 μg/mL. All the phenolic substances were dissolved in sterile water at a stock concentration of 10−4 M. The stock solutions were serially diluted in a modified Krebs− Ringer buffer (KRB) supplemented with 16.7 mM glucose. The KRB contained 125 mM NaCl, 5.9 mM KCl, 1.2 mM MgCl2, 1.28 mM CaCl2, and 25 mM HEPES (pH 7.4; all Sigma). OB-Ph and OG-Ph were diluted to the final concentrations of 5, 0.05, and 0.0005 μg/mL, while the phenolic substances were dissolved to final concentrations of 10−6, 10−8, and 10−10 M. The NMRI female mice islets were rinsed twice with a modified KRB supplemented with 3.3 mM glucose and 0.1% bovine serum albumin (Sigma). After 30 min of preincubation in normal atmosphere at 37 °C, single islets were handpicked and incubated in 100 μL of KRB containing glucose (3.3 and 16.7 mM) and glucose (16.7 mM) in the presence of OB-Ph and OG-Ph (5, 0.05, and 0.0005 μg/mL) or caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, and vicenin-2 (10−6, 10−8, and 10−10 M). After 60 min of incubation in normal atmosphere at 37 °C, 50 μL of the medium was frozen for RIA analysis of insulin. Incubation of Pancreatic Islets for Acute Glucose-Dependent Exposure Assay. The effect of major OB and OG phenolic substances (caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, and vicenin-2) was evaluated on insulin secretion in the presence of different glucose concentrations. The stock solutions (10−4 M) were serially diluted in a modified KRB supplemented with 3.3, 6.6, 11.1, or 16.7 mM glucose to the final concentration of 10−6 M. After twice rinsing with modified KRB supplemented with 3.3 mM glucose and 0.1% bovine serum albumin (Sigma) and preincubation for 30 min in normal atmosphere at 37 °C, single islets were handpicked. Then, these islets were incubated in 100 μL of KRB containing glucose (3.3, 6.6, 11.1, and 16.7 mM) in the presence or absence of caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, and vicenin-2 (10−6 M). After 60 min of incubation in normal atmosphere at 37 °C, 50 μL of the medium was frozen for RIA analysis of insulin. Incubation of Pancreatic Islets for Chronic Exposure Assay. The major OB and OG phenolic substances (caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, and vicenin-2) were evaluated in chronic insulin secretion assays. The 10−4 M stock solutions of the phenolics were serially diluted in RPMI with 11.1 mM glucose medium (supplemented with penicillin G, 10% inactivated fetal calf serum, and 2.06 mM L-glutamine) to the final concentration of 10−8 M. After overnight incubation, mice pancreatic islets were preincubated for 72 h in RPMI with 11.1 mM glucose in the presence or absence of caftaric acid, caffeic acid, chicoric acid, rosmarinic acid, or vicenin-2. After this period, islets were rinsed once with modified KRB supplemented with 3.3 mM glucose and 0.1% bovine serum albumin (Sigma) and preincubated for 30 min in normal atmosphere at 37 °C. Single islets were handpicked and incubated in 100 μL of KRB with glucose concentrations of 3.3 or 16.7 mM. After 60 min of incubation in normal atmosphere at 37 °C, 50 μL of the medium was frozen for RIA analysis of insulin. Insulin Assay. Insulin was analyzed by radioimmunoassay using a guinea pig anti-porcine insulin antibody (Novo Nordisk, Bagsvaerd, Denmark) and mono-125I-(Tyr A14)-labeled human insulin (Novo Nordisk) as tracer and rat insulin as standard (Novo Nordisk). The separation of bound and free radioactivity was performed using ethanol, as previously described by Jeppesen et al.69 Isolation of RNA from Islets. For each group, islets from two to three mice (180−200 islets) were pooled in 1 mL of Trizol reagent (Gibco/Invitrogen) before RNA purification. Total RNA was extracted according to the manufacturer’s instructions. RNA was quantified by measuring absorbance at 260 and 280 nm. The quality of the RNA was checked by an Agilent 2100 bioanalyzer (Agilent, Santa Clara, CA, USA). Quantitative Real-Time PCR. Quantitative real-time PCR was performed using the Fluidigm BioMark System (AROS, Applied Biotechnology AS, Denmark). The samples were analyzed for expression of different gene transcripts using mouse-specific TaqMan

assays. Two specific assays were used as endogenous control: 18S (ABI, Hs99999901_s1) and Hprt1 (ABI, Mm00446968_m1). A list of the TaqMan assays used is given in Table 2. The sequence of the

Table 2. Applied Biosystem TaqMan Assays for Real-Time Quantitative RT-PCR gene symbol

assay ID, TaqMan, Applied Biosystem

GCGR INS1 INS2 INSR IRS1 MAFB PDX1 Slc2a2 (GLUT2)

Rn00597158_m1 Mm01259683_g1 Mm00731595_gH Mm00439693_m1 Mm01278327_m1 Mm00627481_s1 Mm00435565_m1 Mm00446229_m1

primers can be found at https://www.thermofisher.com/dk/en/ home/brands/applied-biosystems.html. Samples were analyzed using Fluidigm 96.96 Dynamic (Fluidigm catalog no. BMK-M-96.96) arrays with assay triplicates in accordance with the manufacturer’s protocol. One hundred nanograms of RNA was used as input in a 20 μL reverse transcript reaction using the High Capacity cDNA reverse transcription kit (ABI, PN4368813) in accordance with the manufacturer’s protocol. After reverse transcription, the cDNA samples were amplified according to the instructions given in the Fluidigm Specific Target Amplification Quick Reference Manual. In short, the cDNA was amplified using a target-specific assay (diluted 1:100) and TaqMan PreAmp Master mix (2×) (ABI, PN 4391128) in a 14-cycle thermal cycler reaction: 95 °C for 10 min and 14 cycles of 95 °C for 15 s and 60 °C for 4 min. Amplification was performed using the standard conditions: 50 °C for 2 min, 95 °C for 10 min, and 40 cycles of 95 °C, 15 s, and 60 °C, 1 min. The relative gene expression was calculated using the (1 + efficiencies) − ΔCT method, and the fold change was used to compare the expression levels. ΔCT is the difference in cycle threshold (CT) value between each target gene based on the average of the triplicates and the geometric mean CT values of 18S and Hprt1. Statistical Analysis. All statistical analyses were performed with GraphPad Prism Software (San Diego, CA, USA). Statistical significance between two groups was evaluated using unpaired Student’s t test. Data are presented as means ± SEM; p-values of

Phenolic Substances from - American Chemical Society

Recommend Documents