Compositions comprising urolithins and uses thereof for the stimulation of insulin secretion

Abstract

The present invention relates to a composition comprising urolithin A, urolithin B, urolithin C, urolithin D, or a combination thereof, for the stimulation of insulin secretion, and to the use of a compound chosen among urolithin A, urolithin B, urolithin C, urolithin D, or a combination thereof, intended for the stimulation of insulin secretion. The present invention also relates to a composition comprising an effective amount of urolithin B, urolithin C, urolithin D, or a combination thereof, for the treatment or the prevention of diabetes mellitus, and in particular for the treatment or the prevention of type 2 diabetes, and to the use of a compound chosen among urolithin B, urolithin C, urolithin D, and a combination thereof, intended for the treatment or the prevention of diabetes mellitus, and in particular of type 2 diabetes.

Claims

1. A composition comprising urolithin C and a second compound able to stimulate insulin secretion selected from the group consisting of sulfonylureas, GLP-1 analogs, and DPP4 inhibitors.

2. The composition of claim 1, wherein the compound able to stimulate insulin secretion is a sulfonylurea.

3. The composition of claim 2, wherein the compound able to stimulate insulin secretion is a sulfonylurea selected from the group consisting of Metahexamide, Glibenclamide (Glyburide), Carbutamide, Acetohexamide, Chlorpropamide, Tolbutamide, Tolazamide, Glipizide, Gliclazide, Glibornuride, Gliquidone, Glisoxepide, Glyclopyramide, and Glimepiride.

4. The composition of claim 3, wherein the compound able to stimulate insulin secretion is Glibenclamide.

5. The composition of claim 1, wherein the compound able to stimulate insulin secretion is a GLP-1 analog.

6. The composition of claim 5, wherein the compound able to stimulate insulin secretion is a GLP-1 analog selected from the group consisting of exenatide and liraglutide.

7. The composition of claim 6, wherein the compound able to stimulate insulin secretion is exenatide.

8. The composition of claim 1, wherein the compound able to stimulate insulin secretion is a DPP4 inhibitor.

9. The composition of claim 8, wherein the compound able to stimulate insulin secretion is a DPP4 inhibitor selected from the group consisting of Alogliptin, Gemigliptin, Linagliptin, Saxagliptin, Sitagliptin, and Vildagliptin.

10. A pharmaceutical formulation for the treatment of diabetes mellitus comprising an effective amount of the composition of claim 1 in a pharmaceutically acceptable vehicle.

11. A food product or a nutritional supplement for the stimulation of insulin secretion comprising an effective amount of the composition of claim 1 and a nutritionally acceptable vehicle.

12. A swallowable tablet, chewable tablet, effervescent tablet, capsule, pill, powder, granule, oral solution, oral suspension, sublingual dosage form, or buccal dosage form, comprising the composition of claim 1 and a carrier that is acceptable for nutritional usage.

13. A pharmaceutical formulation, food product, or a nutritional supplement comprising the composition of claim 1 for the treatment of a patient chosen from among the group consisting of a prediabetic patient, a patient whose fasting glycemia is comprised between 1 and 1.25 g L.sup.−1 and a patient whose post-prandial glycemia is comprised between 1.40 and 1.99 g L.sup.−1.

14. A pharmaceutical formulation, food product, or a nutritional supplement comprising the composition of claim 1 for the treatment of a diabetic patient, a patient whose fasting glycemia is equal or superior to 1.26 g.Math.L.sup.−1 and a patient whose post-prandial glycemia is equal or superior to 2 g.Math.L.sup.−1.

15. A combination product comprising the components of the composition of claim 1, provided for simultaneous, separate or sequential use.

Description

LEGENDS OF THE FIGURES

(1) FIG. 1: Representation of potential pathways for the conversion of ellagitannins to urolithins

(2) The acorn ellagitannins release ellagic acid, which is metabolized sequentially by intestinal microbiota, producing urolithin D, urolithin C, urolithin A and urolithin B.

(3) FIGS. 2A to 2D: Histogram representation of the concentration-response study of urolithins on insulin secretion determined in INS-1 insulin-secreting β-cells on insulin secretion under low-glucose (1.4 mmol.Math.L.sup.−1) or glucose-stimulated (8.3 mmol.Math.L.sup.−1) secretion conditions.

(4) FIGS. 2A-D illustrate the effects of urolithin A (FIG. 2A), urolithin B (FIG. 2B), urolithin C (FIG. 2C) or urolithin D (FIG. 2D) on insulin secretion determined under low-glucose (1.4 mmol.Math.L.sup.−1, left bars) or glucose (8.3 mmol.Math.L.sup.−1)-stimulated secretion (right bars) conditions. Data are expressed as percentage of insulin secretion in the presence of 8.3 mmol.Math.L.sup.−1 glucose alone (“control 100%”). In each case, insulin secretion was determined in the presence of urolithins at concentrations varying from 0 to 20 gmol.Math.L.sup.−1′. Results are presented as means±SEM of 4-6 separate experiments. A multiple comparison analysis of data was performed for each experimental condition. For each figure, different letters at the top of the bars correspond to statistically significant differences (p<0.05) between data obtained under the same experimental conditions.

(5) FIGS. 3A to 3D: Histogram representation of the concentration-response effect of urolithins on insulin secretion determined in INS-1 insulin-secreting β-cells, in the presence of glibenclamide.

(6) FIGS. 3A-D illustrate the effects of urolithin A (FIG. 3A), urolithin B (FIG. 3B), urolithin C (FIG. 3C) or urolithin D (FIG. 3D) on insulin secretion determined in the presence of 1.4 mmol.Math.L.sup.−1 glucose and 0.01 gmol.Math.L.sup.−1 glibenclamide. In each case, insulin secretion was determined in the presence of urolithins at concentrations varying from 0 (“control 100%”) to 20 gmol.Math.L.sup.−1. Results are presented as means±SEM of 4-6 separate experiments. A multiple comparison analysis of data was performed for each experimental condition. For each figure, different letters at the top of the bars correspond to statistically significant differences (p<0.05) between data obtained under the same experimental conditions.

(7) FIGS. 4A to 4D: Histogram representation of the concentration-response effect of urolithins on insulin secretion determined in INS-1 insulin-secreting β-cells in the presence of exendin.

(8) FIGS. 4A-D illustrate the effects of urolithin A (FIG. 4A), urolithin B (FIG. 4B), urolithin C (FIG. 4C) or urolithin D (FIG. 4D) on insulin secretion determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose and 0.001 gmol.Math.L.sup.−1 exendin. In each case, insulin secretion was determined in the presence of urolithins at concentrations varying from 0 (“control 100%”) to 20 gmol.Math.L.sup.−1′. Results are presented as means±SEM of 4-6 separate experiments. A multiple comparison analysis of data was performed for each experimental condition. For each figure, different letters at the top of the bars correspond to statistically significant differences (p<0.05) between data obtained under the same experimental conditions.

(9) FIGS. 5A to 5D: Histogram representation of the concentration-response effect of urolithins on insulin secretion determined in INS-1 insulin-secreting β-cells in the presence of oxidative stress.

(10) FIGS. 5A-D illustrate the effects of urolithin A (FIG. 5A), urolithin B (FIG. 5B), urolithin C (FIG. 5C) or urolithin D (FIG. 5D) on insulin secretion determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose and 50 μmol.Math.L.sup.−1 H.sub.2O.sub.2. In each case, insulin secretion was determined in the presence of urolithins at concentrations varying from 0 to 20 μmol.Math.L.sup.−1. Data are expressed as percentage of insulin secretion determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose (“control 100%”). Results are presented as means±SEM of 3 determinations. A multiple comparison analysis of data was performed for each experimental condition. For each figure, different letters at the top of the bars correspond to statistically significant differences (p<0.05) between data obtained under the same experimental conditions.

(11) FIG. 6: Histogram representation of the effect of urolithins on cell viability determined in INS-1 insulin-secreting β-cells in the presence of oxidative stress.

(12) FIG. 6 illustrates the effects of 20 μmol.Math.L.sup.−1 of urolithin A, urolithin B, urolithin C or urolithin D on cell viability determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose and 50 mol.Math.L.sup.−1 H.sub.2O.sub.2. Data are expressed as percentage of cell viability determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose (“control 100%”). Results are presented as means±SEM of 4 determinations. A multiple comparison analysis of data was performed for each experimental condition. Different letters at the top of the bars correspond to statistically significant differences (p<0.05) between data obtained under the same experimental conditions.

(13) FIGS. 7A and 7B: Effect of urolithin C on insulin secretion determined in the rat isolated perfused pancreas.

(14) FIGS. 7A and 7B illustrate the effect of urolithin C (10 μmol.Math.L.sup.−1) under 8.3 or 5 mmol.Math.L.sup.−1 glucose conditions, respectively. Rat pancreas was surgically isolated and perfused (2.5 ml/min) with modified Krebs-Ringer bicarbonate buffer containing 8.3 mmol.Math.L.sup.−1 (FIG. 7A) or 5 mmol.Math.L.sup.−1 (FIG. 7B) glucose in the absence or presence of urolithin C (10 μmol.Math.L.sup.−1). Samples collected for 1 minute were taken at the indicated times and insulin concentration determined (HTRF insulin assay kit, Cisbio International, Bagnols-sur-Ceze, France). Amounts of insulin (ng) secreted per minute are indicated as a function of time (minutes).

EXAMPLES

Example 1: Insulin Secretion in the Presence of Catabolites and Metabolites of Polyphenols in the INS-1 Insulin Secreting β-Cells

(15) 30 polyphenol metabolites (see Table I), used at the concentrations of 2 and 20 gmol.Math.L.sup.−1, were blind-screened for their capacities to modulate glucose (8.3 mmol.Math.L.sup.−1)-stimulated insulin secretion. In table I, data are expressed as percent of glucose (8.3 mmol.Math.L.sup.−1)-stimulated secretion (“internal standard”).

(16) Obtention of Molecules:

(17) Urolithins A and B were synthesized by Cu.sup.II-mediated coupling of 1,3-dihydroxybenzene and 2-bromo-5-hydroxybenzoic acid (urolithin A) or 2-bromobenzoic acid (urolithin B) in concentrated aqueous NaOH according to already reported procedures (Bialonska D. et al., 20092-Bromo-5-hydroxybenzoic acid was obtained from commercially available 2-bromo-5-methoxybenzoic acid through demethylation by BBr3 (3 equiv.) in CH.sub.2Cl.sub.2/hexane (3:2, v/v) at low temperature (−20 to 0° C.). Urolithins were isolated by precipitation and their purity checked by HPLC analysis. Their NMR and MS characteristics were in agreement with the literature. Urolithin C and D were purchased from Dalton Pharma Services (Toronto, Canada) and the certificates of analysis are available. Other compounds were commercially available.

(18) Reagents for the Determination of Insulin Secretion

(19) RPMI-1640 media, fetal calf serum (FCS), HEPES solution, sodium pyruvate solution and Dulbecco's phosphate buffered (PBS) were purchased from Lonza (Levallois Perret, France). All the other chemicals and compounds as dimethyl sulfoxide (DMSO), 2-mercaptoethanol, L-glutamine-penicillin-streptomycin solution, albumin from bovine serum (BSA), poly-L-lysine, HEPES, NaHCO.sub.3, KH.sub.2PO.sub.4, NaCl, KCl, CaCl.sub.2, MgSO.sub.4, exendin and glibenclamide were obtained from Sigma-Aldrich (St. Louis, Mo., USA).

(20) INS-1 Cell Culture

(21) The insulin-secreting cell line INS-1 (a gift from Professor C. B. Wollheim) was cultured in RPMI-1640, supplemented with 10% fetal calf serum (FCS), 100 U.Math.mL.sup.−1 penicillin, 100 μg.Math.mL.sup.−1 streptomycin, 2 mmol.Math.L.sup.−1 L-glutamine, 10 mmol.Math.L.sup.−1 HEPES, 1 mmol.Math.L.sup.−1 sodium pyruvate, and 50 mmol.Math.L.sup.−1 2-mercaptoethanol, as previously described by Youl et al. (2010). Cells were seeded in 24-well plates (4×10.sup.5 cells per well) and were used for experiments after 5 days of culture.

(22) Determination of Insulin Secretion

(23) Before the treatment, RPMI medium was removed and the cells were washed twice with HEPES-balanced Krebs-Ringer bicarbonate buffer (KRB) containing (in mmol.Math.L.sup.−1): 123 NaCl, 5.4 KCl, 1.3 KH.sub.2PO.sub.4, 2.7 MgSO.sub.4, 2.9 CaCl.sub.2, 5 NaHCO.sub.3 and 20 HEPES, pH 7.5, with 2 g.Math.L.sup.−1 bovine serum albumin (KRB/BSA). Cells were incubated for 1 h (5% CO.sub.2, 37° C.) in KRB medium containing 8.3 mmol.Math.L.sup.−1 glucose (glucose-stimulated secretion) in the presence or absence of the compounds tested. At the end of the 1 h incubation period, the medium was sampled and stored at −20° C. until the insulin assay. Insulin concentration in cell supernatants was determined using the homogeneous time resolved fluorescence technology (HTRF), according to the manufacturer's instructions (HTRF insulin assay kit, Cisbio International, Bagnols-sur-Ceze, France). Briefly, two anti-insulin antibodies were used; one labelled with Eu.sup.3+-Cryptate and one labelled with XL665 recognizing distinct epitopes. When these two fluorophores bind to insulin molecules, the two antibodies come into close proximity, allowing fluorescence resonance energy transfer (FRET) to occur between the Eu.sup.3+-Cryptate and the XL665. This FRET increases proportionally with the insulin concentrations. All the experiments were performed at least in quadruplicate.

(24) Results

(25) Result of blind-screening of polyphenols metabolites are indicated on Table I which shows the effect of various human phase II and colonic metabolites/catabolites of polyphenols on glucose (8.3 mmol.Math.L.sup.−1)-stimulated insulin secretion. A positive control, quercetin, was tested under the same conditions. Control response corresponds to insulin secretion determined in the presence of 8.3 mmol.Math.L.sup.−1 glucose alone.

(26) TABLE-US-00001 TABLE I % control response Metabolites/Catabolites 2 μM 20 μM From flavonols Phloroglucinol 100.00 105.00 Protocatechuic acid 108.00 92.00 4-hydroxybenzoic acid 96.00 98.00 3,4-dihydroxyphenylacetic acid 105.00 84.00 Quercetin-3 -glucuronide 89.00 103.00 Kaempferol-3 -glucuronide 111.00 104.00 From chlorogenic acids Caffeic acid 104.00 105.00 Dihydrocaffeic acid 110.00 83.00 Ferulic acid 109.00 115.00 Dihydroferulic acid 106.00 103.00 Feruloylglycine 102.00 102.00 Isoferuloylglycine 80.00 82.00 From flavan-3-ols 4′-hydroxyphenylacetic acid 102.00 74.00 3-O-Methylgallic acid 93.00 105.00 4-O-Methylgallic acid 104.00 86.00 Pyrogallol 93.00 98.00 Homovanillic acid 93.00 83.00 From cyanidin-based anthocyanins Tyrosol 86.00 89.00 4′-hydroxyhippuric acid 98.00 111.00 3-(4′-hydroxyphenyl)lactic acid 97.00 79.00 3-(3′-hydroxyphenyl)propionic acid 98.00 103.00 3-hydroxyphenylacetic acid 98.00 104.00 4′-hydroxyphenylacetic acid 107.00 81.00 Pyrocatechol 87.00 96.00 4′hydroxymandelic acid 103.00 86.00 From sanguiin H-6 ellagitanin Urolithin A (URO A) 109.82 162.50 Urolithin B (URO B) 108.77 145.61 Urolithin C (URO C) 107.14 194.64 Urolithin D (URO D) 88.79 158.62 Ellagic acid 93.36 100.92 Quercetin (positive control) 93.80 237.00

(27) The blind screening of 30 polyphenolic metabolites indicates that the only metabolites/catabolites able to amplify glucose (8.3 mmol.Math.L.sup.−1)-stimulated insulin secretion are urolithins A, B, C and D. The urolithin precursor, ellagic acid, is not active.

(28) Example 2: Effect of Urolithins a, B, C or D Determined in INS-1 Insulin Secreting β-Cells in the Presence of Low Glucose (1.4 Mmol.Math.L.sup.−1) or Glucose (8.3 Mmol.Math.L.sup.−1)-Stimulated Secretion Conditions (FIGS. 2A-D)

(29) In the experiments described below, concentration-response studies of active compounds selected from example 1 were performed and reproduced four times under low-glucose (1.4 mmol.Math.L.sup.−1) or glucose (8.3 mmol.Math.L.sup.−1)-stimulated secretion conditions.

(30) Material and methods were the same as described in example 1. For the determination of insulin secretion, cells were incubated for 1 h (5% CO.sub.2, 37° C.) in KRB medium containing either 1.4 mmol.Math.L.sup.−1 glucose (low-glucose) or 8.3 mmol.Math.L.sup.−1 glucose (glucose-stimulated secretion) in the presence or absence of the compounds tested. At the end of the 1 h incubation period, mediums were sampled and stored at −20° C. until the insulin assay. Insulin concentration in cell supernatants was determined as described in Example 1.

(31) Data are expressed as percent of glucose (8.3 mmol.Math.L.sup.−1)-stimulated secretion (“control 100%”) in FIGS. 2A-D.

(32) Results:

(33) In low-glucose condition, urolithin A did not significantly stimulate insulin secretion while urolithins B, C and D induced some stimulation at the maximal concentration tested (20 gmol.Math.L.sup.−1). In glucose-stimulated secretion condition, urolithin C appeared as the compound producing the greatest stimulation (about 200%) at the maximal concentration tested (20 gmol.Math.L.sup.−1). Regarding active concentrations, urolithins A, C and D displayed a clear concentration-dependent effect, the concentration inducing 50% of the maximal stimulation (EC.sub.50) being around 5 to 10 gmol.Math.L.sup.−1. Urolithin B was active for 20 gmol.Math.L.sup.−1 only.

Example 3: Effect of Urolithins A, B, C or D Determined in INS-1 Insulin Secreting β-Cells in the Presence of Glibenclamide (FIGS. 3A-D)

(34) Glibenclamide is an insulin secretion stimulant sulfonylurea used in the treatment of diabetes. Like other sulfonylureas, glibenclamide is able to stimulate insulin secretion even at low glucose concentrations. In the experiments described below, glibenclamide concentration (0.01 μmol.Math.L.sup.−1) was chosen as able to induce a 4/6-fold increase of secretion as compared to low-glucose condition.

(35) Material and methods were the same as described in example 1. For the determination of insulin secretion, cells were incubated for 1 h (5% CO.sub.2, 37° C.) in KRB medium containing 1.4 mmol.Math.L.sup.−1 glucose (low-glucose condition) and 0.01 μmol.Math.L.sup.−1 glibenclamide (glibenclamide-stimulated secretion), in the presence or absence of urolithins. At the end of the 1 h incubation period, the medium was sampled and stored at −20° C. until the insulin assay. Insulin concentration in cell supernatants was determined as described in Example 1.

(36) Data are expressed as percent of glibenclamide-stimulated insulin secretion (“control 100%”) in FIGS. 3A-D.

(37) Results:

(38) In the presence of glibenclamide, urolithins A, B, C and D concentration-dependently stimulated insulin secretion, again with some differences regarding amplitude of responses and active concentrations. Regarding maximal response, urolithin C appeared as the most active compound (about 200% stimulation as compared to glibenclamide alone), followed by urolithins A/D (150-160%) and B (130%). Regarding active concentrations, maximal effect was obtained for urolithin A at 5 μmol.Math.L.sup.−1. For the same concentration, urolithin C stimulated glibenclamide-induced response (about 150%), EC.sub.50 value being in this case close to 5 μmol.Math.L.sup.−1.

Example 4: Effect of Urolithin A, B, C or D Determined in INS-1 Insulin Secreting β-Cells in the Presence of Exendin (FIGS. 4A-D)

(39) Exendin, a GLP-1 analog used for the treatment of diabetes, stimulates insulin secretion under high- but not low-glucose condition. Therefore, experiments were conducted in the presence of 8.3 mmol.Math.L.sup.−1 glucose and 0.001 μmol.Math.L.sup.−1 exendin.

(40) Material and methods were the same as described in example 1. For the determination of insulin secretion, cells were incubated for 1 h (5% CO.sub.2, 37° C.) in KRB medium containing 8.3 mmol.Math.L.sup.−1 glucose (glucose-stimulated condition) and exendin (0.001 μmol.Math.L.sup.−1) in the presence or absence of the compounds tested. Data were expressed as percent of exendin (0.001 μmol.Math.L.sup.−1)+glucose (8.3 mmol.Math.L.sup.−1)-stimulated insulin secretion (“control 100%”) (FIGS. 4A-D).

(41) Results.

(42) In the presence of exendin and glucose, the addition of urolithins A or B did not induce a significant raise of insulin secretion, as shown in FIGS. 4A and B. As opposed to urolithin A and B, the addition of urolithin C or D induced a significant concentration-dependent stimulation of insulin secretion, as illustrated in FIGS. 4C and 4D. Regarding maximal response, urolithin C appeared as the most active compound (about 185% stimulation as compared to exendin alone), while urolithin D induced a 140% stimulation.

(43) Regarding active concentrations, maximal effects were obtained in both cases at the maximal concentration tested (20 gmol.Math.L.sup.−1). EC.sub.50's from both compounds could be estimated between 5 and 10 gmol.Math.L.sup.−1.

Example 5: Evaluation of the Protecting Effects of Urolithins Against Oxidative Stress in INS-1 Insulin Secreting β-Cells (FIGS. 5A-D and 6)

(44) As previously stated, one of the mechanisms potentially involved in the prevention of diabetes could be the protection of β-cells from insults induced by inflammation and oxidative stress (Bonora, 2008).

(45) As oxidative stress is able to impair both β-cells insulin secretion capacity and viability (Kaneto et al., 1999), it is important to determine on insulin secretion and viability the effects of compounds potentially able to prevent or delay the evolution of diabetes.

(46) It has been previously shown (Youl et al., 2010) that quercetin, but not resveratrol nor the anti-oxidant N-acetyl cysteine, was able to prevent H.sub.2O.sub.2-induced insulin secretion impairment. The possible protective effect of urolithins A-D under was therefore examined under the same experimental conditions. Insulin secretion was determined as previously described by measuring insulin accumulation after a one-hour incubation period and viability determined from the MTT test, an indicator of mitochondrial function.

(47) Material and Methods:

(48) a—INS-1 Treatment

(49) For protection experiments, INS-1 cells were pre-incubated with the indicated concentrations of urolithins for one hour (5% CO.sub.2, 37° C.) in RPMI medium. After two washes in HEPES-balanced Krebs-Ringer bicarbonate buffer (123 mmol.Math.L.sup.−1 NaCl, 5.4 mmol.Math.L.sup.−1 KCl, 1.3 mmol.Math.L.sup.−1 KH.sub.2PO.sub.4, 2.7 mmol.Math.L.sup.−1 MgSO4, 2.9 mmol.Math.L.sup.−1 CaCl.sub.2, 5 mmol.Math.L.sup.−1 NaHCO.sub.3 and 20 mmol.Math.L.sup.−1 HEPES, pH 7.5) containing 2 g.Math.L.sup.−1 bovine serum albumin (KRB/BSA), INS-1 cells were incubated for one hour (5% CO.sub.2, 37° C.) in KRB/BSA containing urolithins at the indicated concentrations in the presence of 1.4 mmol.Math.L.sup.−1 glucose (basal condition) or 8.3 mmol.Math.L.sup.−1 glucose (stimulant condition). When used, H.sub.2O.sub.2 (50 gmol.Math.L.sup.−1) was added at the beginning of the one-hour incubation period. Control experiments were performed in the basal condition, in the absence of urolithins during the pre-incubation and incubation periods.

(50) b—Determination of Insulin Secretion

(51) Insulin secretion was determined as described above by measuring insulin accumulation after a one-hour incubation period.

(52) c—Determination of INS-1 Cells Viability:

(53) Cell viability was determined using the 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. At the end of the one-hour incubation period, cells were washed with KRB/BSA and KRB/BSA containing 5 mg/ml MTT was added to each well. Plates were then incubated for 3 h in the dark in a humidified atmosphere (5% CO.sub.2, 37° C.). Cells were washed with phosphate buffered saline (PBS) and precipitates were dissolved in 50 μl dimethyl sulfoxide (DMSO). Absorbance of the reduced intracellular formazan product was read at 492 nm on a microtiter plate reader (Tecan, Lyon, France). Experiments were performed in quadruplicate.

(54) Results:

(55) The effects of Urolithins A-D on insulin secretion in the presence of oxidative stress are illustrated on FIGS. 5A-D.

(56) As expected, H.sub.2O.sub.2 induced a major reduction of the degree of stimulation of insulin secretion by 8.3 mmol.Math.L.sup.−1 μlucose.

(57) All urolithins were able to prevent insulin secretion impairment induced by H.sub.2O.sub.2. However urolithins B, C and D were active for a concentration as low as 5 gmol.Math.L.sup.−1 while urolithin A was active for 20 gmol.Math.L.sup.−1 only.

(58) In addition, Uro B (10 gmol.Math.L.sup.−1) was able to induce a 150% stimulation, while Uro B and C (20 μmol.Math.L.sup.−1) were both able to induce a 200% insulin secretion stimulation as compared to control conditions.

(59) The effects of Urolithins A-D on cell viability are illustrated on FIG. 6.

(60) As expected, H.sub.2O.sub.2 induced a major reduction in cell viability as determined from the MTT test (mitochondrial function).

(61) As shown, Uro C and D were both able to totally prevent viability impairment while Uro B was partially active and Uro A had no effect.

(62) In summary, Uro B—and at a lower degree Uro C—possess a particular effect of insulin secretion protection, while Uro C and D—and at lower degree Uro B—are active in preventing impairment of viability.

(63) Due to structural similarities between urolithin C and isourolithin A (FIG. 1), a similar effect is expected for both urolithin C and isourolithin A.

Example 6: Effect of Urolithins A, B, C or D on Insulin Secretion Determined on a Physiologically Relevant Model, the Rat Isolated Pancreas Preparation (FIGS. 7A-B)

(64) The isolated pancreas preparation is a model reproducing the in vivo situation, as pancreas is surgically extract and perfused through its normal circulation. Therefore, islets of Langherans and insulin secreting β-cells are exposed to compounds modulating insulin secretion through the normal circulation (arteries, capillaries and veins). It is the most physiologically relevant and predictive model for the determination of the effects and physiologically active concentrations of compounds on insulin secretion.

(65) Material and Methods:

(66) Pancreas was isolated from male Wistar rats of 250-300 g body weight under pentobarbital anaesthesia (60 mg/kg i.p.). Previously described technique (Cadene et al., 1996) was used to isolate the pancreas from neighboring tissues. The organ was then transferred into a plastic chamber maintained at 37.5° C. Perfusion medium was Krebs-Ringer bicarbonate buffer containing 2 g/l bovine serum albumin and 5 mmol.Math.L.sup.−1 (non stimulant condition) or 8.3 mmol.Math.L.sup.−1 (stimulant condition) glucose, and continuously bubbled with a mixture of 95% O.sub.2/5% CO.sub.2. Infusion pressure was selected to provide a pancreatic outflow of 2.5 ml/min. The first sample was taken 30 min after initiation of perfusion to allow for an adaptation period. Two additional control samples were collected 10 and 15 min later, immediately followed by switching to the same buffer containing urolithins (10 μmol.Math.L.sup.−1). Pancreatic effluents were then collected at the following times (minutes): 17, 18, 19, 20, 21, 27, 32, 42. Pancreas was then washed using the same buffer in the absence of urolithin. Two additional samples were taken 5 and 15 min after urolithin withdrawal.

(67) All samples were collected for 1 min allowing determinations of pancreatic effluent output, and immediately frozen for insulin assay (Cisbio HTRF method). Insulin output rate (ng/min) was calculated by multiplying the hormone concentration (ng/ml) in the effluent by the corresponding flow rate (ml/min).

(68) Amplitude of stimulation of insulin secretion by urolithins was estimated by calculating the Area Under the Curve (AUC) of insulin produced above the basal level (trapezoidal rule). Before administration of urolithins, insulin secretion levels were 3.39±0.58 ng/min (n=14 pancreases) and 67.5±7.80 ng/min (n=15 pancreases), for 5 mmol.Math.L.sup.−1 and 8.3 mmol.Math.L.sup.−1 glucose, respectively.

(69) Results:

(70) Urolithins (10 μM) were all able to induce insulin secretion (change in basal insulin secretion rate), with some differences in their potencies, though. The corresponding AUC (ng insulin secreted for 32 min) were as follows: urolithin A: 2609.5 urolithin B: 142.0 urolithin C: 9160.0 urolithin D: 387.5

(71) In accordance with insulin secretion data obtained in insulin secreting INS-1 β-cells, urolithin C appeared as the most active compound, followed by urolithins A, D and B.

(72) FIG. 7A-B illustrates the insulin secreting response to urolithin C obtained either under 8.3 mmol.Math.L.sup.−1 stimulating glucose condition (7A) or 5 mmol.Math.L.sup.−1 non-stimulating glucose condition (7B). As illustrated, urolithin C induced a major amplification under stimulating glucose condition and had no effect under non-stimulating glucose condition.

(73) In addition, we found that urolithin C was able to induce some stimulation of insulin secretion for a concentration as low as 1 μM (not illustrated) in stimulating glucose condition but not in non-stimulating glucose condition.

(74) These results validate on a physiologically relevant model reproducing the in vivo situation the potential use of urolithins as glucose-dependent insulin-secreting compounds. They also suggest that insulin-secreting activity will not occur under normo-glycemic conditions in vivo, reducing the risk of hypoglycemia, a common side-effect of insulin secretion stimulants (e.g. sulfonylureas) that stimulate insulin secretion even under normo-glycemic conditions.

(75) This is an important feature for the development of urolithins for the prevention of the treatment of type 2 diabetes.

CONCLUSION

(76) Insulin secretion regulatory effects of urolithins were first studied on INS-1 insulin-secreting β-cells under low-glucose or glucose-, glibenclamide- and exendin-stimulated insulin secretion conditions.

(77) Urolithins stimulated insulin secretion in various experimental conditions at the maximal concentration tested (20 μmol.Math.L.sup.−1), with the notable exceptions of: urolithin A, which did not stimulate insulin secretion under low-glucose condition. urolithins A and B, which did not amplify exendin-stimulated insulin secretion.

(78) These exceptions may indicate some different cellular mechanism between urolithins.

(79) Noticeably, the degree of stimulation of insulin secretion by urolithins appeared similar between the different insulin secretion conditions: about 160% for urolithin A, 140% for urolithin B, 185-200% for urolithin C and 135-155% for urolithin D, suggesting that urolithin C was the most active compound (C>A>B≈D)

(80) Urolithin-induced stimulation does occur either under high or low level of insulin secretion conditions and in the presence of glucose and/or insulin secretion stimulating agents, suggesting that urolithins act as amplifiers of insulin secretion stimulants.

(81) The potential of urolithins to protect β-cells against oxidative stress were also studied on the INS-1 β-cells both on insulin secretion and viability, knowing that quercetin—but not resveratrol or the antioxidant N-acetyl cystein—was previously shown to fully prevent viability and insulin secretion impaired by oxidative stress (Youl et al., 2010). Results (FIGS. 5 and 6) indicate that the various urolithins do possess differential capacities to protect cells against the oxidative stress-induced impairments of insulin secretion (notably urolithin B and C) or viability (notably urolithins C and D).

(82) The effects of urolithins on insulin secretion were also determined on a physiologically relevant model, the rat isolated perfused pancreas.

(83) On this model, urolithins (10 μM) were able to amplify insulin secretion in a 8.3 mmol.Math.L.sup.−1 stimulating glucose condition, but not under a 5 mmol.Math.L.sup.−1 non-stimulating glucose condition. Order of potencies of urolithins to amplify insulin secretion in this model appeared as C>>A>D>B, in good agreement with data obtained on INS-1 β-cells. Again, urolithin C appeared as the most active compound and some response was obtained for a concentration as low as 1 μM. Also, our results suggest that the effects of urolithins are glucose-dependent and, as opposed to sulfonylureas, may not induce hypoglycemia when exposed to normo-glycemic conditions.

(84) In summary, our data suggest that urolithins act as amplifiers of physiological (glucose)- or drug-induced insulin secretion, their EC.sub.50's being around 5 to 10 μmol.Math.L.sup.−1.

(85) In summary, urolithin C seems to be the most active compound as a glucose-dependent stimulant of insulin secretion, although other urolithins may display additional protective effects on β-cells.

BIBLIOGRAPHIC REFERENCES

(86) Bardy et al., Br J Pharmacol. 2013, 169:1102-13. Bialonska et al., J. Agric. Food Chem. 2009, 57: 10181-10186. Bonora, Nutr. Metab. Cardiovasc. Dis. 2008, 18: 74-83 Cadene A et al., Eur J Pharmacol. 1996, 318:145-51. Cerda et al., Eur J Nutr. 2004, 43: 205-20. Cerda et al., J Agric Food Chem. 2005, 53:5571-6. Del Rio et al., Antioxid Redox Signal. 2013, 18: 1818-1892. Kaneto et al., Diabetes. 1999, 48: 2398-406. Garber, Diabetes, Obesity and Metabolism. 2012, 14: 399-408. Gonzilez-Sarrias et al., J Agric Food Chem. 2013, 61:4352-9. Li et al., Biochem Soc Trans 2008, 36: 930-934 Lucas et al., Carbohydr. Res. 2009, 344:1340-1346. Palsamy and Subramanian, Biomed Pharmacother. 2008, 62: 598-605. Pinto et al., J Med Food. 2010, 13:1027-35. Poitout and Robertson, Endocr Rev 2008, 29: 351-366. Tulipani et al., J Agric Food Chem. 2012, 60: 8930-40. Youl et al., Br J Pharmacol. 2010, 161: 799-814.