Epoxypukalide Induces Proliferation and Protects against Cytokine-Mediated Apoptosis in Primary Cultures of Pancreatic β-Cells

There is an urgency to find new treatments for the devastating epidemic of diabetes. Pancreatic β-cells viability and function are impaired in the two most common forms of diabetes, type 1 and type 2. Regeneration of pancreatic β-cells has been proposed as a potential therapy for diabetes. In a preliminary study, we screened a collection of marine products for β-cell proliferation. One unique compound (epoxypukalide) showed capability to induce β-cell replication in the cell line INS1 832/13 and in primary rat cell cultures. Epoxypukalide was used to study β-cell proliferation by [3H]thymidine incorporation and BrdU incorporation followed by BrdU/insulin staining in primary cultures of rat islets. AKT and ERK1/2 signalling pathways were analyzed. Cell cycle activators, cyclin D2 and cyclin E, were detected by western-blot. Apoptosis was studied by TUNEL and cleaved caspase 3. β-cell function was measured by glucose-stimulated insulin secretion. Epoxypukalide induced 2.5-fold increase in β-cell proliferation; this effect was mediated by activation of ERK1/2 signalling pathway and upregulation of the cell cycle activators, cyclin D2 and cyclin E. Interestingly, epoxypukalide showed protection from basal (40% lower versus control) and cytokine-induced apoptosis (80% lower versus control). Finally, epoxypukalide did not impair β-cell function when measured by glucose-stimulated insulin secretion. In conclusion, epoxypukalide induces β-cell proliferation and protects against basal and cytokine-mediated β-cell death in primary cultures of rat islets. These findings may be translated into new treatments for diabetes.


Introduction
Diabetes is one of the most devastating diseases of our era. The most common forms are type 1 and type 2 diabetes. b-cell mass decreases approximately by 70-100% in type 1 diabetes and up to 65% in type 2 diabetes. Both forms of the disease display increased b-cell apoptosis [1,2].
Finding therapeutic targets and molecules that preserves and/or enhances functional beta-cell mass is essential in the long-term treatment of diabetes. It is known that adult pancreatic b-cells, in rodents and humans, are generated from the proliferation of differentiated beta-cells [3,4]. Some growth factors have been shown to induce b-cell proliferation in vitro and in vivo, among them HGF (hepatocyte growth factor), PL (placental lactogen) and PTHrP (Parathyroid hormone related protein) have also shown enhancement of b-cell survival [5].
Classical signalling pathways involved in beta-cell proliferation comprise PI3K/AKT, ERK1/2, PKC and JAK2/STA5 [5]. Activation of these intracellular effectors ends on the regulation of the proteins that control G1/S interphase of the cell cycle. Among all the proteins that regulate this interphase, pRb is the key checkpoint gatekeeper. The complexes that regulate pRb phos-phorylation (and its inactivation to allow cell cycle progression) are cdk4/cyclin D and cdk2/cyclinE [6]. Both, cyclin D2 and cyclin E have been found to have a relevant role in several models of b-cell proliferation [5,7,8,9,10,11,12,13], being essential to induce b-cell replication.
Epoxypukalide was reported by Schmitzs group in 1984 by the first time [14], they purified it from the gorgonian Leptogorgia setacea. Epoxypukalide and other furanocembranolides were identified in 2007 by Dorta et al. [15] who searched for marine natural products in benthic species around the Isthmus of Panama. Finally Grote et al. [16] described a group of furanocembranoids isolated from the soft coral Sinularia asterolobata, being one of them epoxypukalide. They studied the cytotoxic and anti-proliferative effect of those compounds, reporting that epoxypukalide did not inhibit HeLa (human epithelial carcinoma) cell proliferation nor produces cytotoxicity in the cell lines L-929 (murine fibrosarcoma) or K-562 (human erythroleukemic cell line) [16].
In the current report, we demonstrate a proliferative effect of epoxypukalide on primary cultures of rat b-cells, in parallel to a protective effect on b-cell death while preserving b-cell function. Epoxypukalide-induced b-cell proliferation is mediated by ERK1/ 2 activation and targets cyclin D2 and cyclin E. Taken together these data support a potential use of epoxypukalide in the treatment of diabetes.
Rat islets were isolated from 2 months old male Wistar rats provided by Animal Production and Experimentation Service (SEPA, University of Cádiz) by a standard procedure and incubated at 37uC and 5% CO2 in a humidified atmosphere. Culture medium was RPMI-1640 with 2 mmol/l L-glutamine supplemented with 5.5 mmol/l D-glucose, 10% fetal bovine serum, 100 U/ml penicillin, and 100 mg/ml streptomycin.

Proliferation in INS-1 Cells
For the first screening, INS-1 832/13 cells were seeded at a density of 20,000 cells per well in 96-well plates. Natural products were assayed at a final concentration of 0.1 mM in culture medium with 5.5 mM glucose. Proliferation was measured after 24 hours by BrdU kit (Roche, Germany), following manufacturer's instructions.

Proliferation in Rat Islets
Rat islets were isolated and incubated in serum free medium with 0, 0.01, 0.1 and 1 mM epoxypukalide after an overnight recovery. 24 hours later, 100 IEq (islet equivalent) per well were seeded in 24-well plates with fresh medium without FBS containing [ 3 H]thymidine (1 mCi/well). [ 3 H]thymidine incorporation was measured 24 h later; radioactivity was corrected for protein levels measured by the BCA kit (Thermo Fisher, USA). Results are expressed as percentage of counts per minute incorporated per microgram of protein in control cells (100%).
To determine specific b-cell proliferation, isolated rat islets were treated with medium without serum containing 0.1 mM epoxypukalide after an overnight recovery. 24 hours later, islets were serum depleted and incubated with 1 ml/ml BrdU (Sigma-Aldrich, USA). Islets were then incubated for 24 hours. Afterwards islets were fixed with Bouins Solution (Sigma-Aldrich, USA) for 1 hour and then with formalin until treated for paraffin blocks. 5 mm sections were stained with rat anti-BrdU antibody (Abcam, UK) and with guinea pig anti-Insulin antibody (Invitrogen, USA). Fluorescence images of the sections were acquired using a Olympus BX40 fluorescence axial microscope. The BrdU-positive nuclei of beta cells and the total nuclei of beta cells were counted manually with the assistance of the ImageJ software. At least 1000 insulin-positive cells for each preparation were counted.

Western-blot
Islets used to study signalling pathways activation (AKT and ERK1/2) were treated for 0, 5, 15 or 30 minutes with 0.1 mM epoxypulide after overnight recovery from islet isolation.
Islets used to study cell cycle proteins expression were treated for 24 h with 0.1 mM epoxypukalide after overnight recovery from islet isolation.

Beta-cell Death
5 mm sections obtained as explained above were used for TUNEL-staining. DeadEnd Fluorometric Tunel System Kit (Promega, USA) was used following manufacturer's instructions. TUNEL-positive/beta-cells were counted and represented as a percentage of total number of beta-cells. Nuclei of beta cells were counted manually with the assistance of the ImageJ software. At least 1000 TUNEL-positive cells for each preparation were counted.

Glucose-stimulated Insulin Secretion and Insulin Content
Secretion assays was performed on treated rat islets and their controls. Islets were plated on cell culture inserts onto 24-well plates at a density of 20 IEq groups in HEPES balanced salt solution (HBSS) (114 mmol/l NaCl, 4.7 mmol/l KCl, 1.2 mmol/l KH2PO4, 1.16 mmol/l MgSO4, 20 mmol/l HEPES, 2.5 mmol/l CaCl2, 25.5 mmol/l NaHCO3, and 0.2% bovine serum albumin [essentially fatty acid free], pH 7.2). Islets were washed twice in 1 ml HBSS with 2.2 mmol/l glucose followed by 10 minutes of preincubation in 2 ml of the same buffer. Insulin secretion was stimulated by using static incubation for 30 minutes in 1 ml of the same buffer, followed by 30 minutes of incubation in HBSS containing 22 mmol/l glucose. Secretion samples were measured by ultra-sensitive rat insulin ELISA (Alpco Diagnostics, USA).
20 IEq by triplicate of each condition (control, 0.01, 0.1 and 1 mM epoxypukalide) were harvested after 24 h treatment. Insulin was extracted using 1 mL of acid-ethanol (70% ethanol and 0.02% HCl). Insulin was measured by rat insulin ELISA.

Statistics
Statistical analyses of data were performed by Student-t test when two conditions were compared and ANOVA when more than two conditions were compared. Data were expressed as mean 6 SD. P values ,0.05 were considered significant.

Screening of a Collection of Marine Natural Products
Seven furanocembranolides (Figure 1) obtained from the soft coral Leptogorgia spp. [15] were screened for induction of b-cell proliferation in the rat cell line INS-1 832/13. This group of compounds was chosen from a larger collection of marine natural products, after a preliminary screening, being the unique group of compounds inducing beta-cell proliferation of the INS1 832/13 cells (data not shown). Cell proliferation was measured using BrdU incorporation. Four products, epoxypukalide (1), pukalide (2), (Z)deoxypukalide (3) and leptolide (4), induced b-cell proliferation over the threshold ($1.5-fold compared to control) ( Figure 1).

Epoxypukalyde Induces b-cell Proliferation in Primary Rat Islets
We used isolated rat islets to test ex-vivo the proliferative capability of compounds 1-4. Proliferation was measured using [ 3 H]thymidine incorporation. Epoxypukalide (1) showed an increase of 30% in proliferation compared to control, conversely pukalide (2), leptolide (4) and (Z)-deoxypukalide (3) did not induce proliferation in primary cell cultures (Figure 2A). Then, we performed a dose response evaluation of epoxypukalide to test cell proliferation, using concentrations from 0.01-1 mM. Only 0.1 mM showed a significant increase in cell proliferation ( Figure 2B). Thus, for following experiments involving beta-cell proliferation we have used 0.1 mM epoxypukalide. As a second approach and in order to measure specific b-cell proliferation, we used BrdU incorporation followed by insulin/BrdU staining to quantify BrdU+insulin+ cells in respect to total insulin+ cells. Epoxypukalide treatment showed 250% increased b-cell proliferation compared to control cells ( Figure 2C, D).

Activation of Signalling Pathways by Epoxypukalyde in Primary Rat Islets
The analysis by western-blot of two of the most relevant signalling pathways involved in b-cell proliferation indicated that epoxypukalide did not activate AKT pathway ( Figure 3A, B), but ERK1/2 was activated within the first 30 minutes after treatment ( Figure 3C, D). To determine whether ERK1/2 could mediate the proliferative effect of epoxypukalide, a specific pharmacological inhibitor of ERK1/2 (PD98059) was used and its effect on epxypukalide-b-cell proliferation was examined using [ 3 H]thymidine incorporation. PD98059 decreased the proliferation induced by epoxypukalide to basal levels ( Figure 3E), showing that this pathway is involved in epoxypukalide-induced b-cell proliferation.

Epoxypukalide Induces Expression of Cell Cycle Activators
To further investigate the mechanisms by which epoxypkalide induces b-cell proliferation, we performed western-blots for the proteins involved in pRb phosphorylation in G1/S (cyclin D1, cyclin D2, cyclin D3, cdk4, cyclin E and cdk2) in rat islets in the presence or absence of epoxypukalide. Of the six tested proteins, only cyclin D2 and cyclin E were up-regulated, showing ,15%   more expression in epoxypukalide-treated islets compared to control islets ( Figure 4A, B).

Epoxypukalide Protects Basal and Cytokine-induced bcell Death
To investigate the effects of epoxypukalide on b-cell survival, we performed a dose response evaluation of epoxypukalide, using concentrations from 0.01-1 mM. We tested cell death using caspase 3 activation detected by western-blot. Only 0.1 mM showed a significant decrease in cell death ( Figure 5A, B), thus, for following experiments involving beta-cell death we have used 0.1 mM. To confirm epoxypukalide effect on beta-cell death we performed TUNEL/insulin staining on sections of rat islets that were previously exposed to epoxypukalide and their untreated control. After quantification of TUNEL+insulin+ cells, we detected 40% lower basal b-cell death in epoxypukalide treated islets than in control islets ( Figure 5C, D). To test epoxypukalideprotective effect in conditions of stress for the pancreatic islets, we treated rat islets with cytokines (CK), in the presence or absence of epoxypukalide, and we determined activation of the apoptotic pathway using caspase-3 activation. Islets treated with CK showed 80% increase in cleaved-caspase 3 compared to control, this effect was reverted when islets were treated with 0.1 mM epoxypukalide ( Figure 5E, F).

Epoxypukalyde does not Impair b-cell Function
Finally, we tested whether epoxypukalide alters b-cell function. To this end we performed glucose-stimulated insulin secretion on epoxypukalide-treated rat islets to check for b-cell function in vitro. Insulin secretion after glucose overload was completely normal ( Figure 6A). Furthermore we have measured insulin content in control islets and epoxypukalide-treated islets, finding that there are not significant differences between epoxypukalidetreated and control islets ( Figure 6B). Both results support that there is not deleterious effect of epoxypukalide on b-cell function.

Discussion
Search for new molecules to preserve functional b-cell mass is important for diabetes. Research groups and pharmaceutical companies are involved in the chase of new molecules that can induce b-cell proliferation and/or enhance b-cell survival; their searches are based on the detection of activation of signalling pathways involved in b-cell death or b-cell proliferation [17,18]. The interaction between chemists that purify and identify new compounds and biologists who test their function is essential for the discovery of new treatments.
Accordingly to the structures of compounds 1-4 and the proliferation results, it can be observed that although they possess the same carbon skeleton, differences in their functionalizations are critical to induce proliferation in primary cell cultures. From comparison of epoxypukalide (1) and leptolide (4) it can be deduced that the presence of a methyl ester instead of the alhdeyde group at C-18 is a key to the potency of furanocembranolides. Also from the observation of epoxypukalide (1), pukalide (2) and (Z)deoxypukalide (3) it can be deduced that epoxydation, especially at C-112C-12, is crucial to show proliferative activity in primary cell cultures.
From the dose-dependence studies we have learned that 0.1 mM is the optimal concentration to induce beta-cell proliferation and to inhibit beta-cell death. This should be the reference concentration to design in vivo experiments in rodents.
Two of the most relevant pathways involved in b-cell proliferation (AKT and ERK1/2) were studied. Analysis of these two pathways indicated that epoxypukalide did activate ERK1/2 pathway. Some growth factors have shown to induce b-cell proliferation through ERK1/2 signalling including HGF [19], insulin and IGF1 (insulin growth factor-1) [5].
It has been shown that G1/S interphase is essential in the regulation of b-cell cycle proliferation. Cyclin D2 and E, individually, have been shown previously to induce b-cell replication under the stimulus of several mitogens including high glucose, GH (growth hormone), PRL (prolactin) and PTHrP [5,7,8,9,10,11,12,13]. Cyclin D2 joins to cdk4 and cyclin E to cdk2 maintaining pRb phosphorylation in different residues. We can argue that since both cyclins are upregulated by epoxypukalide, it induces a synergistic effect responsible to maintain beta-cell proliferation, even though their expression levels increase modestly.
CK (cytokine)-induced b-cell apoptosis is a hallmark of diabetes [1,20,21,22]. Our experiments showed epoxypukalide protects bcells from basal and CK-induced apoptosis. Other small molecules have shown similar effects on protection of CK-induced b-cell apoptosis [18], but their proliferative capability was not tested. It is important to expand b-cell mass in order to have an efficient control of glucose homeostasis in diabetes. In support of this notion, Zhao et al. showed that diabetic mice (leptin receptor nulldb/db) treated with a sphingosine 1-phosphate receptor modulator (FTY720) counteracted hyperglycemia by induction of b-cell proliferation and b-cell mass expansion [23].
Increased b-cell proliferation may involve loss of differentiation and cell function failure. Thus, we performed glucose-stimulated insulin secretion on epoxypukalide-treated rat islets to check for bcell function in vitro. Insulin production and secretion after glucose overload was completely normal, meaning that there is not deleterious effect of epoxypukalide on b-cell function. This is important in order of a possible use of this compound in the treatment of diabetes. In this direction, Grote et al. showed that epoxypukalide was not cytotoxic for L-929 (murine fibrosarcoma) or K-562 (human erythroleukemic cell line) [16]. Their results sustain that epoxypukalide has not adverse effects on other cell types.
In conclusion, our study supports that epoxypukalide is a potential drug for the treatment of diabetes, further work in granted in order to study its in vivo effects.