AMPK is indicated as cellular energy sensor by switching AMP/ATP status. It is comprised of three subunits, catalytic α (α1or α2) subunit, regulatory β (β1, β2 or β3) and γ (γ1, γ2 or γ3) subunits (Hardie and Carling, 1997, Hardie et al., 1998). With various splicing, 12
10
heterotrmeric combinations mainly distribute in particular tissues. For instance, α1 subunit is abundant in the cytoplasm of kidney pancreatic β-cell, the lung and the adipose tissue, while α2 subunit is in the heart and skeletal muscles; most of β1 subunit is in the liver and β2 in the skeletal muscle (Birk and Wojtaszewski, 2006, Treebak et al., 2007). The
N-terminal of α subunit contains serine/threonine protein kinase domain which can be phosphorylated on threonine172 within the activation loop (T-loop) by upstream kinases so that AMPK can be activated (Hardie et al., 2003), while γ subunit is the main site for AMP or ATP by binding on its specific four cystathionine-β synthase (CBS) domains (Xiao et al., 2007). AMP binding on AMPK would promote allosteric activation which is predominantly phosphorylated by LKB1/STRAD/MO25 complex, whereas recent studies indicated that another alternative
CaMKK β was capable to activate AMPK via increasing calcium without raising amount of AMP (Woods et al., 2003, Momcilovic et al., 2006, Towler and Hardie, 2007). Several environmental stresses, such as hypoxia and hyperglycemia, are capable to enhance the non-metabolic activation of AMPK.
1.5
1.6 AMPK activator Nstpbp168
Since AMPK involves in multiple cellular and physiological
interactions, its activators are applied in the therapy of diabetes or other metabolic diseases. One common anti-diabetic drug is metformin which promote AMPK activity through binding to the γ unit of AMPK.
Metformin can’t induce LKB1 phosphorylation and fails to increase calcium influx in skeletal muscles, even in pancreatic β-cells, but the effects is reverse in its main target organ, liver. The other pharmaceutical compound, AICAR can be metabolized to AMP analog ZMP
(5-aminoimidazole-4-carboxamide-1-β-D-ribofuranotide) to activate AMPK by interacting with γ subunit(Corton et al., 1995). Similarly, ZMP can inhibit glucogenesis via blocking fructose-1,6-biphosphatase in
heptocytes, whereas it could increase glucose uptake in skeletal muscle.
Therefore, it was concerned as a potent drug for increasing insulin sensitivity in peripheral tissues. However, it can also interact with other enzymes, such as glycogen phosphorylase, meaning that it lack a
specificity for AMPK(Gruzman et al., 2009).
Traditional Chinese herbs have been found from many potential resources for the production of drugs to treat diabetes. Nstpbp168, a pure compound purified from methanol extracts of a fern plant Hypolepis punctata (Thunb.) Mett. Its chemical structure was identified as a small molecular weight natural product, pterosin A. Previous studies found that streptozotocin-induced diabetic mice (Type-1 diabetes) treated with Nstpbp168 could against glucose intolerance. In differentiated C2C12
cell
stress/lipotoxicity caused by H2O2/palmitate. We hope this study can provide more information of Nstbp168 in regards of its protective effects on pancreatic β-cell/liver cell in the development of diabetes.
16
Chapter 2. Materials and Methods
2.1 Reagents
The materials/reagents were purchased from companies indicated in the parentheses. RPMI-1640 and Dulbecco’s modified Eagle’s medium (DMEM) (Gibco-BRL-Life Technologies, Grand Island, NY, USA); Fetal bovine serum (Thermo Scientific, South Logan, Utah, USA);
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT), hydrogen peroxide (H2O2), nitrotetrazolium blue chloride (NBT) and 2′,7′-Dichlorofluorescin diacetate (DCFH-DA) (Sigma Chemical Co, St.
Louis, MO, USA);
6-[4-[2-(1-Piperidinyl)ethoxy]phenyl]-3-(4-pyridinyl)-pyrazolo[1,5-a]pyr imidine dihydrochloride (Compound C) (Tocris Bioscience, Bristal, UK);
Anti-phospho AMPK alpha (Thr172) antibody (Cell Signaling
Technology); Horseradish peroxidase anti-rabbit secondary antibody (Jackson, West Grove, PA, USA).Nstpbp168 (provided from Dr.
Feng-Lin Hsu, Taipei Medical University, Taipei, Taiwan)
2.2 Cell Culture
Rat pancreatic insulin secreting cell line, RINm5F, were obtained from the American Type Culture Collection (Rockville, MD). RINm5F were maintained in RPMI-1640 medium containing 10% fetal bovine serum, 1mM sodium pyruvate, 2 mM glutamine, 11.1 mM glucose, 24 mM NaHCO3, 2.5 g/l HEPES, 10,000 units/l of penicillin and 100 mg/l of streptomycin at 37°C in an incubator with a humidified atmosphere of 5%
CO2. The growth rate of RINm5F was slow that the suspension would be able to attach onto culture dishes for 3 days after passage and it was necessary to maintain for additional 3 days to reach the confluence.
Therefore, when performing pharmaceutical treatment, it is batter to let cell grow for at least 4 days so that the cell numbers could be enough to be applied with various concentrations of drugs.
Human hepatocellular liver carcinoma cell line, HepG2 cells were purchased from American Type Culture Collection (Rockville, MD) and grown in DMEM supplemented with 10% fetal bovine serum, 100 units/mL of penicillin, 100 lg/mL of streptomycin at 37°C under a humidified, 5% CO2 atmosphere.
2.3 Palmitate/BSA complex preparation
A stock palmitate solution was dissolved in absolute alcohol and 1%
FFA-free BSA solutions were prepared in serum free medium. To prepare 250 μM palmitate /1% BSA stock solutions, 10 μl of 25 mM palmitate stock solution was added to 990 μl of 1% FFA-free BSA solution and then mixed by vortexing for 30 seconds followed by additional 1~2 hours of incubation at room temperature . All control conditions contained a solution of vehicle (absolute alcohol) mixed with FFA-free BSA as the abovementioned ratio. This mixture was so called palmitate/BSA solution.
2.4 Cell viability assay
Cell viability was measured by MTT assay. MTT was yellow
18
tetrazole which could be reduced to insoluble purple formazon dye
crystals by mitochondrial succinate dehydrogenase in living cells. Briefly, the RINm5F cell was seeded in a 24-well plat at a density of 2×105
cell/0.5 ml. After adhering for three days, cells began to expose to
different treatment for variant time. Then, 50 μl MTT solution (1 mg/ml in phosphate-buffered saline (PBS)) was added to each well. The plate was further incubated for another 2 hours at 37 °C. The supernatant was removed and then 200 μl dimethyl sulphoxide (DMSO) was added to thoroughly dissolve the dark blue crystal formazan. The absorbance of the royal purple solution at 570 nm was detected with a
spectrophotometer.
2.5 Western blotting
Cells were seeded in a 6-well plate at a density of 5×105 cells/2 ml and incubated for 6 days w/o drug treatment when the cell confluence was reaching 70-80%. After cells were harvested, 30μl of lysis buffer (20 mM Tris base, pH 7.9, 20 mM NaCl, 1 mM EDTA, 5 mM EGTA, 20 mM β-glycerolphosphate, 1 mM DTT, 1 mM PMSF in isopropanol, protease inhibitor, 25nM Calyculin A, 05.% TritonX-100) was added. Lysates were centrifuged at 13,000 rpm for 15 minutes. Aliquots of samples containing equal amounts of protein, measured by Bradford assay (Bio-Rad,
Hercules, CA, USA), was resolved by SDS-PAGE for separation and transferred to PVDF membrane for immunoblotting. The membrane were blocked with TBST including 5% skim milk at room temperature for 1 h and then incubated with primary antibodies for AMPK alpha, phospho
AMPK alpha(Thr172) in TBST at 4℃overnight. After washing with TBST, blots were visualized using horseradish peroxidase-conjugated secondary antibody followed by ECL detection (Millipore Corporation, Billerica, MA, USA).
2.6 Determination of reactive oxygen species (ROS)
The generation of ROS was examined by NBT assay. NBT is membrane permeable yellow-colored nitroblue tetrazolium (Choi et al., 2006). After absorbed by cells, NBT would interact with O2- to be reduced to blue formazon. Cells were seeded in a 24-well plate at a density of 2×105 cell/0.5 ml and were grown for three days. Cells were begun to expose to various drugs under variant time. After treatment, the plate was further incubated for 90 minutes at incubator. The supernatant was withdrawn and the pellet was washed with PBS one more time. The dark blue crystal formazan was solved in 400 μl dimethylsulphoxide (DMSO) and 600μl KOH (1N). The absorbance of the bluish green solution at 630 nm was detected with a spectrophotometer.
2.7 Evaluation of ROS production by DCFH-DA
A fluorometric assay, 2′, 7′-Dichlorofluorescin diacetate
(DCFH-DA) is performed to detected intracellular oxidation. DCFH-DA crosses the cell membrane and then undergoes deacetylation by cellular esterases to non-fluorescent DCFH, which is quickly oxidized to highly fluorescent DCF by ROS (Rosenkranz et al., 1992). Cells were seeded on a cover-slip-loaded 12-well plate at a density of 3×105 cell/ ml. After
20
adhering for three days, cells began to expose to different treatment for variant time. The cells were added with 10 μM DCFH-DA in no serum medium at incubator for 20 min. After washing the cells with PBS twice, digital images of DCF fluorescence were observed by a fluorescence microscope system at an excitation wavelength of 488 nm (argon laser) and a 515-nm long-pass emission filter.
2.8 Glucose stimulated insulin secretion (GSIS)
Isolated islets were incubated w/o Nstpbp168 in RPMI-1640 without glucose for 2 hours. Islets were washed once in RPMI-1640 without glucose and washed twice in RPMI-1640 containing 2.8 mM glucose, and then they were pre-incubated in the same medium for 30 min at 37°C.
This buffer was then replaced with RPMI-1648 containing 2.8 mM or 16.7 mM glucose as indicated for a further 60 min at 37°C. Supernatants was then collected for analysis of insulin secretion by ELISA.
2.9 Statistical Analysis
Data are presented as mean ± SEM. Treatment effects were evaluated by using a two-tailed Student’s t test. A p value < 0.05 was considered to be statistically significant.
Chapter 3. Results
3.1 The effects of Nstpbp168 on cell survival
In previous study, we have found that Nstpbp168 could enhance fatty acid oxidation and glycolysis in animal model. Even more, it could reduce pancreatic cell hyperplasia in diabetic mouse model. This implied that Nstpbp168 may play a protective role in β-cell dysfunction during the progression of type 2 diabetes. Thereby, we intended to investigate the toxicity of Nstpbp168 on RINm5F by utilizing cell viability assay at first.
After treated with Nstpbp168, 1 μM, 10 μM, 30 μM, 100 μM and DMSO for 18 hours, cells were applied to MTT assay. As shown in Fig 1A, Nstpbp168 did not display any cellular toxicity even up to 100 μM. Thus, the concentrations from 0 to 100 μM of Nstpbp168 would be further employed in the present study.
H2O2 was indicated as an ideal stimulator for oxidative stress instead of β-cell-specific toxins, such as alloxan and streptozotocin because H2O2 is the prevailing production in most oxidative stress processes
(Szkudelski, 2001). For determining the appropriate dose of H2O2, we exerted the prior study to test the dose-dependent viability losses caused by H2O2 on RINm5F cells. After treated with different doses of H2O2 (25–
100 μM) for 2 hours, cell survival was estimated by MTT assay. As indicated in Fig 1B, H2O2 dose-dependently induced cell death, with a maximum effect at 100 μM. Since 40 μM H2O2 caused 50% cell impairment, it was identified as efficient dose in the following experiments.
22
3.2 Scavenging capacity of Nstpbp168 on H2O2-induced ROS In advance, cells treated with various concentrations of Nstpbp168 in the absence or presence of 40 μM H2O2 were measured by MTT assay as indicated in Fig 1C. The survival rate of Nstpbp168 combined with 40 μM H2O2 was elevated in a little tendency compared to H2O2 treatment, implied that Nstpbp168 may have mild protective effects on
H2O2-induced cell death, just by sustaining but not promoting cell viability. Therefore, we tried to raise Nstpbp168 up to 200 and 600 μM.
Interestingly, efficient increasing survival rate was obtained in this range of higher concentration. Although the protective effect of 100 μM
Nstpbp168 was mild in MTT assessment, we are still interested in
analyzing the change of cellular ROS under these conditions. Intracellular ROS levels in H2O2-treated RINm5f β-cell were determined by using NBT assay which was shown in Fig 2. The ROS scavenging capacity of Nstpbp168 was presented in a dose dependent manner, especially at 100 μM. Hence, these data provided evidence that Nstpbp168 might exert as an anti-oxidant drug to reduce oxidative stress- induced cell death in β-cell.
3.3 The anti-oxidant protective ability of Nstpbp168 is through activating AMPK
In our previous research, we found that Nstpbp168 was a novel AMPK activator. Some reports had mentioned that phosphorylated
AMPK would suppress nitric oxide induced apoptosis on β-cell (Nyblom et al., 2008). Therefore, in order to determine whether the cytoprotection
of Nstpbp168 was due to activate AMPK, RINm5F cells were pretreated with the AMPK inhibitor, Compound C for 2 hours before applied with Nstpbp168 or H2O2 as presented in Fig 3C. Consequently, 20 μM
Compound C attenuated the protective effect of Nstpbp168 on
H2O2-induced oxidative stress, implying that the cytoprotective effect of Nstpbp168 might be partly mediated through AMPK activation. In order to further clarify the association, we tried to detect the AMPK
phosphorylated level under different doses of Nstpbp168 in a short-term or long-term manner. Regardless short-term or long-term treatment, the pattern was similar as shown in Fig 3A. Phosphorylated AMPK was apparently increased under 75~100 μM Nstpbp168 treatment for 3 or18 hours, revealing that Nstpbp168 could activate AMPK in a short time incubation and the effect could prolong for an extend time scale. Because 100 μM Nstpbp168 performed well in activating AMPK no matter in 3hr or 18hr treatment, we subsequently chose this concentration to investigate its influence on H2O2-induced damage. Markedly, as indicated in Fig 3B, additional Nstpbp168 introduced into cells with 40 μM H2O2 pretreatment could promote higher AMPK activation than that of treated with
Nstpbp168 or H2O2 alone. Owing to the protective role of 100 μM
Nstpbp168 in preventing H2O2-inducing cell death (Fig 1C), preliminary data might also provide the insight that Nstpbp168 could regulate cellular physiological status through activating AMPK, even under oxidative stress.
24
3.4 Nstpbp168 prevented Palmitate-induced lipotoxicity
From the protective effects of Nstpbp168 under H2O2 -induced oxidative stress on pancreatic β cells, it is suggested that Nstpbp168 might behave as an anti-oxidant drug as well as an activator of AMPK.
However, H2O2 is the production of oxidative stress which may be resulted from excess high glucose or lipid intake. Therefore, we were interested in studying how Nstpbp168 regulated cellular physiology or signal transduction under the stress of fatty acid overload.
Here, we treated cells with various doses of palmitate to create an environment with over-loaded lipid. As indicated in Fig. 4A, the MTT assay revealed that palmitate treatment for 24 hours would cause cell death significantly in a dose-dependent manner from 100 to 500 μM and 250 μM palmitate led 50% leaithality efficiently. Therefore, we chose 250 μM palmitate for further studying its lipotoxic effects on pancreatic/liver cells. Notably, in Fig 4B and 4C, the MTT results showed that the
cytotoxic effect of palmitate treatment would be reversed by
administrated with Nstpbp168, particularly at 600 μM. While pretreated with Compound C before palmitate plus Nstpbp168 treatment, cells would display decreased survival rate implying that Nstpbp168 might protect cells from lipotoxicity through activating AMPK (Fig. 4D).
Furthermore, in Fig 5, we used western blotting analysis to prove that phosphorylated level of AMPK would increase under different doses (50, 100, 250, 500 μM) of palmitate treatment for 24 hours. Nevertheless, the increased phosphorylated AMPK level would be further promoted in presence of palmitate-induced injury after exposured to Nstpbp168 for at
least 24 hours as indicated in Fig 5B. According to the results, it is suggested that Nstpbp168 might have a protective role in reducing lipotoxicity- induced cell death in β-cell through activated AMPK.
In advance, liver cell is also an important organ for lipid metabolism.
In order to assess the protective effects of Nstpbp168 against the
lipotoxicity on liver cell, HepG2, MTT assay was conducted. As shown in Fig. 6A, cells were treated with different dosage of palmitate for 24 hours and cell viability was declined respectively. The correspondent viability loss was achieved 50% under 350 μM palmitate treatment in HepG2.
Pre-treated with Nstpbp168 at 10, 50 or 100 μM for 48 hours could present the protective potential to sustain cell viability from palmitate injury, although there was no significant variation when compared to the result of control as presented in Fig 6B.
Fig 5A showed that the activity of AMPK was enhanced by palmitate treatmenton RINm5F cells. Thus, we would like to verify whether it also employed the same effects on HepG2. Cells were applied with different doses of palmitate for 24 hours and compared the following results with that on RINm5F. However, from Fig 7A, the pattern of
AMPK phosphorylation was not similar with that in RINm5F. Instead, 250μM palmitate could induce higher AMPK activity significantly, but not 350μM palmitate.
As abovementioned, we knew that Nstpbp168 was an AMPK activator. Previous studies also indicated that a phenylpropanoid dibenzylbutyrolactone lignan from Arctium lappa L, Arctigenin could efficiently suppress HepG2 cells from palmitate-induced cell death via
26
activating AMPK (Gu et al., 2012). These intrigued us to further investigate whether Nstpbp168 also exerted its suppressive effects on palmitate-induced damage through activating AMPK in HepG2. As shown in Fig. 7B, 10 μM and 50 μM Nstpbp168 pretreatment could enhance the palmitate-inducied phosphorylation of AMPK but not in 100 μM Nstpbp168 pretreatment , which is not consistent to that in RINm5F cells. This could only provide the possibility that Nstpbp168 might activate AMPK to prevent cell death from palmitate in HepG2 which is needed more evidence to be proved.
3.5 Inhibition of palmitate-induced ROS production by Nstpbp168 Palmitate is recognized as a lipototoxic fatty acid which can elevate oxidative stress and ROS production to cause insulin resistance in
hepatocytes (Shimano et al., 1999). Recent studies indicate that Luteolin could reduce lipid accumulation in HepG2 cells which was related to the activation of AMPK and mitigation of oxidative stress (Liu et al., 2011).
Accordingly, the effects of Nstpbp168 on palmitate-induced ROS production would be detected by using DCFH-DA. Here we utilized fluorescence microscopy to observe the distribution of luminescence resulting from DCF formation. As illustrated in Fig. 8 & 9, treated
RINm5F and HepG2 cells with palmitate could enhance the ROS level at the 24th hour apparently. However, from 10 to 100 μM Nstpbp168
attenuated the ROS generation in the palmitate-stimulated RINm5F and HepG2 cells dose dependently. AICAR was known as an AMPK
activator also could diminish the palmitate-induced ROS production.
Thus, corresponding to previous observation, we proposed that the
ameliorated effect of palmitate-induced lipotoxicity by Nstpbp168 might through regulating AMPK activation to reduce ROS generation.
3.6 Effects of Nstpbp168 on glucose-stimulated insulin secretion in mouse islets.
Since Nstpbp168 could protect cells from H2O2 or palmitate-induced cell death or ROS injury and activate AMPK, we would like to evaluate whether Nstpbp168 stimulated rapid GSIS from mouse islets and whether the phenomenon is similar to that of AMPK activato, AICAR treatment.
As shown in Fig. 10, after mouse islets were pre-incubated with 100 μM Nstpbp168 or AICAR for 2 hours, the response of islets to high glucose significantly decreased in the last 1 h incubation without Nstpbp168. This apparently indicated the inhibition effects of Nstpbp168 on insulin
secretion like the result of AICAR.
28
Chapter 4. Discussion
Previous study has shown that Nstpbp168 employed as an AMPK activator is beneficial in promoting the ability of glucose uptake and
improving insulin sensitivity in vivo and also in C2C12 muscular cell line.
In present study, we further examined the role of Nstpbp168 on
insulin-secreting pancreatic β-cells. The present result demonstrated that Nstpbp168 could significantly reduce H2O2/palmitate-induced oxidative stress and reduce ROS generation in insulin secreting cell line RINm5F.
Additionally, the protective effect of Nstpbp168 on H2O2-induced cell death was significantly suppressed by AMPK inhibitor Compound C. It could enhance AMPK activity in RINm5F cells both in short-term and long-term treatment. Taken together, it is suggested that Nstpbp168 might have a protective potential on ROS-induced cell damage through
mediating AMPK signaling pathway in insulin secreting cells.
4.1 Nstpbp168 has antioxidant potential
In our study, we utilized H2O2 as the source of oxidative stress but not oxidant toxin, such as streptozotocin. Although H2O2 is only
by-production from oxidative stress, it is proposed to be good for studying oxidative issues due to its derived from oxygen-derived intermediate. Pancreatic abnormal glucose metabolism and long-term treatment of free fatty acid (FFAs) can elicit deficient mitochondrial function to raise H2O2 production gradually (Carlsson et al., 1999, Evans et al., 2003, Wang et al., 2004). Hence, H2O2 is supposed to be more
closed to the real physiological status to mimic the oxidative situation
closed to the real physiological status to mimic the oxidative situation