Cytoprotective Effects of Polyenoylphosphatidylcholine (PPC) on β-cells During Diabetic Induction by Streptozotocin

Abstract

Polyenoylphosphatidylcholine (PPC), a phosphatidylcholine-rich phospholipid extracted from soybean, has been reported to protect liver cells from alloxan-induced cytotoxicity. The present study aimed to investigate whether PPC protects pancreatic β-cells from the cytotoxic injury induced by streptozotocin, thus preserving insulin synthesis and secretion. β-Cells of the PPC-treated rats showed a significant reduction of cell death with lesser destruction of plasma membrane on streptozotocin insult. They demonstrated a rapid recovery of GLUT-2 expression, whereas almost irreversible depletion of membranebound GLUT-2 was seen in β-cells of the rats treated with only streptozotocin. A similar cytoprotective effect of PPC was also monitored in the PPC-pretreated MIN6 cells. These β-cells retained their ability to synthesize and secrete insulin and no alteration of glucose metabolism was detected. These results strongly suggest that PPC plays important roles not only in protecting β-cells against cytotoxicity but also in maintaining their insulin synthesis and secretion for normal glucose homeostasis.

Keywords

polyenoylphosphatidylcholine pancreas cytoprotection β-cell insulin streptozotocin diabetes

Impaired insulin secretion of β-cells results in abnormal glucose homeostasis, leading to diabetes. Type I diabetes is caused by selective and progressive destruction of pancreatic β-cells (Doberson et al. 1980; Lernmark et al. 1981; Lernmark 1985; Eisenbarth 1986; Bach 1997; Chervonsky et al. 1997). β-Cells are highly susceptible to cytotoxic agents, including alloxan and streptozotocin (STZ), which induce a cascade of cellular events including DNA strand breaks, activation of poly (ADP-ribose) synthase, and NAD depletion, resulting in impaired function and cell death (Durkacz et al. 1980; Yamamoto et al. 1981a,b; Elsner et al. 2000). It has been known that a primary action of these diabetogenic chemicals is at the plasma membrane level of β-cells (Norlund et al. 1984; Orci et al. 1976). STZ acts directly on the plasma membrane and induces alterations of islet cell membrane properties (Orci et al. 1976). These imply that preservation of membrane integrity could protect β-cells from cytotoxic insults. Field et al. (1990) reported that dietary fat could modify the membrane phospholipid composition of β-cells in diabetic rats and also improve insulin responsiveness. This may imply that dietary fatinduced modifications of membrane composition could improve β-cell function in diabetes. We have reported that dietary soybean protects the β-cells in STZ-treated rats and stimulates insulin synthesis and secretion (Lee and Park 2000). However, it is still not clear which component in the soybean is involved in cytoprotection. It should be underlined that phosphatidylcholine (PPC), a major component of the phospholipid fractions of the islet cells (Diaz et al. 1988), exerts diverse biological actions including receptor processing for insulin responsiveness (Clore et al. 1998,2000), cell growth, and transformation (Turk et al. 1986; Larrodera et al. 1990; Cantafora et al. 1992). PPC extracted from soybean is mainly composed of phosphatidylcholine containing two molecules of linoleic acid. Buko et al. (1996) reported that dietary PPC reduced total lipids and triglycerides in serum and liver of alloxan-treated rats, resulting in alleviation of liver injuries. We therefore propose that dietary PPC could prevent induction and development of diabetes by protecting the β-cell membrane. In the present study we examined the anti-diabetogenic effects of dietary PPC by monitoring the viability and function of β-cells using in vivo and in vitro models.

Materials and Methods

Animals and PPC Feeding

Male Sprague-Dawley rats (Daehan Experimental Animal Lab; Seoul, Korea) weighing 150–200 g with normal fasting blood glucose levels (70–120 mg/dl) were used for PPC feeding and STZ treatment. Fifty-six rats were randomly assigned to the experimental and control groups. PPC (Rhône-Poulenc; Köln, Germany) was dissolved in drinking water and given to the rats at a daily dose of 200 mg/kg (body weight) for 3 weeks according to previous reports (Buko et al. 1994,1996). One group of rats was treated with STZ 7 days after PPC feeding. STZ (60 mg/kg body weight; Sigma, St Louis, MO) was given as a single IV injection into the jugular vein after anesthesia with inhalation of gaseous nitrous oxide-oxygen and isofrane. A second group of rats was injected with STZ without any PPC feeding, and the animals became diabetic, whereas a normal control group of rats was subjected to neither STZ nor PPC treatment. Animals were allowed free access to a normal diet (Samyang Feed; Seoul, Korea). Fasting blood glucose was monitored daily with blood samples collected from the tail vein using a glucometer (Roche; Mannheim, Germany). Three weeks after the experiment, the entire pancreatic tissue was excised from the rats after laparotomy and blood samples were collected from the heart for insulin radioimmunoassay (RIA) determination. The experimental animals were treated humanely and surgical procedures were performed according to the Animal Use and Care protocol.

Tissue Preparation

For light microscopic examination, the pancreatic tissues were fixed overnight with Bouin's solution or 4% paraformaldehyde and 5-μm-thick paraffin sections were mounted on silanized slides (Fisher Scientific; Pittsburgh, PA). For electron microscopic studies, small fragments of tissue were fixed with 4% paraformaldehyde and embedded in Unicryl resin (BBI International; Cardiff, UK).

Immunocytochemical Labeling for Light Microscopy

Immunostaining for light microscopy was carried out using the avidin-biotin-peroxidase complex (Hsu et al. 1981). Briefly, the tissue sections were incubated with the antibodies, either mouse anti-insulin (BioGenex; San Ramon, CA), rabbit anti-glucagon (Dako; Carpinteria, CA), rabbit anti-GLUT-2 (Alpha Diagnostic; San Antonio, TX), or mouse anti-PCNA (Zymed; South San Francisco, CA) and subsequently reacted with the biotinylated anti-mouse or antirabbit immunoglobulins (Vector; Burlingame, CA). After treatment with ABC (Vector), the tissues were oxidized by diaminobenzidine (Sigma; St Louis, MO) and counterstained with hematoxylin for microscopic observation. Some tissues sections fixed with 4% paraformaldehyde were subjected to TUNEL.

Immunoelectron Microscopic Labeling for Insulin

To locate insulin in the secretory granules of β-cells at the subcellular level, electron microscopic immunogold labeling was performed on ultrathin sections as described previously by Bendayan (1984). The tissue sections were incubated with guinea pig anti-insulin antibody (Dako), and the reaction was detected by the protein A-gold complex. After immunogold labeling, the grids were counterstained with uranyl acetate and lead citrate.

In Situ Hybridization

The tissues fixed with 4% paraformaldehyde were subjected to in situ hybridization as described previously (Lee and Park 2000; Kim et al. 2001). Briefly, tissue sections were incubated in the buffer containing digoxigenin-tagged oligonucleotide for insulin (2 μg/ml; R&D Systems, Minneapolis, MN) after digestion with proteinase K (5 μg/ml in Tris-buffered saline, pH 8.0). After hybridization, the sections were incubated with alkaline phosphatase-conjugated anti-digoxigenin antibody (Boehringer Mannheim; Mannheim, Germany). Hybridization signal was revealed by nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate toluidinium.

Image Analysis for Evaluation of Immunocytochemistry and In Situ Hybridization

The β-cell number and islet mass were assessed by morphometry on tissue sections immunostained for insulin or glucagon. The entire area of each tissue section was photographed and digitalized for planimetry. The percent area of islet to pancreatic tissue was assessed using an image analyzing system (Image-Pro Plus; Media Cybernetics, Silver Spring, MD). Populations of β- and α-cells were estimated by manual calculation of insulin or glucagon positive cells in each islet. This morphometric assessment was carried out on all islets present in random sections obtained from at least five different rats of each experimental group. It is possible that the actual numbers of insulin cells are higher than the estimated numbers because only the immunoreactive cells displaying their nuclei were counted as positive cells. On the other hand, insulin content in secretory granules of the β-cell was determined in terms of the insulin-immunogold labeling index at the electron microscopic level (Bendayan 1984,1989). After determining the number of immunogold particles (Ni) and the area of each secretory granule of the beta cell (Sa), the labeling index (I) was calculated according to the formula

I = Ni / sa

Northern Blotting Analysis for Insulin

Total cellular RNA was isolated from pancreatic tissue samples according to the protocol previously described (Chomczynski and Sacchi 1987). Total RNA was transferred to a nylon membrane (Roche) and prehybridized in hybridization buffer (7% SDS, 50% formamide, 5 X SSC, 2% blocking reagent, 50 mM sodium phosphate, pH 7.0, 0.1% N-lauroylsarcosine). Hybridization was performed with digoxigenin-labeled rat insulin cRNA, and the hybridization signal was detected with CSPD (Roche) in X-ray film.

Assessment of Insulin Secretion by RIA

Insulin levels were determined in sera and culture media. Culture media were collected each day. An insulin RIA kit was used according to the manufacturer's protocol (Linco Research; St Charles, MO).

Determination of Proliferation Activity of Islet Cells

To identify the proliferating activity of islet cells, immunostaining for PCNA was carried out on pancreatic tissues using a specific antibody (Zymed). Proliferation rate is presented as the percentage of the PCNA-labeled cells in each islet or in cell culture.

Cell Culture

We induced in vitro cytotoxic injury on a mouse β-cell line (MIN6) which was highly susceptible to STZ. The cells were grown in RPMI 1640 medium (Gibco Life Technologies; Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum and 1% antibiotics. The cells were cultured with PPC (0.5 mM) for 1 day and then incubated with STZ (4 mM) for 2 additional days in the presence of PPC. In a second experiment, pretreatment with PPC was omitted, while both PPC and STZ treatments were excluded for control experiment. Media collected from each culture condition were subjected to RIA for insulin.

Assessment of Glucose-stimulated Insulin Secretion

The cells cultured under the above conditions were washed with PBS and then transferred to serum-free Krebs-Ringer bicarbonate buffer (KRBB; Sigma) for the test of insulin secretion responding to glucose stimulation. The cells were consequently exposed to 5 mM glucose in KRBB and 20 mM for 2 hr each, and insulin contents in KRBBs were determined by combining immunoprecipitation and western blotting (Yang et al. 1994). Briefly, the media were immunoprecipitated with the rabbit anti-insulin antibody. After denaturation they were subjected to 17% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane (0.22 μm; Schleicher & Schuell, Keene, NH). Insulin was detected by chemiluminescence.

Assessment of Cell Death

The cultured cells undergoing apoptosis were determined by TUNEL assay (Sgonc et al. 1994) using a detection kit (Roche Diagnostics). Briefly, after fixation with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 10 min, the cells were digested with proteinase K (1 μg/ml in 0.1 M Tris buffer, pH 7.4) and incubated with TUNEL reaction mixture containing terminal deoxynucleotidyl transferase and fluorescein-labeled nucleotides. The reaction was detected by reactions with horseradish peroxidase-conjugated sheep anti-fluorescein antibody and with diaminobenzidine. The same procedure was applied to the paraffin sections of pancreatic tissues fixed with 4% paraformaldehyde.

Figure 1

Blood glucose concentration (A) and body weight (B) of normal (open circles), diabetic rats (solid triangles), and rats fed PPC before STZ treatment (solid circles). Blood glucose levels were monitored daily at 1100–1200 hr after 3 hr of fasting. STZ (arrow) was injected on day 7 except for the controls. Data are mean values ± SE; n=12 for control, n=15 for diabetic animals, n=17 for PPC-fed rats.

Results

Modulation of Diabetes Induction in PPC-Treated Rats

The rats fed with PPC maintained normal blood glucose levels (139.28 ± 5.79 mg/dl) after STZ treatment (Figure 1A) and demonstrated normal growth and regular weight gain. However, the rats not treated with PPC but having STZ showed severe hyperglycemia (427.92 ± 27.40 mg/dl) and growth retardation with diabetic symptoms, including polyuria (Figure 1B).

Modification of GLUT-2 Expression

We examined modification of GLUT-2 expression in islet cells by immunocytochemistry (Figure 2). Because GLUT-2 is specifically localized in the plasma membrane of β-cells, it can also demonstrate alteration of membrane integrity of β-cells. GLUT-2 expression was significantly reduced by STZ intoxication in both PPC-pretreated and non-treated control rats, but the latter were more severely affected. GLUT-2 was seen in all β-cell membranes, with higher immunoreactivity in normal pancreas, but only some β-cells showed a weak immunoreaction for GLUT-2 in PPC-pretreated rats (Figure 2A and 2D). In non-pretreated control rats, moreover, no GLUT-2 was detectable in the islet cells 24 hr after STZ treatment (Figure 2B). This may be caused not only by loss of the plasma membrane of dying β-cells but also by complete impairment of GLUT-2 expression by STZ. Membrane-bound GLUT-2 was remarkably recovered after the cytotoxic insult in the PPC-pretreated rats than in those treated only with STZ (Figures 2B,2C,2E and 2F). As shown in Figure 2G, β-cells showing GLUT-2 expression in the plasma membrane were rapidly increased by 75% of the population of normal ones 3 weeks after STZ insult, while poor retrieval of GLUT-2 expression was seen in the diabetic controls at the corresponding period.

Figure 2

Modification of GLUT-2 expression in β-cells. GLUT-2 expression detected by immunocytochemistry demonstrates the contour of the β-cell membrane in normal controls (A), diabetic controls at 48 hr (B), and 3 weeks after STZ treatment (C). Modification of membranebound GLUT-2 is determined by immunocytochemistry in the islets of PPC-treated rats 24 hr (D), 48 hr (E), and 3 weeks after STZ injection (F). Number of β-cells encircled by the GLUT-2-labeled plasma membrane was counted in the islets of normal control (N), diabetic control (DM), and PPC-treated rats (PPC). Data are mean values ±SE of seven tissue sections from three to five different rats in each group. ^∗ p<0.05 vs DM. Bars = 25 μm.

Modification of Islet Cell Death and Proliferation

β-Cell death and alteration of islet cell population were prominent in the STZ diabetic rats that were not treated with PPC. β-Cell necrosis showing chromatin condensation and dissolution of plasma membrane was seen 24–48 hr after STZ injection, and some cells demonstrated apoptotic cell death (Figure 3A and 3C). β-Cell death in the core of islets led to shrinkage in islet mass, particularly at 24–48 hr after STZ treatment. However, subsequent α-cell proliferation resulted in a certain regaining of islet mass and modification of the β-cell to α-cell ratio in the islets. In contrast, such loss of β-cells and proliferation of non-β-cells were not seen in the islet cells of the PPC-fed rats in spite of the STZ treatment. Moreover, mitotic features were frequently seen in the islet cells of the rats fed PPC (Figure 3B). We examined proliferation activity of the islet cells by immunostaining for PCNA, a specific marker for proliferating cells (Figure 3D and 3E). The PCNA labeling index, the number of labeled cells in each islet (Figure 3F), was higher in PPC-fed rats (2.91 ± 0.33) than in the diabetic (0.07 ± 0.05) or control (1.17 ± 0.22) animals.

Figure 3

Islet cell death and replication represented by hematoxylineosin (A,B), TUNEL assay (C), and immunolabeling for PCNA (D,E). The islet cells of diabetic rats after 24 hr of STZ treatment (A) showed extensive cell lysis, representing loss of plasma membrane with condensed nuclei (arrows) and dissolved cytoplasm in wide intercellular spaces (asterisks), while some islet cells demonstrated apoptotic cell death in the TUNEL assay (brown nuclei indicated by arrows in C). In contrast, the islet cells of PPC-fed rats (B) were well preserved and even manifested a mitotic figure in the nucleus (arrow). Proliferating cells labeled for PCNA are increased in an islet of PPC-treated rat (brown nuclei in D) compared to normal controls (brown nuclei in E). PCNA labeling index (number of labeled cell/islet) is presented in F. Data are mean values ± SE of nine sections from five different rats in each group. ^∗ p<0.05 vs normal (N) and diabetic controls (DM); †p<0.05 vs N, PPC, PPC-fed rats. Bars = 25 μm.

Morphological Alteration of Pancreatic Islets and β-cells

The islets of the PPC-treated rats retained their conventional features. They showed a normal population of β-cells centrally located, with α-cells at the periphery (Figure 4). However, diabetic rats demonstrated a remarkable increase in α-cells in the central area where only few β-cells remained, resulting in significant modifications of β- to α-cell ratio (Figures 4C-4F and 4H). At the light microscopic level, the highest insulin immunoreactivity was found in β-cells of the PPC-fed rats, whereas only few cells displayed immunoreaction with variable intensities in the diabetic animals (Figure 4B and 4C). Immunoreaction for insulin at the electron microscopic level reflects subcellular insulin content in β-cells (Bendayan 1989). The PPC-treated rats demonstrated typical profiles of β-cell ultrastructure as seen in control rats. Their secretory granules were characterized by a relatively wide halo surrounding inner dense cores, which were labeled by the insulin antibodyimmunogold complex. They demonstrated labeling densities similar to those of normal animals (Figures 5A,5B, and 5D). In contrast, poor immunolabeling was found on secretory granules of β-cells in diabetic animals. These granules were difficult to find due to the considerable degranulation of the cells (Figure 5C and 5D).

Figure 4

Population of insulin (A-C) and glucagon cells (D-F) in the pancreatic tissues of normal (A,D), PPC-fed (B,E), and diabetic rats (C,F). Note the proliferated glucagon cells (F) and few insulin cells (C) in diabetic animals. In the PPC-treated rats, these types of cells (B,E) show similar distributions as the controls (A,D). Percentage of β-cell area to pancreatic tissue (G) and ratio of β-cells to a-cells (H) were measured in tissue sections of PPC-treated rats (PPC), normal controls (N), and diabetic animals (DM). Data are mean values ± SE of nine sections from five different rats in each group. ^∗ p<0.05 vs DM; †p<0.05 vs N; NS, not significant. Bars = 50 μm.

Insulin mRNA Expression

Expression of insulin mRNA, as an indication of insulin synthesis, was evaluated by in situ hybridization and Northern blotting analysis. Hybridization signals for insulin mRNA were more intense in β-cells of the PPC-treated rats than in those of the normal and diabetic animals (Figures 6A-6C). Higher expression of insulin mRNA was also detected in the pancreatic tissue of the rats fed with PPC by Northern blotting analysis, showing intense expression of insulin mRNA equivalent to that of the normal animals (Figure 6D and 6E).

Figure 5

Ultrastructure and insulinimmunogold labeling of β-cells in pancreas of normal (A), PPC-treated (B) and diabetic rats (C). The β-cell granules of diabetic rats are few in number and small in size, whereas PPC-treated rats (B) show intense immunogold labeling on the secretory granules, which present normal profiles similar to those of controls (A). Immunogold labeling for insulin in unit area (μm²) of the granules is shown in D. N, normal; DM, diabetic animals; PPC, PPC-treated group. Data are mean values ± SE from five different rats in each group. ^∗ p<0.05; NS, not significant. Bars = 1 μm.

Modulation of Insulin Secretion and Glucose Metabolism

Fasting plasma insulin level in the rats fed with PPC (0.38 ± 0.05 ng/ml) was within the normal range, whereas that of the diabetic animals (0.05 ± 0.01 ng/ml) was significantly lower (Figure 7A). PPC-fed rats showed a normal glucose tolerance value of 75 g IPGTT, whereas the diabetic rats manifested impaired glucose tolerance (Figure 7B).

Preservation of Cell Viability and Insulin Secretion by PPC in βcell Culture

β-Cell death was examined in MIN6 cell culture by TUNEL assay. A higher rate of apoptotic cell death (7.65 ± 3.04%) was monitored after 48 hr of treatment with STZ, whereas a much lesser number of apoptotic cells was demonstrated in PPC-pretreated cells before the STZ insult (1.47 ± 0.378%) and in normal control cells (0.60 ± 0.20%) (Figures 8A-8D). Alteration of insulin secretion and glucose sensitiveness in the MIN6 cells was examined by RIA and immunoprecipitation-Western blotting analysis. Insulin concentration in culture media significantly declined after STZ treatment, whereas the media from PPC-treated cells showed no decrease in insulin concentration in spite of the STZ treatment (Figure 9A). We assessed glucose-stimulated insulin secretion in these cultured cells. As shown in Figure 9B, PPC-pretreated MIN cells produced higher levels of insulin level in response to glucose stimulation, whereas STZ-treated cells without PPC treatment displayed loss of glucose responsiveness.

Figure 6

Insulin mRNA expression detected by in situ hybridization (A-C) and Northern blotting analysis (D) with its densitometric evaluation (E). In the islets of diabetic animals (C), lower hybridization signal is seen in some β-cells compared to controls (A), whereas intense signal is found in all β-cells of the PPC-fed rats (B). Northern blotting analysis (D,E) shows similar expression patterns of in situ hybridization. The mRNA expression levels of Northern blots from four independent experiments represent percentage of the controls (N). Data are mean values ± SE. ^∗ p<0.05; NS, not significant; DM, diabetic; PPC, PPC-treated group. Bars = 50 μm.

Figure 7

Changes in plasma insulin concentration (A) and IPGTT (B). Plasma samples were collected at time of sacrifice after 3 hr of fasting from normal rats (N), PPC-treated group (PPC), and diabetic animals (DM). Data in A are mean values ± SE; n=5 for N, n=7 for PPC, n=6 for DM. ^∗ p<0.05; NS, not significant. (B) Glucose tolerance test after 3 weeks of experiments in normal rats (open circles), diabetic rats (solid triangles), and rats fed with PPC (solid circles). Blood glucose levels were determined after IP injection of a glucose bolus (75 g/kg bw) at given time points for 120 min. Data in B are plotted as mean values ± SE; n=4 for normal control, n=4 for diabetic animals, n=5 for PPC-treated group.

Discussion

We investigated the effects of PPC as a cytoprotector of β-cells preventing development of diabetes induced by STZ. Membrane phospholipids are important regulators of cell functions such as growth, viability, homeostasis, and signal transduction (Awad et al. 1996; Wang et al. 1998; Flores et al. 2000). Modification of phospholipid composition could therefore deteriorate cell functions by altering membrane fluidity, permeability, and receptor activity (Buko et al. 1994; Saragovi et al. 1998; Alvarez et al. 2001; Valcarcel et al. 2001). STZ, a diabetogenic compound that induces death of β-cells, modifies the molecular structure of phospholipids, particularly phosphatidylcholine (PC), which is a major phospholipid of the outer leaflet of the plasma membrane, resulting in decrease of membrane fluidity in β-cells (Vecchini et al. 2000; Detmar et al. 1994; Wieder et al. 1995; Cui et al. 1996). In the present study we found that β-cell death was considerably inhibited with amelioration of membrane destruction in the PPC-treated rats, although their membrane-bound GLUT-2 expression was significantly suppressed. GLUT-2 is not only a structural protein specific for β-cell membrane but is also a crucial constituent for recognition and entry of glucose as well as the glucose analogue STZ (Noel and Newgard 1997). This indicates that suppression of β-cell death in PPC-treated animals is not associated with the facility of STZ entry by GLUT-2 in the β-cell membrane. PPC-treated rats, moreover, exhibited an increased proliferation activity with a rapid retrieval of membranebound GLUT-2 in the islet cells after STZ insult. PPC is a polyunsaturated phosphatidylcholine bearing two linoleic acids in its molecule. Dietary PPC can provide unsaturated fatty acid-rich phospholipids to the β-cell membrane, thus facilitating reconstruction of membrane integrity as reported in rat liver tissue (Buko et al. 1996). It is therefore likely that PPC facilitates retrieval membrane constituents of β-cells as well as their cell replicability.

Figure 8

In-vitro detection of cell death by TUNEL assay. The nuclei of apoptotic cells (brown nuclei indicated by arrows in A-C) were seen in MIN6 cells cultured with PPC before STZ treatment (PPC-STZ) or STZ treatment only (STZ). The cells for normal controls (N) were cultured without PPC or STZ treatment. The percentage of apoptotic cells to total cells is shown in D. Data in D are mean values ± SE (n=4 from independent experiments; ^∗ p<0.05 vs STZ; NS, not significant). Bars = 25 μm.

Figure 9

Insulin concentrations of culture media from different conditions. Media were collected after 24 hr of culture without STZ treatment (STZ [-]) and 24 hr of culture in the presence of STZ (STZ [+]). There is a significant decrease in insulin concentration after STZ treatment in the culture without PPC treatment (solid triangles), whereas no reduction was noted in both the PPC-pretreated culture (open circles) and the untreated control (solid circles). Data in A are mean values ± SE (n=3 independent experiments). Glucose-stimulated insulin secretion was determined in these cultured cells by combining immunoprecipitation and Western blotting (B). Streptozotocin was given to the PPC pretreated (PPC-STZ) or untreated MIN6 cells (STZ), then exposed to 5 mM and 20 mM glucose for 2 hr each with serum-free Krebs-Ringer bicarbonate buffer (KRBB). The culture media were collected for insulin analysis from those cultures as well as from normal control cells. PPC-pretreated MIN cells produced higher levels of insulin in response to glucose stimulation, whereas the STZ-treated cells without PPC treatment display loss of glucose responsiveness. Data in B are representive of three independent experiments.

The β-cells, along with preservation of their population in PPC-treated rats, did not manifest impaired insulin secretion. They displayed well-preserved insulin immunoreactivity and insulin mRNA expression as well as glucose-stimulated insulin secretion, indicating an almost intact process for insulin synthesis and secretion. In fact, the secreted insulin in bloodstream or in culture media was within the normal range in PPC-treated groups.

We suggest that a sufficient supply of phospholipid with unsaturated fatty acids should be considered not only to improve β-cell function but also to suppress the diabetogenic process in diabetes-prone subjects.

Footnotes

Acknowledgements

Supported by a grant from the Oriental Medicine R&D Project, Ministry or Health and Welfare, Republic of Korea (HMP-00-CO-06-0006).

We wish to thank Dr V. U. Buko (Institute of Biochemistry, National Academy of Science of Belarus) for kind discussion and suggestions.

References

Alvarez

Ruiz-Gutierrez

Sobrino

Santa-Maria

(2001) Age-related changes in membrane lipid composition, fluidity and respiratory burst in rat peritoneal neutrophils. Clin Exp Immunol 124:95–102

Awad

Ntanios

Fink

Horvath

(1996) Effect of membrane lipid alteration on the growth, phospholipase C activity and G protein of HT-29 tumor cells. Prostaglandins Leukotrienes Essent Fatty Acids 55:293–302

Bach

(1997) Autoimmunity and type I diabetes. Trends Endocrinol Metab 8:71–74

Bendayan

(1984) Protein A-gold electron microscopic immunocytochemistry: methods, applications and limitations. J Electron Microsc Tech 1:243–270

Bendayan

(1989) Ultrastructural localization of insulin and C-peptide antigenic sites in rat pancreatic B-cell applying the quantitative high resolution protein A-gold approach. Am J Anat 185:205–216

Buko

Artsukevich

Maltsev

Nikitin

Ignatenko

Gundermann

Schumacher

(1994) Effect of polyunsaturated phosphatidylcholine on lipid structure and cAMP-dependent signal transduction in the liver of rats chronically intoxicated with ethanol. Exp Toxicol Pathol 46:375–382

Buko

Lukivskaya

Nikitin

Tarasov

Zavodnik

Borodinsky

Gorenshtein

. (1996) Hepatic and pancreatic effects of polyenoylphosphatidylcholine in rats with alloxaninduced diabetes. Cell Biochem Funct 14:131–137

Cantafora

Masella

Angelico

Gandin

Blount

Peterson

(1992) Effect of intravenous polyunsaturated phosphatidylcholine infusion on insulin receptor processing and lipid composition of erythrocytes in patients with liver cirrhosis. Eur J Clin Invest 22:777–782

Chervonsky

Wang

Wong

Visintin

Flavell

Janeway

Jr Matis

(1997) The role of fas in autoimmune diabetes. Cell 89:17–24

10.

Chomczynski

Sacchi

(1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159

11.

Clore

Harris

Azzam

Gill

Zuelzer

Rizzo

. (2000) Changes in phosphatidylcholine fatty acid composition are associated with altered skeletal muscle insulin responsiveness in normal man. Metabolism 49:232–238

12.

Clore

Gill

Gupta

Spencer

Azzam

Zuelzer

. (1998) Skeletal muscle phosphatidylcholine fatty acids and insulin sensitivity in normal humans. Am J Physiol 275:E665–670

13.

Cui

Houweling

Chen

Record

Chap

Vance

Terce

(1996) A genetic defect in phosphatidylcholine biosynthesis triggers apoptosis in Chinese hamster ovary cells. J Biol Chem 271:14668–14671

14.

Detmar

Geilen

Wieder

Orfanos

Reutter

(1994) Phospholipid analogue hexadecylphosphocholine inhibits proliferation and phosphatidylcholine biosynthesis of human epidermal keratinocytes in vitro. J Invest Dermatol 102:490–494

15.

Diaz

Cortizo

Garcia

Gagliardino

(1988) Lipid composition of normal male rat islets. Lipids 23:1125–1128

16.

Doberson

Scharff

Ginsberg-Fellner

Notkins

(1980) Cytotoxic autoantibodies to beta cells in the serum of patients with insulin-dependent diabetes mellitus. N Engl J Med 303:1493–1498

17.

Durkacz

Omidiji

Gray

Shall

(1980) (ADP-ribose)n participates in DNA excision repair. Nature 283:593–596

18.

Eisenbarth

(1986) Type I diabetes mellitus: a chronic autoimmune disease. N Engl J Med 314:1360–1368

19.

Elsner

Guldbakke

Tiedge

Munday

Lenzen

(2000) Relative importance of transport and alkylation for pancreatic beta-cell toxicity of streptozotocin. Diabetologia 43:1528–1533

20.

Field

Ryan

Thomson

Clandinin

(1990) Diet fat composition alters membrane phospholipid composition, insulin binding, and glucose metabolism in adipocytes from control and diabetic animals. J Biol Chem 265:11143–11150

21.

Flores

Jones

Merida

(2000) Changes in the balance between mitogenic and antimitogenic lipid second messengers during proliferation, cell arrest, and apoptosis in T-lymphocytes. FASEB J 14:1873–1875

22.

Hsu

Raine

Fanger

(1981) Use of avidin-biotin peroxidase complex (ABC) in immunoperoxidase technique: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 29:577–580

23.

Kim

Han

Shin

Min

Park

(2001) Clusterin expression during regeneration of the pancreatic islet cells in the streptozotocin-induced diabetic rats. Diabetologia 44:2192–2202

24.

Larrodera

Cornet

Diaz-Meco

Lopez-Barahona

Diaz-Laviada

Guddal

Johansen

. (1990) Phospholipase C-mediated hydrolysis of phosphatidylcholine is an important step in PDGF-stimulated DNA synthesis. Cell 61:1113–1120

25.

Lee

Park

(2000) Effects of soybean diet on the beta cells in the streptozotocin treated rats for induction of diabetes. Diabetes Res Clin Pract 47:1–13

26.

Lernmark

(1985) Molecular biology of type I (insulin-dependent) diabetes mellitus. Diabetologia 28:195–203

27.

Lernmark

Haggloff

Freedman

Irvine

Ludvigsson

Holmgren

(1981) A prospective analysis of antibodies reacting with pancreatic islet cells in insulin-dependent diabetic children. Diabetologia 20:471–474

28.

Noel

Newgard

(1997) Structural domains that contribute to substrate specificity in facilitated glucose transporters are distinct from those involved in kinetic function: studies with GLUT-1/GLUT-2 chimeras. Biochemistry 3:5465–5475

29.

Norlund

Grankvist

Norlund

Taljedal

(1984) Ultrastructure and membrane permeability of cultured pancreatic beta-cells exposed to alloxan or 6-hydroxydopamine. Virchows Arch [A] 404:31–38

30.

Orci

Amherdt

Malaisse-Lagae

Ravazzola

Malaisse

Perrelet

Renold

(1976) Islet cell membrane alteration by diabetogenic drugs. Lab Invest 34:451–454

31.

Saragovi

Bhandoola

Moreau

Lavine

Gagnon

Lemercier

Theze

(1998) Functional and physical association of a cell surface phospholipid and interleukin-2 receptor p⁵⁵(alpha) subunits. Biochim Biophys Acta 1414:51–64

32.

Sgonc

Boeck

Dietrich

Gruber

Recheis

Wick

(1994) Simultaneous determination of cell surface antigens and apoptosis. Trends Gene 10:41–42

33.

Turk

Wolf

Lefkowith

Stump

McDaniel

(1986) Glucose-induced phospholipid hydrolysis in isolated pancreatic islets: quantitative effects on the phospholipid content of arachidonate and other fatty acids. Biochim Biophys Acta 879:399–409

34.

Valcarcel

Dalla Serra

Potrich

Bernhart

Tejuca

Martinez

Pazos

. (2001) Effects of lipid composition on membrane permeabilization by sticholysin I and II, two cytolysins of the sea anemone Stichodactyla helianthus. Biophys J 80:2761–2774

35.

Vecchini

Del Rosso

Binaglia

Dhalla

Panagia

(2000) Molecular defects in sarcolemmal glycerophospholipid subclasses in diabetic cardiomyopathy. J Mol Cell Cardiol 32:1061–1074

36.

Wang

Chang

Xing

(1998) The interference effects of hexadecylphosphocholine on proliferation and membrane phospholipid metabolism in human myeloid leukemia cell lines. Int J Tissue React 20:101–107

37.

Wieder

Haase

Geilen

Orfanos

(1995) The effect of two synthetic phospholipids on cell proliferation and phosphatidylcholine biosynthesis in Madin-Darby canine kidney cells. Lipids 30:389–393

38.

Yamamoto

Uchigata

Okamoto

(1981a) DNA strand breaks in pancreatic islets by in vivo adminstration of alloxan or streptozoticin. Biochem Biophys Res Commun 103:1014–1020

39.

Yamamoto

Uchigata

Okamoto

(1981b) Streptozotocin and alloxan induce DNA strand breaks and poly(ADP-ribose) synthetase in pancreatic islets. Nature 294:284–286

40.

Yang

Rhee

Williamson

(1994) Epidermal growth factorinduced activation and translocation of phospholipase C-gamma 1 to the cytoskeleton in rat hepatocytes. J Biol Chem 269:7156–7162