Advertisement
CARBOHYDRATES, LIPIDS, AND OTHER NATURAL PRODUCTS| Volume 273, ISSUE 49, P32487-32490, December 04, 1998

Lipoapoptosis in Beta-cells of Obese Prediabeticfa/fa Rats

ROLE OF SERINE PALMITOYLTRANSFERASE OVEREXPRESSION*
Open AccessPublished:December 04, 1998DOI:https://doi.org/10.1074/jbc.273.49.32487
      We reported that the lipoapoptosis of beta-cells observed in fat-laden islets of obese fa/fa Zucker Diabetic Fatty (ZDF) rats results from overproduction of ceramide, an initiator of the apoptotic cascade and is induced by long-chain fatty acids (FA). Whereas the ceramide of cytokine-induced apoptosis may be derived from sphingomyelin hydrolysis, FA-induced ceramide overproduction seems to be derived from FA. We therefore semiquantified mRNA of serine palmitoyltransferase (SPT), which catalyzes the first step in ceramide synthesis. It was 2–3-fold higher in fa/fa islets than in +/+ controls. [3H]Ceramide formation from [3H]serine was 2.2–4.5-fold higher in fa/faislets. Triacsin-C, which blocks palmitoyl-CoA synthesis, andl-cycloserine, which blocks SPT activity, completely blocked [3H]ceramide formation from [3H]serine. Islets of fa/fa rats are unresponsive to the lipopenic action of leptin, which normally depletes fat and prevents FA up-regulation of SPT. To determine the role of leptin unresponsiveness in the SPT overexpression, we transferred wild type OB-Rb cDNA to their islets; now leptin completely blocked the exaggerated FA-induced increase of SPT mRNA while reducing the fat content. Beta-cell lipoapoptosis was partially prevented in vivo by treating prediabetic ZDF rats withl-cycloserine for 2 weeks. Ceramide content and DNA fragmentation both declined 40–50%. We conclude that lipoapoptosis of ZDF rats is mediated by enhanced ceramide synthesis from FA and that blockade by SPT inhibitors prevents lipoapoptosis.
      ZDF
      Zucker Diabetic Fatty
      FA
      fatty acid
      SPT
      serine palmitoyltransferase
      PBS
      phosphate-buffered saline
      TG
      triglyceride.
      Obesity has become the single most prevalent health problem in the United States and non-insulin-dependent diabetes mellitus its most common complication (
      • National Center for Health Statistics
      ). Yet the mechanism by which obesity can damage remotely situated tissues such as the pancreatic beta-cells has not been established. Recently however, considerable evidence has implicated fatty acid excess as the major factor in obesity-associated beta-cell dysfunction and damage in a rodent model of adipogenic non-insulin-dependent diabetes mellitus, the Zucker Diabetic Fatty (ZDF)1rat (
      • Lee Y.
      • Hirose H.
      • Ohneda M.
      • Johnson J.H.
      • McGarry J.D.
      • Unger R.H.
      ,
      • Lee Y.
      • Hirose H.
      • Zhou Y.-T.
      • Esser V.
      • McGarry J.D.
      • Unger R.H.
      ,
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ).
      Recent studies from this laboratory have demonstrated that long-chain fatty acids induce apoptosis of the beta-cells of obese ZDF rats (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). Furthermore, the evidence pointed to de novo ceramide synthesis as a major factor in the lipoapoptosis (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). Obese ZDF rats are homozygous for the fa mutation (fa/fa), a Gln-269 → Pro substitution in the leptin receptor (OB-R) (
      • Iida M.
      • Murakami T.
      • Ishida K.
      • Mizuno A.
      • Kuwajima M.
      • Shima K.
      ,
      • Phillips M.S.
      • Liu Q.
      • Hammond H.
      • Dugan V.
      • Hey P.
      • Caskey C.T.
      • Hess J.F.
      ). The lipid content of their islets is markedly elevated because of complete unresponsiveness of their fat-laden islets to the lipopenic action of leptin (
      • Shimabukuro M.
      • Koyama K.
      • Chen G.
      • Wang M-Y.
      • Trieu F.
      • Lee Y.
      • Newgard C.B.
      • Unger R.H.
      ). We theorized that the overproduction of ceramide was somehow linked to the overproduction of fat.
      Because each molecule of ceramide contains two molecules of long-chain fatty acids (FA), it seemed likely that fatty acids stimulated ceramide-mediated apoptosis by providing excess substrate for de novo ceramide synthesis rather than by causing sphingomyelin hydrolysis to ceramide. To test this possibility, we studied [3H]ceramide formation from [3H]palmitate and observed a marked increase in islets from fa/fa ZDF rats (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). This overproduction of ceramide was blocked with fumonisin-B1, an inhibitor of ceramide synthetase (
      • Wang E.
      • Norred W.P.
      • Bacon C.W.
      • Riley R.T.
      • Merrill Jr., A.H.
      ), which also blocked the apoptosis (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). Thus de novo synthesis of ceramide appeared to be important in fatty acid-induced apoptosis.
      However, the mechanism of de novo ceramide overproduction was not fully explained by the foregoing studies. Because the high triglyceride (TG) content of these islets provided an expanded source of palmitoyl-CoA for de novo ceramide formation (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ), we considered an increase within the metabolic pathway of ceramide synthesis. The first step in the pathway involves serine palmitoyltransferase (SPT), the enzyme that catalyzes the condensation of palmitoyl-CoA and serine to form dehydrosphinganine, a precursor of sphingosine (
      • Weiss B.
      • Stoffel W.
      ,
      • Hanada K.
      • Hara T.
      • Nishijima M.
      • Kuge O.
      • Dickson R.C.
      • Nagiec M.M.
      ); sphingosine is then acylated to form ceramide (
      • Merrill Jr., A.H.
      • Jones D.D.
      ,
      • Obeid L.M.
      • Linardic C.M.
      • Karolak L.A.
      • Hannun Y.A.
      ). The following study was designed to determine whether there is an increase in SPT activity and, if so, whether fatty acid-induced ceramide-mediated apoptosis of beta-cells in islets of ZDF rats could be blocked by inhibiting SPT activity and if this prevents the partial destruction of beta-cells that accompanies this form of obesity.

      MATERIALS AND METHODS

      Animals

      Lean wild-type (+/+) male ZDF rats and obese homozygous (fa/fa) male ZDF rats were bred in our laboratory from (ZDF/Drt-fa(F10)) rats originally purchased from Dr. R. Peterson (University of Indiana School of Medicine, Indianapolis, IN). In some rats (n = 6), 25 mg/kg/day of l-cycloserine (freshly dissolved in phosphate-buffered saline (PBS)) was injected intraperitoneally for 2 weeks. In controls (n = 6), only PBS was given.

      Islet Isolation and Culture

      Pancreatic islets were isolated by the method of Naber et al. (
      • Naber S.P.
      • McDonald J.M.
      • Jarett L.
      • McDaniel M.L.
      • Ludvigsen C.W.
      • Lacy P.E.
      ) with modifications (
      • Shimabukuro M.
      • Ohneda M.
      • Lee Y.
      • Unger R.H.
      ). Isolated islets were cultured as described previously (
      • Shimabukuro M.
      • Ohneda M.
      • Lee Y.
      • Unger R.H.
      ). In some experiments, islets were cultured with or without 1 mmlong-chain FA (2:1, oleate:palmitate, Sigma) in 2% bovine serum albumin in the absence or presence of 20 ng/ml recombinant mouse leptin (kindly provided by Ron Chance, Ph.D., Lilly), or 10 μmTriacsin-C (Biomol Research Laboratories, Plymouth Meeting, PA).

      De Novo Ceramide Synthesis From [3H]Serine

      Groups of 100–200 islets were cultured for 2 days with [3H]serine (Amersham Pharmacia Biotech) in the absence or presence of 10 μm Triacsin-C or 2 mml-cycloserine (Sigma). Lipids were extracted as described (
      • Bligh E.G.
      • Dyer W.J.
      ) with modification (
      • Shimabukuro M.
      • Koyama K.
      • Chen G.
      • Wang M-Y.
      • Trieu F.
      • Lee Y.
      • Newgard C.B.
      • Unger R.H.
      ). Lipid extracts dissolved in a volume of 80 μl of chloroform were spotted onto high performance thin layer chromatography plates (Merck) and developed with chloroform:methanol:water (65:25:4, v:v:v). The radioactive spot corresponding to [3H]ceramide was counted as described (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ).

      Semiquantitation of SPT mRNA by Reverse Transcription Polymerase Chain Reaction

      SPT mRNA expression was analyzed in cultured islets using the reverse transcription polymerase chain reaction protocol as described previously in detail (
      • Shimabukuro M.
      • Ohneda M.
      • Lee Y.
      • Unger R.H.
      ). Briefly, total RNA was extracted using TRIzol isolation kit (Life Technologies) and treated with RNase-free DNase. First-strand cDNA was obtained using a first strand cDNA synthesis kit (CLONTECH). Primers used to amplify SPT and β-actin cDNA were: 5′-GGTGTGGCTGTGCTTGAATA-3′ (1501–1520) and 5′-AACCCTCTTCCCAAAACTGA-3′ (2031–2050) for SPT (GenBankTMaccession number X95641, 550-base pair fragment) and 5′-TTGTAACCAACTGGGACGATATGG-3′ (1552–1575) and 5′-GATCTTGATCTTCATGGTGCTAGG-3′ (2991–2844) for β-actin (GenBankTM accession number J00691, 764-base pair fragment). The products were electrophoresed on a 1.2% agarose gel. After transferring to a Hybond-N nylon membrane (Amersham Pharmacia Biotech), DNA samples were hybridized with [32P]ATP-labeled internal probes and analyzed in the Molecular Imager (Bio-Rad). The internal probes were: 5′-CGGCACTGCAGAAACCACCAGGGATATGCT-3′ for SPT and 5′-GGTCAGGATCTTCATGAGGTAGTCTGTCAG-3′ for β-actin. Levels of SPT mRNA were expressed as the ratio of the signal intensity relative to that for β-actin.

      Overexpression of Wild-type Leptin Receptor

      As described previously (
      • Wang M-Y.
      • Koyama K.
      • Shimabukuro M.
      • Newgard C.B.
      • Unger R.H.
      ), pancreata of 10-week old male ZDF (fa/fa) rats were perfused with 1 × 1012 plaque-forming units of recombinant adenovirus containing either the full-length wild-type leptin receptor (OB-Rb) cDNA (AdCMV-OB-Rb) or the β-galactosidase (AdCMV-β-gal) cDNA in Krebs-Ringer bicarbonate buffer with 4.5% Dextran-T 70, 1% bovine serum albumin, 5.6 mm glucose, and 5 mm each of sodium pyruvate, sodium glutamate, and sodium fumarate. Pancreatic islets were then isolated and maintained for 2 days in suspension culture in 60-mm Petri dishes at 37 °C in a humidified atmosphere of 5% CO2 and 95% air, as described previously (
      • Wang M-Y.
      • Koyama K.
      • Shimabukuro M.
      • Newgard C.B.
      • Unger R.H.
      ). The culture medium consisted of RPMI 1640 supplemented with 10% fetal calf serum, 1% penicillin and streptomycin, and 8 mm glucose. In some experiments, 20 ng/ml recombinant mouse leptin was added to islets cultured either with or without a 1 mm FA mixture consisting of 2:1 oleate:palmitate in 2% bovine serum albumin.

      DNA Fragmentation Assay

      DNA fragmentation was assayed by a modification (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ) of the method of Duke and Sellins (
      • Duke R.C.
      • Sellins C.B.
      ). Groups of ∼200 islets were washed twice with cold PBS and suspended in 100 ml of lysis buffer (10 mm Tris-HCl, 10 mm EDTA, and 0.5% Triton X-100, pH 8.0). Islets were then homogenized every 5 min by gentle pipetting. Incubation was carried out in an ice bath for 20 min. After centrifugation for 20 min at 4 °C (14000 ×g), the supernatant containing fragmented (soluble) DNA was transferred to another tube. Lysis buffer (100 ml) was added to the pellet containing insoluble DNA. Both samples were treated with RNase A (0.5 mg/ml) for 1 h at 37 °C and then with Proteinase K (Sigma, 0.4 mg/ml) for 1 h at 37 °C. After adding 20 ml of 5m NaCl and 120 ml of isopropyl alcohol, the samples were incubated overnight at −20 °C, and the DNA concentrations were measured by the method of Hopcroft et al. (
      • Hopcroft D.W.
      • Mason D.R.
      • Scott R.S.
      ). Fragmented DNA was calculated as 100% × soluble DNA/(soluble + insoluble DNA). The soluble fraction of DNA was determined by electrophoresis on 1.5% agarose gel and has a ladder-like appearance.

      Statistical Analysis

      All values shown are expressed as mean ± S.E. Statistical analysis was performed by unpaired Student's t test or by one-way analysis of variance.

      RESULTS

      Comparison of Serine Palmitoyltransferase mRNA

      To determine whether the increase in de novo ceramide synthesis in fa/fa islets (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ) was the consequence of overexpression of SPT, its mRNA was semiquantified by reverse transcriptase-polymerase chain reaction. At 7 and 14 weeks of age, the SPT/β-actin mRNA ratio was, respectively, 2 and 3 times greater than that of +/+ controls (Fig. 1).
      Figure thumbnail gr1
      Figure 1Comparison of SPT mRNA expression in islets from 7- and 14-week-old lean +/+ and obesefa/fa ZDF rats. fa/fa ZDF rats are prediabetic at 7 weeks of age and diabetic at 14 weeks. Representative blots are displayed.Bars represent the mean (± S.E.) mRNA ratio of SPT/β-actin (n = 3) in +/+ rats (open bars) and in fa/fa rats (hatched bars). *, p < 0.01 versus +/+ rats; †, p < 0.05 versus7-week-old fa/fa rats.

      Comparison of Ceramide Synthesis from Serine

      To determine whether the increase in SPT mRNA was associated with an increase in de novo ceramide formation from serine, we compared the rate of [3H]ceramide formation from [3H]serine in +/+ and fa/fa ZDF islets (Fig. 2). The rate of [3H]ceramide formation in fa/fa islets was 2.2 times that of +/+ controls islets at 7 weeks of age and 3.5 times greater at 14 weeks of age. Thus, there was enhancement of ceramide synthesis from serine, matching our earlier report of increased ceramide formation from palmitate (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). Moreover, the increase in ceramide synthesis correlated remarkably well with the increase in SPT mRNA. The appearance of [3H]sphingomyelin, [3H]phosphatidylethanolamine, and [3H]phosphatidylcholine were the same in +/+and fa/fa islets, evidence that the increase in synthetic activity was limited to the serine → ceramide pathway.
      Figure thumbnail gr2
      Figure 2Comparison of [3H]ceramide formation from [3H]serine in islets from 7- and 14-week-old +/+ and fa/fa ZDF rats.Although most of the label was present in [3H]sphingomyelin (SM), [3H]phosphatidylethanolamine (PE), and [3H]phosphatidylcholine (PC), the only differences between +/+ and fa/fa islets are in [3H]ceramide formation. A, representative thin layer chromatography blot. B, bars represent the mean ± S.E. of arbitrary densitometric units (n = 3). *,p < 0.01 versus +/+ controls; †, p < 0.05 versus 7-week-oldfa/fa islets.

      Effects of Metabolic Blockers on Ceramide Synthesis

      For additional evidence that the increased ceramide formation in fa/fa islets reflected de novo ceramide formation, we prevented fatty acyl-CoA formation in fa/faZDF islets by culturing them in 10 μm Triacsin-C, which blocks fatty acyl-CoA synthetase (
      • Noel R.J.
      • Antinozzi P.A.
      • McGarry J.D.
      • Newgard C.B.
      ,
      • Antinozzi P.A.
      • Segall L.
      • Prentki M.
      • McGarry J.D.
      • Newgard C.B.
      ), or with 2 mml-cycloserine, which blocks SPT activity (
      • Sundaram K.S.
      • Lev M.
      ). As shown in Fig. 3, both completely blocked the formation of [3H]ceramide from [3H]serine. This indicated that de novooverproduction was the source of the ceramide excess and that blockade of fatty acid activation to fatty acyl-CoA of fatty acyl-CoA condensation with serine prevented excess ceramide formation.
      Figure thumbnail gr3
      Figure 3Effects of l-cycloserine and Triacsin-C on [3H]ceramide formation from [3H]serine in islets of 7-week-old +/+ and fa/fa rats. Bars represent the mean (± S.E.) of arbitrary densitometric units (n = 3). *, p < 0.01 versus +/+ controls; †,p < 0.05 versus control fa/faislets (without cycloserine or Triacsin-C).

      Effects of Long-chain Fatty Acids on SPT mRNA

      Next, we attempted to identify the cause of the overexpression of SPT. Based on the mutation in the leptin receptor, the most plausible explanation for ceramide overproduction seemed to be either lack of a direct inhibitory action of leptin on ceramide synthesis or an indirect effect resulting from the excess lipid content of these islets. Because leptin swiftly depletes lipid content in leptin-responsive islets (
      • Shimabukuro M.
      • Koyama K.
      • Chen G.
      • Wang M-Y.
      • Trieu F.
      • Lee Y.
      • Newgard C.B.
      • Unger R.H.
      ), it would be next to impossible to distinguish between direct lowering of SPT expression by leptin and indirect lowering by leptin-induced reduction in lipids. Therefore, we studied the effect of 1 mm FA on SPT expression in +/+ and fa/fa islets. In both groups of islets, FA caused a significant increase in SPT/β-actin mRNA ratio (Fig. 4). However, the base-line SPT expression (without added FA) in the fa/fa islets was 90% higher, and the fatty acid-induced increment in SPT expression (above the base line) was 60% higher in fa/fa islets than in control +/+ islets. Both the base-line TG content and the fatty acid-induced increase in TG were proportionately higher in fa/fa islets (Table I), raising the possibility that SPT expression was influenced by the ambient islet lipid content.
      Figure thumbnail gr4
      Figure 4Effects of 1 mm FA on SPT mRNA expression in islets from 7-week-old lean +/+ and obese fa/fa ZDF rats. Representative blots are displayed. Bars represent the mean (± S.E.) mRNA ratio of SPT/β-actin (n = 3). *, p < 0.01 versus 0 mm FA.
      Table IEffect of recombinant leptin on islet triglyceride content (ng/islet)
      Fatty acids (mm)0011
      Leptin (ng/ml)020020
      AdCMV-β-gal48 ± 0.146 ± 0.1155 ± 9146 ± 8
      AdCMV-OB-Rb50 ± 0.227 ± 0.2*161 ± 1252 ± 7*
      Data are the mean ± S.E. of three-four samples. *,p < 0.01 versus 0 leptin.

      Effects of Leptin on SPT Overexpression in fa/fa Islets Overexpressing Wild-type OB-Rb

      To obtain further insight into the mechanism of the exaggerated SPT expression, adenoviral gene transfer technology was employed to overexpress the wild-type leptin receptor in the fa/fa islets (
      • Wang M-Y.
      • Koyama K.
      • Shimabukuro M.
      • Newgard C.B.
      • Unger R.H.
      ). We perfused recombinant adenovirus containing either the OB-Rb cDNA (AdCMV-OB-Rb) or, as a control, the β-galactosidase cDNA (AdCMV-β-gal) into fa/fapancreata and immediately thereafter cultured the islets for 2 days. On the third day, we added either 1 mm FA alone or 1 mm FA plus 20 ng/ml recombinant leptin for an additional 24 h. In the absence of FA, SPT mRNA was virtually identical in OB-Rb-overexpressing and β-galactosidase-overexpressing islets. When 1 mm FA was added, SPT mRNA increased to the same degree in both groups of islets. However, the presence of leptin completely blocked the FA-induced increase in SPT mRNA in OB-Rb-overexpressing islets and actually lowered it by 20% below the base line (p < 0.05) (Fig. 5). Leptin had no effect on SPT in the β-galactosidase-overexpressing control islets.
      Figure thumbnail gr5
      Figure 5Effects on SPT mRNA expression of 1 mm FA with or without 20 ng/ml recombinant leptin in islets of obese diabetic fa/fa ZDF rats that overexpress either OB-Rb (AdCMV-OB-Rb) or β-galactosidase (AdCMV-β-gal). Representative blots are displayed.Bars represent the mean (± S.E.) mRNA ratio of SPT/β-actin (n = 3). *, p < 0.01versus 0 mm FA; †, p < 0.05versus 1 mm FA.

      In Vivo Effects of l-cycloserine on de Novo Ceramide Formation and DNA Fragmentation in Islets

      To determine whetherin vivo SPT blockade would reduce de novoceramide synthesis and spare the beta-cells of fa/fa rats from apoptosis, we treated prediabetic ZDF rats with 25 mg/kg/dayl-cycloserine for 2 weeks. Control rats were given injections of saline. After 2 weeks, all rats were sacrificed and the ceramide content of their islets measured, together with percentage of DNA fragmentation. As shown in Fig. 6, ceramide content of l-cycloserine-treated rats was decreased by 50%, and DNA fragmentation was reduced by 43% (from 14 to 8%) (Fig. 6, A and B). The concentration relationship of FFA to lipoapoptosis in islets of prediabetic rats is shown in Table II.
      Figure thumbnail gr6
      Figure 6A, effects of in vivo l-cycloserine treatment of 7-week-old prediabeticfa/fa rats for 2 weeks on [3H]ceramide formation from [3H]serine in islets. Results in +/+ islets are shown for reference purposes. Bars represent the mean ± S.E. of arbitrary densitometric units (n = 3). *, p < 0.01 versus+/+; *, p < 0.05 versus fa/faislets without cycloserine. B, effects of in vivo l-cycloserine treatment of prediabetic fa/farats on DNA fragmentation in islets. +/+ islets are shown for a reference. Bars represent the mean ± S.E. of arbitrary densitometric units (n = 3). *,p < 0.01 versus +/+; †, p< 0.05 versus fa/fa islets without cycloserine.
      Table IIConcentration relationship of FFA to DNA fragmentation (% total DNA) in islets of prediabetic ZDF rats (n = 3)
      FFA% DNA laddering
      mm
      07.2 ± 0.24
      0.17.0 ± 0.93
      0.513.7 ± 3.23
      119.2 ± 1.18

      DISCUSSION

      The results of this study provide several lines of new evidence that the high ceramide content observed in pancreatic islets of obesefa/fa ZDF rats, and implicated in beta-cell lipoapoptosis and adipogenic diabetes (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ), represents newly synthesized ceramide rather than a product of sphingomyelin hydrolysis. First, expression of the enzyme SPT, the enzyme that catalyzes the condensation of serine and palmitoyl-CoA to form the ceramide precursor, dehydrosphinganine (
      • Wang E.
      • Norred W.P.
      • Bacon C.W.
      • Riley R.T.
      • Merrill Jr., A.H.
      ,
      • Weiss B.
      • Stoffel W.
      ,
      • Hanada K.
      • Hara T.
      • Nishijima M.
      • Kuge O.
      • Dickson R.C.
      • Nagiec M.M.
      ,
      • Merrill Jr., A.H.
      • Jones D.D.
      ,
      • Obeid L.M.
      • Linardic C.M.
      • Karolak L.A.
      • Hannun Y.A.
      ), was strikingly increased in the fa/fa islets with high ceramide content. Second, [3H]ceramide formation from [3H]serine, evidence of SPT enzyme activity, was increased in those islets, confirming our previous report of increased [3H]ceramide from [3H]palmitate (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ). Third, the increased ceramide synthesis was inhibited by Triacsin-C blockade of fatty acyl-CoA synthetase, indicating that the excess ceramide formation is dependent on activation of long-chain fatty acids. Triacsin-C also blocked fatty acid-induced apoptosis (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ), providing further evidence of their linkage. Finally, the competitive inhibitor of SPT, l-cycloserine, blocked the increase in [3H]ceramide, clear evidence that it was newly formed.De novo ceramide synthesis had not previously been identified as a source of ceramide in apoptosis; in cytokine-mediated autoimmune destruction of islets, for example, the ceramide is believed to be derived from sphingomyelin of membranes (
      • Sjoholm A.
      ,
      • Welsh N.
      ), although this has been questioned (
      • Kwon G.
      • Bohrer A.
      • Han X.
      • Corbett J.A.
      • Ma Z.
      • Gross R.W.
      • McDaniel M.L.
      • Turk J.
      ).
      It is noteworthy that in the [3H]serine experiments demonstrating the marked increase in [3H]ceramide formation in fa/fa ZDF islets, most of the label was recovered as sphingomyelin, phosphatidylcholine, and phosphatidylethanolamine, but these products were formed equally in fa/fa and +/+ islets. This strongly indicates that the serine → ceramide segment of sphingomyelin synthetic pathway is increased in fa/fa islets without any discernible post-ceramide differences (Fig. 7).
      Figure thumbnail gr7
      Figure 7A possible mechanism for increased ceramide synthesis in islets of fa/fa ZDF rats. SPT, serine palmitoyltransferase. Arrow widthindicates putative enzyme activity. Only SPT was actually quantified; the width of arrows is inferred from the metabolic studies of Fig. .
      The overexpression of SPT in the fa/fa islets appears to be secondary to the fa mutation, a Gln-269 → Pro substitution in the OB-R (
      • Iida M.
      • Murakami T.
      • Ishida K.
      • Mizuno A.
      • Kuwajima M.
      • Shima K.
      ,
      • Phillips M.S.
      • Liu Q.
      • Hammond H.
      • Dugan V.
      • Hey P.
      • Caskey C.T.
      • Hess J.F.
      ). One consequence of this mutation is total loss of the lipopenic action of leptin on islets (
      • Shimabukuro M.
      • Koyama K.
      • Chen G.
      • Wang M-Y.
      • Trieu F.
      • Lee Y.
      • Newgard C.B.
      • Unger R.H.
      ,
      • Zhou Y-T.
      • Shimabukuro M.
      • Koyama K.
      • Lee Y.
      • Wang M-Y.
      • Trieu F.
      • Unger R.H.
      ). In its presence, lipogenic enzymes are overexpressed (
      • Lee Y.
      • Hirose H.
      • Zhou Y.-T.
      • Esser V.
      • McGarry J.D.
      • Unger R.H.
      ) and enzymes of fatty acid oxidation are underexpressed (
      • Zhou Y-T.
      • Shimabukuro M.
      • Koyama K.
      • Lee Y.
      • Wang M-Y.
      • Trieu F.
      • Unger R.H.
      ), resulting in striking increase in islet lipid content exceeding more than 50 times that of normal (
      • Lee Y.
      • Hirose H.
      • Ohneda M.
      • Johnson J.H.
      • McGarry J.D.
      • Unger R.H.
      ,
      • Lee Y.
      • Hirose H.
      • Zhou Y.-T.
      • Esser V.
      • McGarry J.D.
      • Unger R.H.
      ). Based on the foregoing, it seemed reasonable to consider as a mechanism for SPT overexpression either a loss of direct leptin-induced inhibition or a lipid-mediated up-regulatory effect. Because loss of leptin activity is so tightly coupled to lipid accumulation (
      • Shimabukuro M.
      • Koyama K.
      • Chen G.
      • Wang M-Y.
      • Trieu F.
      • Lee Y.
      • Newgard C.B.
      • Unger R.H.
      ), we could not exclude the former possibility and focused instead on the role of lipids. Fatty acids up-regulated SPT mRNA in both normal+/+ and obese fa/fa ZDF rat islets, but the baseline level of SPT expression and the fatty acid-induced up-regulation was far greater in the fa/fa islets. This correlated well with the much higher base-line TG content and the fatty acid-induced increase in TG content in fa/fa islets.
      Evidence consistent with leptin-mediated regulation of islet lipid was also obtained by overexpressing the wild-type OB-Rb or, as a control, β-galactosidase in the fa/fa islets. This conferred leptin responsiveness to the former islets, whereas leptin had no effect in the β-gal controls. In the OB-Rb-overexpressing islets, 20 ng/ml leptin reduced islet TG content to normal and completely blocked fatty acid-induced up-regulation of SPT mRNA.
      Finally, we tested the possibility that SPT blockade might protect the fat-laden islets of prediabetic ZDF rats from ceramide-mediated lipoapoptosis of beta-cells (
      • Shimabukuro M.
      • Zhou Y-T.
      • Levi M.
      • Unger R.H.
      ), which may be a factor in their diabetes. Prediabetic ZDF rats were treated withl-cycloserine for 2 weeks, and ceramide content and DNA fragmentation were measured. The results suggest that agent was partially effective, reducing [3H]ceramide formation from [3H]serine and DNA fragmentation by ∼50%.

      ACKNOWLEDGEMENTS

      We thank Tagan Ferguson and Kay McCorkle for excellent technical work and Tess Perico for secretarial assistance.

      REFERENCES

        • National Center for Health Statistics
        Newsweek. 1994; 127: 62
        • Lee Y.
        • Hirose H.
        • Ohneda M.
        • Johnson J.H.
        • McGarry J.D.
        • Unger R.H.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10878-10882
        • Lee Y.
        • Hirose H.
        • Zhou Y.-T.
        • Esser V.
        • McGarry J.D.
        • Unger R.H.
        Diabetes. 1997; 46: 408-413
        • Shimabukuro M.
        • Zhou Y-T.
        • Levi M.
        • Unger R.H.
        Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 2498-2501
        • Iida M.
        • Murakami T.
        • Ishida K.
        • Mizuno A.
        • Kuwajima M.
        • Shima K.
        Biochem. Biophys. Res. Commun. 1996; 224: 597-604
        • Phillips M.S.
        • Liu Q.
        • Hammond H.
        • Dugan V.
        • Hey P.
        • Caskey C.T.
        • Hess J.F.
        Nat. Genet. 1996; 13: 18-19
        • Shimabukuro M.
        • Koyama K.
        • Chen G.
        • Wang M-Y.
        • Trieu F.
        • Lee Y.
        • Newgard C.B.
        • Unger R.H.
        Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4637-4641
        • Wang E.
        • Norred W.P.
        • Bacon C.W.
        • Riley R.T.
        • Merrill Jr., A.H.
        J. Biol. Chem. 1991; 266: 14486-14490
        • Weiss B.
        • Stoffel W.
        Eur. J. Biochem. 1997; 249: 239-247
        • Hanada K.
        • Hara T.
        • Nishijima M.
        • Kuge O.
        • Dickson R.C.
        • Nagiec M.M.
        J. Biol. Chem. 1997; 272: 32108-32114
        • Merrill Jr., A.H.
        • Jones D.D.
        Biochim. Biophys. Acta. 1990; 1044: 1-12
        • Obeid L.M.
        • Linardic C.M.
        • Karolak L.A.
        • Hannun Y.A.
        Science. 1993; 259: 1769-1771
        • Naber S.P.
        • McDonald J.M.
        • Jarett L.
        • McDaniel M.L.
        • Ludvigsen C.W.
        • Lacy P.E.
        Diabetologia. 1980; 19: 439-444
        • Shimabukuro M.
        • Ohneda M.
        • Lee Y.
        • Unger R.H.
        J. Clin. Invest. 1997; 100: 290-295
        • Bligh E.G.
        • Dyer W.J.
        Can. J. Biochem. Physiol. 1959; 37: 911-914
        • Wang M-Y.
        • Koyama K.
        • Shimabukuro M.
        • Newgard C.B.
        • Unger R.H.
        Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 714-718
        • Duke R.C.
        • Sellins C.B.
        Kaplan J.G. Cellular Basis of Immune Modulation. Liss, New York1989: 311-314
        • Hopcroft D.W.
        • Mason D.R.
        • Scott R.S.
        Horm. Metab. Res. 1985; 17: 559-561
        • Noel R.J.
        • Antinozzi P.A.
        • McGarry J.D.
        • Newgard C.B.
        J. Biol. Chem. 1997; 272: 18621-18627
        • Antinozzi P.A.
        • Segall L.
        • Prentki M.
        • McGarry J.D.
        • Newgard C.B.
        J. Biol. Chem. 1998; 273: 16146-16154
        • Sundaram K.S.
        • Lev M.
        J. Neurochem. 1984; 42: 577-581
        • Sjoholm A.
        FEBS Lett. 1995; 367: 283-286
        • Welsh N.
        J. Biol. Chem. 1996; 271: 8307-8312
        • Kwon G.
        • Bohrer A.
        • Han X.
        • Corbett J.A.
        • Ma Z.
        • Gross R.W.
        • McDaniel M.L.
        • Turk J.
        Biochim. Biophys. Acta. 1996; 1300: 63-72
        • Zhou Y-T.
        • Shimabukuro M.
        • Koyama K.
        • Lee Y.
        • Wang M-Y.
        • Trieu F.
        • Unger R.H.
        Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6386-6390