Studies in rodents with the dipeptidyl peptidase-4 inhibitor vildagliptin to evaluate possible drug-induced pancreatic histological changes that are predictive of pancreatitis and cancer development in man.

AIM: The present report summarizes rodent studies with vildagliptin, relevant to predicting pancreatitis or pancreatic cancer in man. METHODS: As part of the regulatory development program for vildagliptin, a rodent toxicity program included two 104-week rodent (mouse and rat) carcinogenicity studies that were conducted according to guidelines assigned in Food and Drug Administration's Draft Guidance for Industry. RESULTS: Vildagliptin exposure in animals was evaluated for its effects on endocrine and exocrine pancreas. Two-year carcinogenicity studies were conducted in rats at oral doses up to 900 mg/kg (approximately 200 times the human exposure at the maximum recommended dose) and in mice at oral doses up to 1000 mg/kg (up to 240 times the human exposure at the maximum recommended dose). The results from these studies show the expected preservation and growth of the endocrine ß-cells with no significant findings in the exocrine acinar pancreas. There was no evidence of inflammatory infiltrates characteristic of pancreatitis, no palpable mass detection based on gross examination or any microscopic findings indicative of pancreatic islet cell (endocrine), acinar cell (exocrine) or ductal (exocrine) neoplasia in rat or mouse. CONCLUSIONS: Evaluation of vildagliptin in 2-year preclinical carcinogenicity studies in both rats and mice indicates that while vildagliptin results in pharmacological benefits to the endocrine pancreas, this was not associated with any evidence of pancreatitis, pancreatic islet cell, acinar cell or ductal neoplasia. These data predict no increased risk of pancreatic cancer in man.

Trans-repressor activity of nuclear glycosaminoglycans on Fos and Jun/AP-1 oncoprotein-mediated transcription.

Heparin blocks the phorbol ester-induced progression of nontransformed cells through the G0/G1 phase (Wright, T.C., L.A. Pukac, J.J. Castellot, M.J. Karnovsky, R.A. Levine, H.-Y. Kim-Park, and J. Campisi. 1989. Proc. Natl. Acad. Sci. USA. 86: 3199-3203) or G1 to S phase (Reilly, C. F., M. S. Kindy, K. E. Brown, R. D. Rosenberg, and G. E Sonenshein. 1989. J. Biol. Chem. 264:6990-6995) of the cell cycle. Cell cycle arrest was associated with decreased levels of stage-specific mRNAs suggesting transcriptional regulation of cell growth. In the present report, we show that heparin selectively repressed TPA-inducible AP-1-mediated gene expression. Heparin-induced trans-repression was observed in primary vascular smooth muscle cells, as well as in the transformed HeLa cell line and in nondifferentiated F9 teratocarcinoma cells. Inhibition of AP-1-mediated trans-activation occurred with heparin and pentosan polysulfate but not with chondroitin sulfate A or C. Heparin-binding peptides or heparitinase I addition to nuclear lysates of heparin-treated cells allowed enhanced recovery of endogenous AP-1-specific DNA binding activity. We propose a model in which nuclear glycosaminoglycans play a trans-regulatory role in altering the patterns of inducible gene expression.

Dimers, leucine zippers and DNA-binding domains.

Transcription factors can be divided into classes on the basis of their mode of interaction with the target promoter sequence. Different protein domains responsible for DNA recognition have been identified. In this review we discuss the leucine zipper structure, which has been found in several nuclear factors, including the oncoproteins Fos and Jun. Structural considerations are summarized to help understand how dimerization is mediated by the leucine zipper and how this is the prerequisite for optimal target DNA recognition by the adjacent basic domains.

Fos, Jun and CREB basic-domain peptides have intrinsic DNA-binding activity enhanced by a novel stabilizing factor.

Transcription factors with a 'leucine zipper' (LZ) domain bind to DNA cis-elements, often with palindromic structures. DNA-binding by such factors requires the presence of highly basic regions in the proteins found adjacent to the LZ domain. In order to determine if the observed DNA-binding specificity is programmed by the basic region itself, we developed a competitive binding assay to compare relative binding affinities of synthetic peptides to specific promoter elements. In this report we demonstrate that the basic domains of the oncoproteins Fos, Jun and the transcription factor CREB, possess the structural information necessary to compete for promoter-specific binding. To study the relative binding affinity of the basic motifs to specific promoter elements, we used synthetic peptides to compete for intrinsic Fos/Jun and CREB DNA-binding activity present in HeLa cell nuclear extracts. These studies demonstrate that the basic peptides of both Fos and Jun have higher affinity for the TPA responsive element (TRE) and the cAMP responsive element (CRE) relative to the corresponding peptide for CREB. The peptides showed virtually no affinity for either Sp1 or octamer consensus promoter sequences, demonstrating that these basic peptides also retained their promoter selectivity. We further demonstrate that the conserved dipolar arrangement of two basic amino acid clusters is required for selective, competitive binding. A second conserved feature critical for competitive binding relates to the distance separating the dipolar basic clusters. Our competition binding assay has also helped to identify a novel protein factor, termed ABP (auxilliary bridging protein), which interacts to stabilize Fos and Jun basic domain binding-interaction with specific promoter elements. The data suggest a mechanism in which ABP acts to promote more stable protein-DNA complexes.

Human hepatoma (HepG2) cells secrete a single 65 K dalton triglyceride lipase immunologically identical to postheparin plasma hepatic lipase.

A triacylglycerol lipase was isolated from the culture medium of HepG2 human hepatoma cells and its properties were compared to hepatic triglyceride lipase (H-TGL) from human postheparin plasma. The HepG2 cell enzyme bound to heparin-Sepharose, was eluted with 1 M NaCl and was not inhibited by 1 M salt. Western-blotting of the fractions from the heparin-Sepharose column with a monoclonal antibody prepared against postheparin plasma H-TGL and which binds to an epitope in the carboxyl-terminus of H-TGL gave a single immunoreactive protein band of 65 kDa. This finding of immunochemical identity was confirmed with polyclonal antibodies prepared against synthetic peptides of H-TGL corresponding to amino acid residues 82-94 near the amino-terminus and residues 468-477, the carboxyl-terminus of the enzyme. We conclude that HepG2 cells secrete a single triacylglycerol lipase with molecular weight properties and immunological characteristics identical to post-heparin plasma H-TGL.

Cholesteryl ester analogs inhibit cholesteryl ester but not triglyceride transfer catalyzed by the plasma cholesteryl ester-triglyceride transfer protein.

A lipid transfer protein complex (LTC), purified from human plasma by immunoaffinity chromatography, catalyzed the interlipoprotein transfer of cholesteryl esters (CE) and triglycerides (TG). The CE transfer activity of LTC was governed by the structure of the CE. Incubation of LTC with long chain CE both activated and stabilized LTC. Short chain CE also enhanced the CE and TG transfer activity of LTC during the initial time of incubation. However, LTC's incubation with short chain CE induced a subsequent and time-dependent loss of CE transfer activity without concomitant loss of TG transfer activity. The data indicate that the CE and TG transfer activity of LTC can be regulated independently.

Concentration-dependent antioxidant activity of probucol in low density lipoproteins in vitro: probucol degradation precedes lipoprotein oxidation.

The ability of probucol, a lipid-lowering drug with antioxidant properties, to prevent the Cu2+-induced oxidation of human plasma low density lipoproteins (LDL) was examined as a function of the concentration of probucol in LDL. In the absence of probucol, 3 microM Cu2+ induced half-maximal LDL lipid oxidation, as determined by the formation of thiobarbituric acid reactive substances (TBARS). Oxidation was associated with a loss of apolipoprotein B-100 and the appearance of higher molecular weight forms of the protein. In the presence of 0.6 mol% probucol (relative to phospholipid) and with 3 microM Cu2+, the time required to obtain half-maximal LDL lipid oxidation increased from 130 to 270 min and was explained by an increase in the lag time prior to LDL lipid oxidation. Once rapid oxidation of LDL had begun, the rate of TBARS formation was similar to that for LDL containing no probucol. At a probucol concentration of 4.2 mol%, the antioxidant prevented the oxidation of LDL-lipids. The delay in Cu2+-induced LDL oxidation with probucol corresponded to the time required for free radical-mediated processes to convert probucol to a spiroquinone and a diphenoquinone. These in vitro findings suggest that the potent antioxidant property of probucol is directly related to the amount of drug in the LDL particle and may have relevance to its antiatherosclerotic effects observed in vivo.

Plasma cholesteryl ester-triglyceride transfer protein. The catalytic domain is a low molecular weight proteolipid.

The lipid transfer protein complex (LTC) isolated from human plasma by immunoaffinity chromatography transfers cholesteryl esters (CE), triglycerides, and phosphatidylcholine (PC) between lipoproteins in vitro. The molecular weight of this lipid transfer catalyst in sodium dodecyl sulfate-polyacrylamide gels was 65,000. When resolved on a gel filtration column by high performance liquid chromatography (HPLC), LTC was composed of fractions of high (greater than 150,000) to low (18,000) molecular weight, although sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of each fraction revealed bands at Mr 65,000 (major) and 52,000 (minor). The CE and triglyceride transfer activity of the low Mr HPLC fraction (1049 nmol of triglyceride/mg/h and 244 nmol of CE/mg/h) was significantly greater than that of the high Mr HPLC fraction (15-27 nmol of triglyceride/mg/h and 20-30 nmol of CE/mg/h). The PC transfer activity of the HPLC fractions was not determined. LTC proteins were separated by dialysis in acidified chloroform:methanol solution into dialysand and dialysate proteins. The dialysate contained a low Mr proteolipid, designated the catalytic domain Cd, which catalyzed CE and triglyceride transfer at equivalent rates (11.0 versus 9.5 mumol/mg/h, respectively). PC transfer activity was approximately 10% of these levels (1.5 mumol/mg/h). The dialysand consisted of a protein, designated the transfer protein TP, which facilitated CE (3.4 mumol/mg/h) preferentially over triglyceride and PC (1.0 mumol/mg/h) transfer, and a catalytically inactive protein, designated the heparin-binding domain Hd. We propose a model of the LTC protein (based on catalytic activities, monoclonal antibody reactivities, and heparin-binding capacities of the isolated proteins) in which both Hd (approximately 13 kDa) and Cd (approximately 3 kDa) originate from a single lipid transfer protein, TP.

Immunoaffinity purification of the lipid transfer protein complex directly from human plasma.

The human cholesteryl ester (CE) and triglyceride (TG) exchange protein (denoted LTC or lipid transfer complex) was isolated in a single step from plasma using immunoaffinity batch extraction. Antibodies were raised against two preparations of conventionally purified LTC. LTC-I and LTC-II (purified 20,000-fold and 3500-fold, respectively) were used as immunogens. The antiLTC antibodies were isolated by anion-exchange chromatography and coupled to Affi-Gel 10. Chromatography of plasma on antiLTC Affi-Gel removed all of the CE and TG transfer activity. Moreover, LTC prepared from both antiLTC-I and antiLTC-II-Affi-Gel matrices were identical when analyzed by sodium dodecyl sulfate-polyacrylamide gel LTC electrophoresis. LTC exhibited two protein bands of Mr (apparent) 67,000 and 58,000 and a broad, faintly staining region at greater than 150,000. Analysis of LTC by immunoblotting indicated that both antiLTC-I and antiLTC-II antibodies recognized the same LTC proteins. Isoelectric focussing of LTC gave two pI values, 5.2 and 8.7. These data suggest that LTC is a complex of specific proteins and perhaps lipid. Specific CE and TG exchange activities of immunoaffinity-purified LTC were comparable, although the activities were low with respect to that of the antigen used to generate antiLTC-I. This is not due to contamination of LTC by albumin, lecithin:cholesterol acyltransferase, or apolipoproteins AI, AII, B, CIII, D, or E.

Heparin induces the expression of hepatic triglyceride lipase in a human hepatoma (HepG2) cell line.

Hepatic triglyceride lipase (H-TGL) is a key lipolytic enzyme in the metabolism of human plasma high density lipoproteins. The enzyme is bound to glycosaminoglycans on endothelial cells in the liver and is immediately released into the circulation after heparin administration. In addition to releasing H-TGL, heparin-like glycosaminoglycans have also been shown to suppress hepatocyte proliferation and to alter tissue-specific gene expression. In the present study, the effects of heparin exposure on the secretion of H-TGL were examined in a human parenchymal hepatoma (HepG2) cell line. The addition of heparin to serum-supplemented medium induced the secretion of H-TGL in a time- and concentration-dependent manner. At 5.4 micrograms/ml heparin, H-TGL levels, as determined by triacyglycerol hydrolase activity, increased 7-fold after a 44-h incubation. Heparin exposure decreased intracellular H-TGL activity from 21.3 to 4.8 nmol of oleic acids released/h/10(8) cells and increased enzyme activity in the medium from 16.2 to 165.3 nmol of oleic acids released/h/10(8) cells. The heparin-induced secretion of H-TGL was associated with increased levels of H-TGL-specific mRNA. The addition of actinomycin D or cycloheximide reversed the heparin-induced increase in H-TGL activity and mRNA. Heparin treatment did not increase the level of actin mRNA suggesting that elevated H-TGL mRNA is due to enhanced tissue-specific expression of H-TGL. Expression of apolipoprotein E, another protein involved in lipoprotein metabolism, also showed induced levels of mRNA by heparin but to a lesser extent than that for H-TGL. We conclude that heparin stimulates the de novo synthesis of H-TGL in liver parenchymal cells in vivo by influencing both transcriptional and post-transcriptional events.

DNA binding of Jun and Fos bZip domains: homodimers and heterodimers induce a DNA conformational change in solution.

We constructed plasmids encoding the sequences for the bZip modules of c-Jun and c-Fos which could then be expressed as soluble proteins in Escherichia coli. The purified bZip modules were tested for their binding capacities of synthetic oligonucleotides containing either TRE or CRE recognition sites in electrophoretic mobility shift assays and circular dichroism (CD). Electrophoretic mobility shift assays showed that bZip Jun homodimers and bZip Jun/Fos heterodimers bind a collagenase-like TRE (CTGACTCAT) with dissociation constants of respectively 1.4 x 10(-7) M and 5 x 10(-8) M. As reported earlier [Patel et al. (1990) Nature 347, 572-575], DNA binding induces a marked change of the protein structure. However, we found that the DNA also undergoes a conformational change. This is most clearly seen with small oligonucleotides of 13 or 14 bp harboring respectively a TRE (TGACTCA) or a CRE (TGACGTCA) sequence. In this case, the positive DNA CD signal at 280 nm increases almost two-fold with a concomitant blue-shift of 3-4 nm. Within experimental error the same spectral changes are observed for TRE and CRE containing DNA fragments. The spectral changes observed with a non-specific DNA fragment are weaker and the signal of free DNA is recovered upon addition of much smaller salt concentrations than required for a specific DNA fragment. Surprisingly the spectral changes induced by Jun/Jun homodimers are not identical to those induced by Jun/Fos heterodimers. However, in both cases the increase of the positive CD band and the concomitant blue shift would be compatible with a B to A-transition of part of the binding site or a DNA conformation intermediate between the canonical A and B structures.

Kinetics of plasma protein-catalyzed exchange of phosphatidylcholine and cholesteryl ester between plasma lipoproteins.

A lipid transfer complex (LTC) isolated from human plasma catalyzes equimolar exchange of cholesteryl ester and phosphatidylcholine between low density (LDL) and high density (HDL) plasma lipoproteins. Activation parameters for LTC-catalyzed exchange of neutral and polar lipid are equal and are not influenced by the degree of purity of the catalyst. Activation parameters for exchange of both cholesteryl ester and phosphatidylcholine are influenced by the extent of saturation of phosphatidylcholine fatty acyl groups. The activation parameters also depend on the amount of HDL present in the assay. The flux rates of lipid exchange depend on the concentration of both LDL and HDL. At constant HDL concentration, flux rates become independent of LDL concentration when the ratio of [LDL]:[HDL] exceeds 9:1 (based on cholesteryl ester); at constant LDL concentration, facilitated LDL,HDL lipid exchange is inhibited at high HDL concentration, suggesting preferential HDL,HDL exchange. Analysis of the dependence of initial lipid exchange rate on LDL concentration at two constant HDL concentrations suggests that, in the reaction pathway, LTC mediates a productive collision (ternary complex) between LDL and HDL. A kinetic mechanism consistent with the data is one in which lipid exchange occurs in a ternary complex consisting of LTC, HDL and LDL. At low HDL concentration, this complex is formed by a random sequential route; at high HDL concentration, the mechanism is ordered sequential since the reactants are an LTC-HDL complex and LDL.

Induction of vascular cell adhesion molecule-1 expression by IL-4 in human aortic smooth muscle cells is not associated with increased nuclear NF-kappaB levels.

Vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1) are upregulated in vascular endothelial and smooth muscle cells by cytokines produced at sites of inflammation. The cytokine profile for induction of VCAM-1, however, is different for the two cell types. Tumor necrosis factor-alpha (TNF-alpha) induced both VCAM-1 and ICAM-1 expression in human umbilical vein endothelial cells (HUVECs; ED50 approximately 300 and 30 U/ml, respectively). TNF-alpha and interleukin-1beta (IL-1beta) stimulated cell surface ICAM-1 expression, but not VCAM-1 expression, in human aortic smooth muscle cells (HASMCs). Conversely, IL-4 was a potent VCAM-1 inducer in HASMCs (ED50 approximately 100 pg/ml) but did not induce ICAM-1 expression. Nuclear extracts from IL-4-treated cells were compared with untreated cells for relative nuclear factor-kappa B (NF-kappaB) levels by using an electrophoretic mobility shift assay and surface plasmon resonance techniques. No significant increase in nuclear NF-kappaB DNA binding activity was detected in IL-4-treated HASMCs by either method of analysis. IL-1beta and TNF-alpha stimulated nuclear NF-kappaB levels by about fourfold and fivefold, respectively, in HASMCs. The antioxidant pyrrolidine dithiocarbamate (PDTC) similarly inhibited VCAM-1 upregulation in HASMCs incubated with IL-4 and in HUVECs incubated with TNF-alpha (IC50s of 25 and 40 microM, respectively). These data suggest that a significant increase in nuclear NF-kappaB levels is not necessary or sufficient for VCAM-1 upregulation in HASMCs and does not determine the relative sensitivity to inhibition of gene expression by PDTC.

Differential regulation of hepatic triglyceride lipase and 3-hydroxy-3-methylglutaryl-CoA reductase gene expression in a human hepatoma cell line, HepG2.

The level of hepatic triglyceride lipase (H-TGL) synthesis and secretion was examined in response to changes in cholesterol biosynthesis in the human hepatoma cell line HepG2. Cells were first fed a lipoprotein-deficient serum-supplemented medium to eliminate exogenous cholesterol. Mevinolin, a 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase inhibitor, was then added at a concentration (37 microM) which inhibited cholesterol biosynthesis by greater than 85% and decreased total cell cholesterol from 36.1 to 27.4 micrograms/ml of cell protein. Mevinolin treatment caused a 4.9 +/- 0.8-fold increase in the amount of H-TGL activity secreted into the medium, a 1.8 +/- 0.4-fold rise in H-TGL-specific mRNA, and a concurrent 14-fold increase in HMG-CoA reductase mRNA. Addition of 1 mM mevalonic acid to normal or mevinolin-treated cells raised the cellular cholesterol content and decreased the amount of secreted H-TGL activity to levels below control values. Mevalonic acid also prevented mevinolin-induction of H-TGL and HMG-CoA reductase mRNA, suggesting a common regulatory step for H-TGL and HMG-CoA reductase. Exposure of cells to mevinolin and 25-hydroxycholesterol together resulted in a marked repression of HMG-CoA reductase mRNA levels, whereas these conditions further enhanced the secretion of H-TGL activity and the expression of H-TGL mRNA. These results demonstrate a differential role for 25-hydroxycholesterol in the regulation of H-TGL and HMG-CoA reductase expression.

Purification and characterization of apolipoprotein J.

Apolipoprotein J (apoJ), a unique 70-kDa component of high density lipoproteins in human plasma, consists of two disulfide-linked subunits designated apoJ alpha (34-36 kDa), and apoJ beta (36-39 kDa) which share pI values of 4.9-5.4 and which are recognized by a monoclonal antibody (mAb) 11. ApoJ and its subunits were purified to homogeneity from plasma by a combination of immunoaffinity chromatography, using mAb11 linked to Affi-Gel, and reverse-phase high performance liquid chromatography. ApoJ alpha and apoJ beta are both glycoproteins. When deglycosylated, the molecular mass of apoJ alpha is 24 kDa and that of apoJ beta is 28 kDa, suggesting that approximately 30% of the mass of each subunit is carbohydrate. The amino acid compositions of apoJ alpha and apoJ beta are very similar; however, the sequences of the first 30-amino acid residues are distinct. A comparison of peptide maps suggests that apoJ alpha and apoJ beta are not identical but share limited regions of homology. This possibility is supported by immunochemical data. Five additional mAb specific for apoJ were characterized. One of the mAb, like mAb11, reacts with both apoJ alpha and apoJ beta; the others react with apoJ alpha only. All mAb, including those which recognize both apoJ alpha and apoJ beta and those which recognize apoJ alpha only, immunoprecipitate a approximately 50-kDa protein synthesized from a liver mRNA template translated in a rabbit reticulocyte lysate. We propose that the apoJ alpha and apoJ beta subunits, which have limited homology, are derived by proteolytic cleavage of a common precursor.

Human hepatic triglyceride lipase expression reduces high density lipoprotein and aortic cholesterol in cholesterol-fed transgenic mice.

We have produced a line of transgenic mice expressing human hepatic triglyceride lipase (hH-TGL) to examine the in vivo effects of hepatic lipase expression on high density lipoprotein catabolism. Activation of metallothionine I promoter-hH-TGL cDNA transgene produced high levels of lipase mRNA in liver, heart, and kidney and elevated enzyme activity as assayed in post-heparin plasma. In a series of hyperlipidemic diet studies in which zinc was included in the diet to induce the transgene, hH-TGL expression was associated with a 34% lowering of plasma HDL-cholesterol levels (p < 0.01) when compared with animals on the same hyperlipidemic diet without zinc. This lowering of HDL cholesterol was paralleled by a decrease in total cholesterol and a decrease in HDL particle size. SDS-polyacrylamide gel electrophoresis analysis of the smaller HDL particles revealed that apolipoprotein AI was still the major apoprotein associated with the HDL. Quantitative analysis of abdominal aortic cholesterol content from the same animals suggests that the observed changes in plasma HDL by hH-TGL over-expression correlated with a decrease in the accumulation of aortic cholesterol (42%, p < 0.01). These data support the hypothesis that hH-TGL mediates a non-receptor pathway for the clearance of cholesterol from the plasma compartment.

Isolation and cDNA sequence of human postheparin plasma hepatic triglyceride lipase.

Hepatic triglyceride lipase (H-TGL) was isolated from human postheparin plasma by column chromatography on heparin-Sepharose and phenyl-Sepharose and immunoaffinity chromatography with monoclonal antibodies. The purified enzyme had an apparent molecular weight of 65,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and an amino-terminal sequence of Leu-Gly-Gln-Ser-Leu-Lys-Pro-Glu. Partial amino acid sequences of seven cyanogen bromide peptides were obtained. A human hepatoma cDNA library was screened with synthetic oligonucleotides derived from the partial protein sequence. The cloned H-TGL cDNA of 1569 nucleotides predicts a mature protein of 477 amino acids plus a leader sequence of 22 amino acids. Blot hybridization analysis of poly(A)+ mRNA with a putative H-TGL cDNA clone gave a single hybridizing band of 1.7 kilobases. The protein contains four consensus N-glycosylation sequences based on the cDNA sequence. Comparison of the enzyme sequence with that of other lipases reveals highly conserved sequences in regions of putative lipid and heparin binding. The carboxyl terminus of H-TGL contains a highly basic sequence which is not reported to be present in rat H-TGL or other members of the lipase gene family.