Development of an animal model of chronic alcohol-induced pancreatitis in the rat.

This study was designed to develop an animal model of alcoholic pancreatitis and to test the hypothesis that the dose of ethanol and the type of dietary fat affect free radical formation and pancreatic pathology. Female Wistar rats were fed liquid diets rich in corn oil (unsaturated fat), with or without a standard or high dose of ethanol, and medium-chain triglycerides (saturated fat) with a high dose of ethanol for 8 wk enterally. The dose of ethanol was increased as tolerance developed, which allowed approximately twice as much alcohol to be delivered in the high-dose group. Serum pancreatic enzymes and histology were normal after 4 wk of diets rich in unsaturated fat, with or without the standard dose of ethanol. In contrast, enzyme levels were elevated significantly by the high ethanol dose. Increases were blunted significantly by dietary saturated fat. Fibrosis and collagen alpha1(I) expression in the pancreas were not detectable after 4 wk of enteral ethanol feeding; however, they were enhanced significantly by the high dose after 8 wk. Furthermore, radical adducts detected by electron spin resonance were minimal with the standard dose; however, the high dose increased carbon-centered radical adducts as well as 4-hydroxynonenal, an index of lipid peroxidation, significantly. Radical adducts were also blunted by approximately 70% by dietary saturated fat. The animal model presented here is the first to demonstrate chronic alcohol-induced pancreatitis in a reproducible manner. The key factors responsible for pathology are the amount of ethanol administered and the type of dietary fat.

Corn oil rapidly activates nuclear factor-kappaB in hepatic Kupffer cells by oxidant-dependent mechanisms.

N-6 polyunsaturated fatty acids (N-6 PUFAs), major constituents of corn oil and natural ligands for peroxisome proliferator-activated receptors, increase the rate of growth of established tumors. It has been proposed that chemical peroxisome proliferators increase hepatocyte proliferation by mechanisms involving activation of nuclear factor-kappaB (NF-kappaB) and production of low levels of tumor necrosis factor alpha (TNFalpha) by Kupffer cells; however, how N-6 PUFAs are involved in increased cell proliferation in liver is not well understood. Here, the hypothesis that N-6 PUFAs increase production of mitogens by activation of Kupffer cell NF-kappaB was tested. A single dose of corn oil (2 ml/kg, i.g.), but not olive oil or medium-chain triglycerides (saturated fat), caused an approximately 3-fold increase in hepatocyte proliferation. Similarly, when activity of NF-kappaB in whole rat liver or isolated hepatocytes and Kupffer cells was measured at various time intervals for up to 36 h, only corn oil activated NF-kappaB. Corn oil increased NF-kappaB activity approximately 3-fold 1-2 h after treatment exclusively in the Kupffer cell fraction. In contrast, increases were small and only occurred after approximately 8 h in hepatocytes. The activation of NF-kappaB at 2 h and increases in cell proliferation at 24 h due to corn oil were prevented almost completely when rats were pretreated for 4 days with either dietary glycine (5% w/w), an agent that inactivates Kupffer cells, or the NADPH oxidase inhibitor, diphenyleneiodonium (s.c., 1 mg/kg/day). Furthermore, arachidonic acid (100 microM) activated superoxide production approximately 4-fold when added to isolated Kupffer cells in vitro. This phenomenon was not observed with oleic or linoleic acids. Interestingly, a single dose of corn oil increased TNFalpha mRNA nearly 2-fold 8 h after treatment. It is concluded that corn oil rapidly activates NF-kappaB in Kupffer cells via oxidant-dependent mechanisms. This triggers production of low levels of TNFalpha which is mitogenic in liver and promotes growth of hepatocytes.

Differential regulation of hepatocyte DNA synthesis by cAMP in vitro in vivo.

Adenosine 3',5'-cyclic monophosphate (cAMP) prevents epidermal growth factor (EGF)-induced DNA synthesis in many types of cultured cells, including hepatocytes, but its effects on cellular proliferation in vivo are unknown. This study compares the effects of supplemental cAMP on hepatocyte proliferation induced in vivo by 70% partial hepatectomy (PH) and in vitro by EGF and determines the effects of cAMP on AP-1, a family of growth-regulatory transcription factors, and the kinase cascades that normally activate AP-1. Although injection of dibutyryladenosine 3',5'-cyclic monophosphate (30 mg/kgip) at the time of PH increased liver cAMP concentrations at least 100-fold for several hours, it did not inhibit hepatic incorporation of [3H]thymidine or proliferating cell nuclear antigen expression 24 h after PH. cAMP treatment led to a complete inhibition of extracellular signal-related kinase (ERK) activity and transiently reduced NH2-terminal Jun nuclear kinase (JNK) activity after PH but did not decrease the expression of c-jun mRNA or protein. Consistent with the known cAMP stimulation of jun-B in cultured cells, cAMP treatment increased jun-B mRNA, protein, and DNA binding activity post-PH. Surprisingly, cAMP treatment enhanced Raf kinase activity after PH in rats. In primary hepatocyte cultures, supplemental cAMP inhibited JNK and ERK activity, total AP-1 and c-Jun transcriptional activities, and DNA synthesis. Thus elevated cAMP inhibited ERK and JNK activity in culture and in vivo and inhibited hepatocyte proliferation in culture but not in vivo. This suggests that in vivo mechanisms compensate for cAMP inhibition of certain growth-related signaling cascades and emphasizes potential risks of extrapolating from simple cell culture systems to explain physiology in intact animals.

Expression of intracellular adhesion molecule 1 by activated hepatic stellate cells.

The hepatic stellate cell (HSC), following a fibrogenic stimulus, is transformed from a quiescent to an activated cell. HSC activation results in numerous changes in cellular morphology, cellular metabolism, and in the pattern of gene expression. Many of the changes that are observed in activated HSCs in animal models of hepatic fibrosis are also seen when these cells are activated by culturing on plastic. These changes include morphological changes to a myofibroblast-like cell with the appearance of smooth muscle alpha-actin, a loss of the retinol stores, an increase in the rough endoplasmic reticulum, and increases in extracellular matrix production, including a dramatic increase in type I collagen. To identify additional genes that are induced or suppressed during HSC activation, we used the differential polymerase chain reaction (PCR) display technique. Using this technique, we isolated a complementary DNA (cDNA) fragment for the intercellular adhesion molecule 1 (ICAM-1). Northern blotting confirmed that the ICAM-1 messenger RNA (mRNA) was expressed in HSCs activated by culture, but not in quiescent, freshly isolated HSCs. The presence of ICAM-1 protein was demonstrated in culture-activated HSCs, but not in quiescent cells by Western blot analysis and immunohistochemical staining. A functional assay was performed, demonstrating that lymphocytes will adhere to activated HSCs and that treatment of these cells with tumor necrosis factor alpha (TNF-alpha) increases lymphocyte adherence. Furthermore, ICAM-1 mRNA levels were increased in HSCs activated in rats in vivo after 1 week of bile duct ligation (BDL). Together, these data indicate that ICAM-1 expression is induced following HSC activation and that the HSC may have a direct role in the transmigration of leukocytes from the hepatic sinusoid to sites of tissue damage during the inflammatory response in the liver.

Ferrochelatase cDNA delivered by adenoviral vector corrects biochemical defect in protoporphyric cells.

Protoporphyria is generally an autosomal dominant disease characterized genetically by mutations in the ferrochelatase gene. The interaction between the wild-type and mutant ferrochelatase protein is unknown. The aim of this study was to evaluate the ability to correct the enzymatic and biochemical defects in cells from patients with protoporphyria, using a replication-defective human adenovirus for gene transfer. Overexpression of ferrochelatase was accomplished by construction of a vector in which expression of the wild-type ferrochelatase cDNA was driven by the constitutive cytomegalovirus (CMV) promoter, introduction and packaging of the cDNA into human adenovirus dl309, and transduction of normal and protoporphyric fibroblasts. Fibroblasts from controls and patients were infected with the ferrochelatase adenovirus or a control adenovirus and assayed for ferrochelatase activity and the accumulation of protoporphyrin upon challenge with the precursor delta-aminolevulinic acid (ALA). At a multiplicity of infection (moi) of 10, greater than 85% of both the wild-type and protoporphyric fibroblasts were infected. The recombinant adenovirus increased the ferrochelatase protein content and activity in the wild-type and protoporphyric fibroblasts with equal efficiency. Therefore, the presence of the mutant ferrochelatase protein did not inhibit the ferrochelatase activity expressed by the transgene.

Mithramycin selectively inhibits collagen-alpha 1(I) gene expression in human fibroblast.

The products of the collagen-alpha 1(I) and -alpha 2(I) genes form the triple helical molecule collagen type I, which constitutes the major ECM protein in tissue fibrosis. The collagen-alpha 1(I) gene is mainly transcriptionally regulated, and its promoter activity depends on the interaction of the transcription factors NF-I and Sp1 with a tandem repeat of evolutionary conserved NF-I/Sp1 switch elements. An increased affinity of Sp1 to these elements has been observed in experimental liver fibrosis. Here, we demonstrate that the DNA binding drug mithramycin displays a high affinity binding to the GC-rich elements in the collagen-alpha 1(I) promoter as measured by DNAse I protection and gel retardation assays. Mithramycin interferes with Sp1 but not with NF-I binding to these sites. At a concentration of 100 nM, mithramycin efficiently reduces basal and TGF-beta-stimulated alpha 1(I) gene expression in human primary fibroblasts. The transcriptional activity and mRNA steady state levels of other genes, including the collagenase gene, as well as the growth rate of fibroblasts remained unchanged on exposure to this drug. Taken together, our results indicate that the transcriptional activity of the type I collagen gene highly depends on its GC-rich regulatory elements, and further, that these elements can be differentially blocked, thereby changing the balance between ECM structural and degrading gene activities in human fibroblasts.

Study of factors causing excess protoporphyrin accumulation in cultured skin fibroblasts from patients with protoporphyria.

The activity of heme synthetase, which catalyzes the chelation of ferrous iron to protoporphyrin to form heme, is deficient in sonicates of skin fibroblasts cultured from patients with protoporphyria. During culture in Eagle's medium supplemented with fetal calf serum, these cells do not accumulate protoporphyrin, however. This may be due to a minimal requirement for heme synthesis, since glycine is incorporated into heme at a low rate which is similar to that in normal fibroblasts. In addition, the activity of delta-aminolevulinic acid (ALA) synthetase, the first and rate-limiting enzyme of heme biosynthesis which catalyzes the formation of ALA from glycine, is normal in lysates of the fibroblasts. Cultured fibroblasts were therefore incubated with ALA in order to bypass the rate-limiting step of heme biosynthesis. In the presence of 25 muM iron, protoporphyrin was detected in protoporphyria cell lines when the concentration of ALA in the medium reached 50 muM, but not in normal lines. As the concentration of ALA was increased above 50 muM, all lines accumulated protoporphyrin. However, the amount was 2-3 times more in cultured fibroblasts from patients with protoporphyria, reflecting their deficiency of heme synthetase activity. When iron was not added to the medium, protoporphyrin accumulated to a similar degree in normal and protoporphyria fibroblasts; this was significantly more than that in the presence of iron. These studies indicate that excessive protoporphyrin accumulation in protoporphyria, which is due principally to deficient heme synthetase activity, may be modified by the rate of ALA formation in heme-producing tissues, and by the availability of iron.