A new mouse liver-specific gene, encoding a protein homologous to human antimicrobial peptide hepcidin, is overexpressed during iron overload

Considering that the development of hepatic lesions related to iron overload diseases might be a result of abnormally expressed hepatic genes, we searched for new genes up-regulated under the condition of iron excess. By suppressive subtractive hybridization performed between livers from carbonyl iron-overloaded and control mice, we isolated a 225-base pair cDNA. By Northern blot analysis, the corresponding mRNA was confirmed to be overexpressed in livers of experimentally (carbonyl iron and iron-dextran-treated mice) and spontaneously (beta(2)-microglobulin knockout mice) iron-overloaded mice. In addition, beta(2)-microglobulin knockout mice fed with a low iron content diet exhibited a decrease of hepatic mRNA expression. The murine full-length cDNA was isolated and was found to encode an 83-amino acid protein presenting a strong homology in its C-terminal region to the human antimicrobial peptide hepcidin. In addition, we cloned the corresponding rat and human orthologue cDNAs. Both mouse and human genes named HEPC are constituted of 3 exons and 2 introns and are located on chromosome 7 and 19, respectively, in close proximity to USF2 gene. In mouse and human, HEPC mRNA was predominantly expressed in the liver. During both in vivo and in vitro studies, HEPC mRNA expression was enhanced in mouse hepatocytes under the effect of lipopolysaccharide. Finally, to analyze the intracellular localization of the predicted protein, we used the green fluorescent protein chimera expression vectors. The murine green fluorescent protein-prohepcidin protein was exclusively localized in the nucleus. When the putative nuclear localization signal was deleted, the resulting protein was addressed to the cytoplasm. Taken together, our data strongly suggest that the product of the new liver-specific gene HEPC might play a specific role during iron overload and exhibit additional functions distinct from its antimicrobial activity.

cDNA cloning and expression analysis of new members of the mammalian F-box protein family

F-box proteins are critical components of the SCF ubiquitin-protein ligase complex and are involved in substrate recognition and recruitment for ubiquitination and consequent degradation by the proteasome. We have isolated cDNAs encoding a further 10 mammalian F-box proteins. Five of them (FBL3 to FBL7) share structural similarities with Skp2 and contain C-terminal leucine-rich repeats. The other 5 proteins have different putative protein-protein interaction motifs. Specifically, FBS and FBWD4 proteins contain Sec7 and WD40-repeat domains, respectively. The C-terminal region of FBA shares similarity with bacterial protein ApaG while FBG2 shows homology with the F-box protein NFB42. The marked differences in F-box gene expression in human tissues suggest their distinct role in ubiquitin-dependent protein degradation.

Carbonyl-iron supplementation induces hepatocyte nuclear changes in BALB/CJ male mice

BACKGROUND/AIMS

In humans, chronic iron excess may induce hepatic fibrosis and/or hepatocellular carcinoma. This work was undertaken to investigate hepatic iron overload outcome in iron-overloaded mice.

METHODS

BALB/cJ male mice received supplements of 0, 0.5, 1.5 and 3% carbonyl-iron for 2, 4, 8 and 12 months. Histological staining, immunohistochemistry using ferritin antibodies and electron microscopic studies were performed on liver. Liver iron concentration was measured biochemically. Mitotic index and hepatocyte nuclear size were evaluated on Feulgen-stained slides.

RESULTS

Liver iron concentration was increased, reaching 13 times control value after 12 months in 3% iron-overloaded mice, and iron was found predominantly in hepatocytes, with a porto-centrolobular decreasing gradient. Neither hepatic fibrosis nor hepatocellular carcinoma was found. Perls' stain positive inclusions containing ferritin were found within hepatocyte nuclei in 3%-overloaded mice. Electron microscopy disclosed that inclusions consisted of ferritin particle aggregates without a limiting membrane. Mice overloaded with 3% iron for 12 months showed larger hepatocyte nuclei than control mice and a mitotic index increase with presence of abnormal tripolar mitotic figures. In addition, some iron-free hepatocytes were observed.

CONCLUSIONS

Carbonyl-iron supplementation produces significant iron overload in mice but does not result in liver fibrosis or hepatocellular carcinoma after 12 months. However, nuclear changes were produced in hepatocytes, and occasional iron-free hepatocytes were observed: these may represent preneoplastic changes caused by iron overload.

Antiproliferative effect of deferiprone on the Hep G2 cell line

Iron is an essential element in cellular metabolism and the growth of all living species, and is involved in DNA replication. The risk of hepatocellular carcinoma development is associated with an increase in iron availability. The aim of the present work was to investigate the effect of an oral iron chelator, deferiprone (CP20), on HepG2 cell-line proliferation in culture. HepG2 cell cultures were maintained in the absence of fetal calf serum (FCS) and in the presence or not (control cultures) of CP20 at the concentrations of 50 or 100 microM; deferoxamine (DFO) was used as an iron chelator reference. Cell proliferation was investigated by the analysis of DNA synthesis using [3H] methyl-thymidine incorporation and of the cell cycle by flow cytometry. Iron chelation efficiency in the culture model was studied by analyzing the effect of CP20 on radioactive iron uptake, intracellular ferritin level, and transferrin receptor expression. CP20, at the concentration of 50 or 100 microM, inhibited DNA synthesis after 48 hr of incubation and induced an accumulation of the cells in the S phase of the cell cycle. Iron chelators inhibited cellular iron uptake, decreased intracellular ferritin level, and increased transferrin receptor protein and mRNA levels. Our results show that CP20 as well as deferoxamine inhibit HepG2 cell proliferation and block cell cycle in the S phase.

Stearoyl coenzyme A desaturase 1 expression and activity are increased in the liver during iron overload

In humans, hepatic iron overload can lead to hepatocellular carcinoma development. Iron related dysregulation of hepatic genes could play a role in this phenomenon. We previously found that the carbonyl-iron overloaded mouse was a useful model to study the mechanisms involved in the development of hepatic lesions related to iron excess. The aim of the present study was to identify hepatic genes overexpressed in conditions of iron overload by using this model. A suppressive subtractive hybridization was performed between hepatic mRNAs extracted from control and 3% carbonyl-iron overloaded mice during 8 months. This methodology allowed us to identify stearoyl coenzyme A desaturase 1 (SCD1) mRNA overexpression in the liver of iron loaded mice. The corresponding enzymatic activity was also found to be significantly increased. In addition, we demonstrated that both SCD1 mRNA expression and activity were increased in another iron overload model in mice obtained by a single iron-dextran subcutaneous injection. Moreover, we found, in both models, that SCD1 mRNA was not only influenced by the quantity of iron in the liver but also by the duration of iron overload since SCD1 mRNA upregulation was not detected in earlier stages of iron overload. In addition, we found that cellular repartition likely influenced SCD1 mRNA expression. In conclusion, we demonstrated that iron excess in the liver induced both the expression of SCD1 mRNA and its corresponding enzymatic activity. The level and duration of iron overload, as well as cellular repartition of iron excess in the liver likely play a role in this induction. The fact that the expression and activity of SCD1, an enzyme adding a double bound into saturated fatty acids, are induced in two models of iron overload in mice leads to the conclusion that iron excess in the liver may enhance the biosynthesis of unsaturated fatty acids.

Bombesin activation of phospholipase C beta 1 in rat acinar pancreatic cells involves the pertussis toxin-sensitive G alpha i3 protein

Bombesin stimulation of inositol 1,4,5-trisphosphate (Ins P3) formation in rat sonicated pancreatic acinar cells was inhibited by an antibody directed against the pertussis toxin (PTX)-sensitive GTP-binding G alpha i3 protein but not by an anti-G alpha q-11 antibody. After solubilization and gel filtration, [125I-Tyr4]bombesin binding sites were recovered in a peak of protein of 67 approximately 90 kDa with a maximal enrichment corresponding to a molecular mass of 83-kDa. Results obtained from the non-hydrolysable GTP analog guanosine-5'-[gamma-thio]triphosphate (GTP gamma S) binding, PTX-stimulated ADP-ribosylation and immunoblotting showed that the 83-kDa fraction contained the G alpha i3 protein but not the G alpha q-11 protein. Furthermore, GTP gamma S increased the bombesin binding dissociation constant (KD) from 0.32 to 0.60 nM, while the anti-G alpha i3 antibody decreased the maximal binding capacity (Bmax) from 50 to 25 fmol/mg protein without affecting the KD. Mixing solubilized bombesin binding sites with a phospholipase C (PLC) preparation from rat pancreas reconstituted a bombesin-stimulated PLC activity which was markedly inhibited by the anti-G alpha i3 antibody but unaffected by the anti-G alpha q-11 antibody. In addition, this stimulation was inhibited by an anti-PLC beta 1 antibody. This result supports the involvement of the PLC beta 1 isoform in bombesin receptor activation.

[Current data on iron metabolism]

Iron is required for cellular life. However, abnormalities of its metabolism may lead to iron deficiency or iron overload, both conditions which are deleterious. Therefore, stock and distribution of iron in the body must be very stable. Classically, four major proteins are involved in iron metabolism: (a) transferrin which is implicated in its plasmatic transport, (b) transferrin receptor which regulates iron-transferrin uptake, (c) ferritin, the major iron storage protein, and (d) IRP (Iron Regulatory Protein) which regulates both the entry and storage of iron by linking to the IRE (Iron Responsive Element), a nucleotidic sequence found on transferrin receptor and ferritin mRNA. Thus, IRP adapts gene expression to the iron cellular status. Recent data give informations about new proteins involved in iron metabolism: HFE whose gene is mutated in genetic hemochromatosis, ceruloplasmin which permits cellular iron egress and frataxin which is implicated in the exit of iron from mitochondria.

Purification of a histamine H3 receptor negatively coupled to phosphoinositide turnover in the human gastric cell line HGT1

The histamine H3 receptor agonist (R)alpha-methylhistamine (MeHA) inhibited, in a nanomolar range, basal and carbachol-stimulated inositol phosphate formation in the human gastric tumoral cell line HGT1-clone 6. The inhibition was reversed by micromolar concentrations of the histamine H3 receptor antagonist thioperamide and was sensitive to cholera or pertussis toxin treatment. Using [3H]N alpha-MeHA as specific tracer, high affinity binding sites were demonstrated with a Bmax of 54 +/- 3 fmol/mg of protein and a KD of either 0.61 +/- 0.04 or 2.2 +/- 0.4 nM, in the absence or presence of 50 microM GTP[gamma]S, respectively. The binding sites were solubilized by Triton X-100 and prepurified by gel chromatography. They were separated from the histamine H2 receptor sites by filtration through Sepharose-famotidine and finally retained on Sepharose-thioperamide. The purified sites concentrated in one single silver-stained protein band of 70 kDa in SDS-polyacrylamide gel electrophoresis. They specifically bound [3H]N alpha-MeHA with a KD of 1.6 +/- 0.1 nM and a Bmax of 12,000 +/- 750 pmol/mg of protein. This corresponds to a 90,225-fold purification over cell lysate and a purity degree of 84%. Binding was competitively displaced by N alpha-MeHA (IC50 = 5.8 +/- 0.7 nM), (R) alpha-MeHA (IC50 = 9 +/- 1 nM), and thioperamide (IC50 = 85 +/- 10 nM), but not by famotidine (H2 antagonist) or by mepyramine (H1 antagonist). These findings provide the first evidence for solubilization, purification, and molecular mass characterization of the histamine H3 receptor protein and for the negative coupling of this receptor phosphatidylinositol turnover through a so far unidentified G protein.

The 86-kDa subunit of autoantigen Ku is a somatostatin receptor regulating protein phosphatase-2A activity

We previously reported the immunopurification of a somatostatin receptor from the human tumoral gastric cell HGT1 using the monoclonal antibody 30F3 (Reyl-Desmars, F., Le Roux, S., Linard, C., Benkouka, F., and Lewin, M. J. M. (1989) J. Biol. Chem. 264, 18789-18795). Screening of a lambda gt11 HGT1-cDNA library with 30F3 led us to isolate a cDNA encoding an 86-kDa polypeptide displaying 100% structural identity with the 86-kDa subunit (p86-Ku) of the Ku autoantigen. Recombinant p86 expressed in Escherichia coli cross-reacted with 30F3 and specifically bound [125I-Tyr11]somatstatin-14. Binding was totally displaced by somatostatin-14, somatostatin-28, and SMS 201-995, with IC50 values of 0.7, 1.0, and 1.2 nM, respectively. In a search for a biological effect associated with binding, we purified a 36-kDa, okadaic acid-sensitive phosphatase (protein phosphatase-2A (PP2A)) from rat gastric cytosol. PP2A catalyzed 32P release from p34cdc2-phosphorylated histone H1. However, PP2A-induced 32P release was concentration dependently inhibited by recombinant p86-Ku, with a decrease in maximal velocity without a change in Km. Steric exclusion high pressure chromatography indicated that the inhibition resulted from direct interaction of the enzyme with p86-Ku. Furthermore, it was antagonized by increased concentrations of somatostatin-14 and prevented by preincubating p86-Ku with 30F3. Given the key role played by PP2A in cell cycle regulation, the current findings suggest that p86-Ku could be a physiological target of somatostatin antiproliferative action.

[Stimulation of adenylate cyclase by the isoforms of human and rat beta-3 adrenergic receptor expressed in the CHO cells]

OBJECTIVES AND METHOD

The beta 3 adrenergic receptor stimulates lipolysis and colonic relaxation in the rat, and, suggestively, in man. Several human forms generated by different mRNA splicings can occur: the A form of 396 amino acids and the B and C forms extended by 12 and 6 amino acids respectively, in the C-terminus region. In order to characterize these different forms as expressed in CHO cells, we studied adenylyl cyclase stimulation by the beta 3 agonists, SR58611A and BRL37344 and its inhibition by the beta 3 antagonist SR59230.

RESULTS

This antagonist totally inhibited SR58611-adenylyl cyclase stimulation with the following hierarchy of potency: C form >> B > A. In rat, a unique form is expressed which is close to the human B form. This form was the less sensitive to beta 1 and beta 2 antagonists.

CONCLUSION

These findings constitute a molecular pharmacological basis for the design of beta 3 agonists of therapeutic value.

In-vivo effect of somatostatin analog, lanreotide, and/or grp antagonist, bim-26226, on the growth of colon-cancer peritoneal carcinomatosis in the rat

The effect of somatostatin analogue, lanreotide, and bombesin/GRP antagonist, BIM 26226, on the growth of colon cancer peritoneal carcinomatosis in the rat was studied. BDIX rats were i.p. injected with DHD/K12 rat colon cancer cells at day 0 and received from day 3 either lanreotide, BIM 26226, combination of treatments or peptide solvents. At sacrifice, an day 45, no significant difference between groups was observed for peritoneal tumor growth, hepatic metastases, ascite volume and labeling indices in normal colonic mucosa and tumoral tissues. Survival times were similar in other lanreotide-treated and control groups. However, BIM 26226 decreased plasma gastrin level, consistently with a physiological effect of this peptide. Ln all groups, somatostatin and bombesin receptors were found on mucosal and tumoral tissues. Interestingly, bombesin receptor number was higher in severe than in minor cancer stages, contrarily to that of somatostatin receptors. Moreover, an up-regulation of somatostatin and bombesin receptors was observed in BIM 26226- and lanreotide-treated group tumors, respectively, Despite the presence of these specific receptors, lanreotide and BIM 26226 were inactive on tumor growth in this model.

[Fibrochondro-osteosarcoma of the esophageal sub-mucosa in a German Shepherd]

Images